Donate
Independent, objective, nonpartisan research

R 703EHR

Authors

R 703EHR

Tagged with:

Publication PDFs

Database

This is the content currently stored in the post and postmeta tables.

View live version

object(Timber\Post)#3711 (44) { ["ImageClass"]=> string(12) "Timber\Image" ["PostClass"]=> string(11) "Timber\Post" ["TermClass"]=> string(11) "Timber\Term" ["object_type"]=> string(4) "post" ["custom"]=> array(5) { ["_wp_attached_file"]=> string(12) "R_703EHR.pdf" ["wpmf_size"]=> string(7) "1426074" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(256345) "Who Should Be Allowed to Sell Water in California? Third-Party Issues and the Water Market ••• Ellen Hanak 2003 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Hanak, Ellen. Who should be allowed to sell water in California? : third-party issues and the water market / Ellen Hanak. p. cm. Includes bibliographical references. ISBN: 1-58213-075-2 1. Water transfer—California. 2. Water supply—California. I. Title. HD1694.C2H26 2003 333.91'009794—dc21 2003011570 Copyright © 2003 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or state and federal legislation nor does it endorse or support any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword Over 40 years ago, Jack Hirshleifer, James C. DeHaven, and Jerome W. Milliman introduced the idea of California water markets in Water Supply: Economics, Technology, and Policy. Since then, numerous presentations have shown how California’s existing system of aqueducts could be used to move water to the highest bidder. Why then has the development of water markets—a policy that might provide enough water for decades of growth—taken so long to gain acceptance? Part of the answer is provided in Ellen Hanak’s report, Who Should Be Allowed to Sell Water in California? Third-Party Issues and the Water Market. The report notes many obstacles to the expansion of water markets; however, Hanak focuses on a key player—local governments. Local officials, especially in rural areas, are fearful of losing a resource that is a key component of future economic growth. And the specter of bone-dry Owens Valley haunts residents, officials, and investors alike. For these and other Californians, the problem can be put very simply: “No water, no life.” Although the amount of water sold through markets is only 3 percent of all water used in the state, 22 of the state’s 58 counties have adopted ordinances restricting groundwater exports. With the rise of groundwater transfers during the drought of the early 1990s, the fear of uncontrolled “mining” of the aquifers became widespread in many rural counties. In effect, the counties—through a burdensome review process and the prospect of negative public opinion—have discouraged potential sellers from seeking permits in the first place. Hanak finds that, controlling for other factors, these counties have been selling less water, and more of their sales have been to in-county buyers. The report also points to the need for effective policies pertaining to land fallowing, or idling crops to sell water. If fallowing affects other employment and business opportunities, a case can be made for economic mitigation. Some stakeholders are concerned, however, that iii direct compensation to those whose businesses are affected would establish a dangerous legal precedent, generate excessive claims, and create unrealistic expectations about the potential community benefits from water transfers. Given such concerns, communities may prefer the status quo to the risk of losing the benefits that flow from the control of this precious resource. Fair and sustainable rules for fallowing would go a long way toward balancing the needs of local users with the potential gains that result from water transfers. Finally, the report helps explain why it has taken so long to implement water markets. The concerns expressed at the local level— including the prospect of rapidly growing urban centers appropriating water without adequate compensation—are as real as ever. Nevertheless, Hanak shows that there are solutions in the making, and that with proper concern for users, local communities, and the environment, markets could play a key role in addressing California’s water supply problem for decades to come. David W. Lyon President and CEO Public Policy Institute of California iv Summary At current patterns of water use, California faces the prospect of chronic shortages of this vital resource before the year 2020. Among the measures that can alleviate supply and demand imbalances is the development of a water market. A market enables the historical holders of water rights—mainly farmers in the agricultural heartland—to transfer water to other users willing to pay more for it. Potential buyers include urban and industrial users, other farmers with higher-value crops and more limited supplies, and environmental programs to support fish and wildlife habitats. Although significant trading has occurred since the state began promoting this solution in the late 1970s, obstacles remain. In particular, communities in the source regions have raised concerns over the potential adverse effects of water sales on local groundwater users and the local economy. In the absence of clear state-level policy on these “third-party” effects, many counties are attempting to gain an oversight role through local ordinances. This study examines the issue of third-party effects of water transfers in California from the economic, institutional, and legal perspectives. It also evaluates potential mechanisms for resolving the conflicts between those wishing to trade in water and the wider community. Drawing on a range of data sources, including a new database on water transfers and an extensive set of interviews with water users and county officials, the analysis aims to answer the following questions: How has resistance to water transfers affected California’s water market to date, and what are the likely effects of that resistance? What distinguishes cases where conflicts have been successfully resolved from the stalled deals? Are revisions of state water law a necessary or desirable means for dealing with third-party issues, or should solutions be left to local institutions? We begin with some background on the water market and the rise in local resistance to it. v Water Market Trends Jumpstarted by a prolonged drought in the late 1980s and early 1990s, California’s water market is now a firmly established—if modest—feature of the state’s water allocation process, with annual trades accounting for roughly 3 percent of water use. The state has been a major player, notably by running drought year water banks and purchasing water for the environment. As expected, agricultural water districts are the main suppliers, with Central Valley farmers typically accounting for three-quarters of all sales and farmers in the desert valleys of Imperial and Riverside Counties furnishing the rest. Contrary to expectations, urban agencies have played a limited role in market growth. Instead, the main sources of demand have been directly and indirectly linked to new environmental regulations. Direct purchases for instream uses and wildlife reserves have accounted for over one-third of the increase in purchases since 1995. The other growth sector, accounting for over half of market expansion, has been agriculture in the San Joaquin Valley. Farmers there whose contractual water deliveries have been cut back by environmental mitigation programs have turned to the market for replacement water. However, municipal agencies are the principal buyers of long-term and permanent contracts, which account for roughly 20 percent of all sales. Legislation passed in 2001 requiring that local governments demonstrate adequate water supplies for development should increase urban demand for long-term water transfers. Municipalities’ success in forging these deals and ensuring new supplies will depend on their ability to smooth the waters of community resistance in the source regions. The Rise of Local Resistance to the Water Market Concerns in the source regions relate to two distinct types of negative effects of water marketing on third parties. When sales reduce the quantity or degrade the quality of water available to other users, this constitutes a physical externality. California law protects other surface water users, including fish and wildlife, from such effects under the “no injury” statutes of the Water Code. These protections do not extend to groundwater users, however, because groundwater—a major source of vi supply in many regions—is not regulated by the state. Once the state made it clear that the market was open for business during the early 1990s drought, the fear of uncontrolled “mining” of the aquifers became widespread in many rural counties. The other type of negative effect can occur when farmers idle cropland to sell water. Any resulting losses to the local economy—in jobs, sales, or local tax revenues—constitute an economic effect or pecuniary externality. There is no legal tradition in California or elsewhere in the United States for protecting third parties from this type of effect. The state’s widespread use of fallowing contracts to purchase water for the 1991 drought water bank generated considerable discord in some Sacramento Valley counties, where local businesses and farm workers were affected. Rural communities have responded to the lack of state-level, thirdparty protections by putting in place local restrictions on water marketing. By late 2002, 22 of the state’s 58 counties had adopted ordinances requiring a permit to export groundwater or to extract groundwater used in substitution for exported surface water. Counties’ right to invoke police powers to protect groundwater resources was upheld in a 1994 appellate court decision favoring Tehama County. In effect, the absence of state protections for groundwater users provides the legal justification for county-level action. In some counties, the ordinances reflect a broader intent to discourage any type of transfer—whether or not linked to groundwater—that might harm the local economy. Counties do not have the legal authority to ban crop idling for water sales, but some water districts have adopted policies to that effect. This appears mainly to be a practice of districts whose boards are elected by the community at large rather than districts where only landowners have a vote. The recent controversy over a proposed long-term transfer from the Imperial Irrigation District to San Diego erupted when Imperial—whose board is elected by popular vote—was pressured to fallow land despite district policy against the practice. Landowner-run districts have been more likely to fallow land for the water market, especially in periods of low crop prices when the water is less valuable in agricultural uses. vii Effects of County Restrictions on the Water Market To measure the effects of local resistance on the water market, the study assessed the role of county ordinances restricting exports. In counties with ordinances, those wishing to export groundwater or surface water that is replaced by additional groundwater pumping can go through an environmental review process to obtain a county permit. The very low number of permit applications suggests, however, that this process is more useful as a deterrent than as a screening mechanism. High up-front costs and the likelihood of negative public opinion guiding the decision process are both factors discouraging parties from filing. A lack of groundwater permits will not necessarily block transfers if alternatives such as land fallowing are available and acceptable to farmers and their water districts. In the aggregate, however, there is likely to be an effect on the market, both in reducing total sales and in shifting some water to in-county users, who will typically be willing to pay less than outsiders. A statistical analysis of county trading behavior from 1990 to 2001 provides evidence of both effects. In any given year, the presence of an export restriction reduced a typical county’s trades by 14,300 acrefeet and shifted 2,640 acre-feet to in-county buyers. Since 1996, total out-of-county sales, or “exports,” were reduced by 932,000 acre-feet, or 19 percent, and total sales by 787,000 acre-feet, or 14 percent. Overall, the negative market effect of county restrictions cancelled out the positive effect of a generally improved trading environment resulting from state and federal regulatory changes. The Scope for Resolving Third-Party Issues Local resistance is likely to remain a force to reckon with in market development, especially for the long-term, interregional transfers from agricultural users that municipalities will seek to support growth. Moving forward requires finding solutions that provide communities in source regions with adequate safeguards against the potential negative consequences to local water users and the local economy. What have the experiences to date taught us about the scope for positive resolution of viii these conflicts, and what role can policy play in this process? The responses are distinct for the two types of third-party effects. From Groundwater Protection to Groundwater Management Groundwater is a shared resource, with many users drawing from the same aquifer. In the absence of regulation, these users do not have clear incentives to avoid overexploiting the resource. Because the state does not exercise authority over groundwater, the onus for developing management systems falls on local users. Concerns over the groundwater effects of trade have arisen in California’s rural heartland, where local management systems are inadequate or altogether absent. In this context, county ordinances restricting exports can be justified as a firststep precautionary measure to protect local water users from the effects of an unbridled water market. This defensive strategy is nevertheless suboptimal from the standpoint of local as well as statewide interests. A policy limited to restricting exports does little to stabilize the aquifer in places subject to overdraft. It also makes it difficult, if not impossible, to make economic use of the underground storage space through groundwater substitution transfers and banking of imported surface water. Attaining these goals requires a more assertive, comprehensive strategy of groundwater management that protects local users while providing opportunities to address supply and quality problems and allowing those with sound transfer and banking projects to participate in the market. California’s rural areas have so far eschewed the more comprehensive management systems that govern groundwater in Southern California and in many coastal counties. In these regions, high population densities and special technical problems such as saltwater intrusion have led to the introduction of adjudicated basins and special districts with full regulatory authority over the resource. Nevertheless, there is a movement under way toward more active groundwater management in some of California’s rural counties. In some places, the county itself or a special district with countywide jurisdiction has played a convening role for county water users; in others, water districts overlying a shared basin have grouped together to develop a groundwater management plan. ix Key ingredients of active management include the establishment of effective basin monitoring systems and the development of guarantees to mitigate any harm to third parties from market-related activity. A question that remains on the table is whether a strictly voluntary management principle is adequate—a policy still favored by many rural Californians—or whether target levels and pumping restrictions need to be developed for the program to be effective. Developing effective local groundwater management systems places a central responsibility on local authorities—water districts and city and county governments. But the state also has a key role to play, given the statewide benefits of sound local management. Three current forms of state support are appropriate: providing technical assistance, making funds available to support system development, and encouraging the adoption of programs with sound content by attaching conditions to the release of state funds. Once systems are in place, there is also an opportunity for private funding of groundwater infrastructure, especially for municipal supply projects. Mitigating the Economic Effects of Land Fallowing For fallowing, the problem is one of determining the ground rules under which those with access to water rights may take land out of production and sell water to others. Available studies suggest that the aggregate local effects of fallowing have been quite small for programs idling anywhere from 6 to 29 percent of acreage, with local gains from the program largely balancing out local losses. But the modern track record is limited, and popular sentiment in rural areas tends to be shaped by the dire consequences of fallowing for the local economy in the Owens Valley almost a century ago. The key policy issues on the table concern the rules to limit negative community effects: rules on the scale and content of fallowing program design and rules concerning financial mitigation. Both state law and locally determined guidelines already address the first point. Section 1745.05 of the Water Code requires public review of fallowing that exceeds 20 percent of the local water supply. In designing fallowing programs, water districts increasingly include restrictions to maintain the viability of the idled land and to make sure that participating farmers are x not solely in the business of selling water. The economics of fallowing also plays a natural mitigating role. Farmers have incentives to fallow the crops that generate the least profit per acre-foot, and these tend to be the low-value, highly mechanized commodities that generate the lowest onfarm employment and the least value-added through further processing. Even with this combination of operating rules and incentives to limit negative effects to the local economy, there remains the question of whether the community should receive some sort of compensation. At the federal level, there are some precedents for mitigating economic effects when policy changes shift employment and business opportunities in some sectors or regions. With different degrees of success, federal mitigation programs have aimed to assist affected workers and businesses to make a transition to other economic activities. A parallel case could be made for mitigating the economic effects of sizable, long-term fallowing operations, especially if they generate systematic hardships for low-income groups or local governments. In two large long-term deals pending approval, a transfer from the Palo Verde Irrigation District to the Metropolitan Water District of Southern California and one from the Imperial Irrigation District to San Diego, funds have been earmarked for local communities. This will no doubt become a standard component of any future deals of this type, where large volumes of water are sold to distant urban agencies over more than a decade, with expectations of some systematic effects on local employment opportunities affecting low-income immigrant communities. For temporary or intermittent fallowing operations, such as those undertaken in the Sacramento Valley since 2001, there are larger questions about the appropriateness of mitigation. Two buyers, the Department of Water Resources and Metropolitan, have developed a policy to provide mitigation funds, but it remains unclear what damages, if any, merit mitigation. Many are uncomfortable with the term mitigation because it implies the direct compensation of affected parties. In part, this wariness stems from an expectation that the fallowing programs will generate little if any hardship to low-income workers, given the highly mechanized nature of production process for the rice crop being fallowed and the considerable workload generated by land xi maintenance and improvement activities on fallowed acreage. It also stems from a concern that a direct compensation program would establish a dangerous legal precedent, generate excessive claims, and ultimately create unrealistic expectations about the potential community benefits from water transfers. For these reasons, it may make more sense to think of such funds as providing opportunities for community development rather than mitigation. Since 1998, the legislature has considered three bills to institutionalize mitigation, but none has met with approval. Further legislative actions on the fallowing question should be avoided for the time being, for two reasons. First, there is a limited track record on fallowing and no experience with implementing mitigation funds. Second, in the major short- and long-term fallowing programs slated to occur, the transacting parties themselves have been adopting design measures to limit negative effects and setting up funds to benefit the community. These cases provide the opportunity both to assess the consequences of responsible fallowing and to experiment with use of funds for community benefit. If, as the farmers in the Sacramento Valley and Palo Verde argue, the overall effects are not harmful to the local economy, this may help build wider confidence in a new model for fallowing that can displace the ghost of Owens Valley. xii Contents Foreword..................................... iii Summary..................................... v Figures ...................................... xvii Tables ....................................... xix Acknowledgments ............................... xxi Acronyms .................................... xxiii 1. WATER MARKETING AND THIRD PARTIES ....... Water Marketing as a Component of California’s Water Future ................................ State and Federal Support for Water Marketing ......... The Rise of Third-Party Concerns in the Selling Regions ... The Scope for Resolving Third-Party Issues ............ 2. CALIFORNIA’S WATER MARKET, BY THE NUMBERS ................................ Who Can Sell Water and What Kinds Can They Sell? ..... Overall Market Trends ......................... Water for the Environment: A Key Factor in Market Growth ................................ Agriculture’s Leading Role in Market Supply ........... Most Transfers Are Local or Regional ................ From Farms to Cities: A Key Element of Long-Term and Permanent Transfers ....................... Summing Up ............................... 3. THE RISE OF LOCAL RESTRICTIONS ON WATER MARKETING .............................. The Mobilization of Rural Counties ................. An Overview of Export Restrictions ................. Legal Issues ................................ Economic Issues ............................. Operational Issues ............................ Summing Up ............................... 1 2 3 4 6 9 9 12 15 18 20 22 24 25 25 29 30 33 35 36 xiii 4. WHY DO SOME COUNTIES ADOPT EXPORT RESTRICTIONS? ............................ Factors That Make a Difference ................... Data on County Water Economy and Institutions ...... Cross-County Results ........................ Regional Issues .............................. The Mountain Counties: The Legacy of Owens Valley ... Sacramento Valley: A Balancing Act Between the Surface Water “Haves” and “Have-Nots” ............... San Joaquin Valley: Coping with Overdraft and Surface Water Scarcity ........................... Summing Up ............................... 5. WATER MARKET EFFECTS: DO COUNTY RESTRICTIONS HAVE TEETH? ................. Export Permitting: A Largely Uncharted Territory ....... Effects on the Water Market ...................... Data Sources .............................. Results .................................. Summing Up ............................... 6. MITIGATING THE ECONOMIC EFFECTS OF LAND FALLOWING .............................. California’s Recent Experiences with Land Fallowing ...... DWR’s Dry-Year Programs ..................... Long-Term Fallowing Along the Colorado River ....... Fallowing in the San Joaquin Valley ............... Future Trends? ............................. Economic and Legal Issues of Land Fallowing for Water Sales .................................. Economic Incentives and Third-Party Effects ......... Lack of Legal Provisions for Mitigation ............. Are Landowner-Run Water Districts More Likely to Fallow (and Less Likely to Propose Mitigation)? ...... Limiting the Aggregate Negative Effects of Land Fallowing .. Putting Together a Viable Mitigation Program .......... Determining the Scale of Losses .................. Program Content: Targeted or General? ............ 37 37 38 40 41 42 46 52 57 59 59 62 63 65 68 71 71 71 72 75 77 78 78 80 83 85 88 88 89 xiv Program Administration: Counties or Special Institutions? ............................. “Mitigation” or Community Development? .......... Summing Up ............................... 91 93 94 7. FROM GROUNDWATER PROTECTION TO GROUNDWATER MANAGEMENT .............. 97 Mitigating the Effects of Groundwater Transfers ......... 97 Economic and Legal Issues ..................... 97 Mitigation Options and Experiences ............... 99 Mitigation Design Issues ...................... 103 Ingredients of Effective Groundwater Management Systems.. 106 Role of Counties and Local Agencies ............... 109 Role of the State ............................ 111 Summing Up ............................... 114 8. THE SCOPE FOR RESOLVING THIRD-PARTY ISSUES ................................... 117 Local Groundwater Management as the Linchpin of Water Marketing and Water Banking ................. 119 Land Fallowing and Community Development.......... 122 Appendix A. Tracking the Water Market: Data Sources and Caveats .... 127 B. Groundwater Institutions and Basins ................ 139 C. Predicting County Adoption of Export Restrictions ....... 145 D. Measuring the Effect of Export Restrictions on County Water Sales ................................ 149 References .................................... 161 About the Author ............................... 169 Related PPIC Publications .......................... 171 xv Figures 2.1. Short- and Long-Term Water Transfers in California Since 1985 .............................. 13 2.2. Share of Total Water Transfers, by Type of Market, 1988–2001 .............................. 15 2.3a. Annual Volumes Purchased, by Type of End User ..... 16 2.3b. Market Share of End Users .................... 16 2.4. Nonenvironmental Water Purchases, by Location of Selling Party ............................. 21 3.1. Adoption of County Ordinances Restricting Exports ... 27 3.2. California Counties with Groundwater Export Restrictions, 2002 .......................... 28 5.1. California’s Water-Trading Counties, 1990–2001 ..... 64 5.2. Market Effects of Export Restrictions Since 1996 ...... 66 5.3. Market Effects of State and County Policy Environments ............................ 67 B.1. California’s Counties ........................ 141 B.2. Adjudicated Groundwater Basins ................ 142 B.3. Special Groundwater Management Districts ......... 143 B.4. Critically Overdrafted and Special Problem Groundwater Basins Listed in Bulletin 118-80................. 144 xvii Tables 2.1. Regional Sources and Destinations of Water ......... 19 4.1. Average Characteristics of Counties With and Without Export Restrictions ......................... 40 4.2. Regional Characteristics (County Averages) ......... 43 6.1. Study Estimates of the Overall Economic Effect of Land Fallowing ............................... 82 A.1. California Short- and Long-Term Water Transfers, by Type of Market ........................... 132 A.2. Water Purchases, by Type of End User ............ 133 A.3. Transfers, by Region of Origin and Region of Destination .............................. 134 A.4. Source Regions for Environmental Water Purchases .... 135 A.5. Long-Term Transfers Since 1985 ................ 136 A.6. Permanent Transfers of Surface Water Since 1985 ..... 137 B.1. Counties with Groundwater Protection Ordinances, by Region ................................. 139 C.1. Effects of County Characteristics on the Probability of Adopting an Export Restriction ................. 147 D.1. Summary Statistics for Annual County Water Sales and Water Exports, 1990–2001 .................... 154 D.2. Determinants of Annual County Water Sales in 34 Water-Trading Counties, 1990–2001 ............. 157 D.3. Determinants of Annual County Water Exports in 34 Water-Trading Counties, 1990–2001 ............. 158 D.4. Determinants of Annual County Water Sales in 18 Central Valley Counties, 1990–2001 ............. 159 D.5. Determinants of Annual County Water Exports in 18 Central Valley Counties, 1990–2001 ............. 160 xix Acknowledgments This study would not have been possible without the generous assistance of many members of California’s water community. Individuals in a wide range of institutions, from local water districts to state and federal agencies, provided the building blocks of the original database on California’s water market. County and water district officials and other water experts graciously agreed to be interviewed regarding local responses to the rise of the market. At various stages, the study also benefited from the input of an informal advisory panel. Special thanks go to the attendees of a December 2002 meeting in Sacramento, convened to review a draft of the study’s findings and to discuss the policy implications. Attendees included Curt Aikens of the Yuba County Water Agency; Will Boschman of the Semitropic Water Storage District; Chris Campbell of Baker, Manock and Jensen; Jim Easton of Easton Water Resources; David Guy of the Northern California Water Association; Carl Hauge, Jerry Johns, and Rob Swartz of the Department of Water Resources; Richard Howitt of UC Davis; Wendy Illingworth of Economic Insights; Dave Orth of the Kings River Conservation District; Bill Phillimore of Paramount Farms and the Kern Water Bank; Tim Quinn of the Metropolitan Water District of Southern California; Allen Short of the Modesto Irrigation District; Tracy Slavin and Donna Tegelman of the U.S. Bureau of Reclamation; Steve Stroud of the South San Joaquin Irrigation District; Van Tenney of Glenn-Colusa Irrigation District; and Greg Thomas of the Natural Heritage Institute. Although responsibility for the conclusions rests entirely with the author, this feedback was invaluable. I would also like to extend a special thanks to Dave Sunding of UC Berkeley, Mike Kahoe of the San Joaquin Valley Water Users Association, Carol Whiteside and Holly King of the Great Valley Center, xxi and Jim Ganulin of Baker, Manock and Jensen for help in defining the issues in the early stages of the work. Two very capable research associates were key to the research effort. Antonina Simeti did much of the painstaking work of creating the database on the water market and pulling together other datasets used in the study. Caitlin Dyckman, a graduate student in the UC Berkeley Department of City and Regional Planning, worked on the evolution of county groundwater policies and conducted many of the interviews with county officials. The study was finalized with the help of Greg Thomas, Richard Howitt, Howard Schatz, and Ricardo Ramirez, who provided written reviews, and Peter Richardson and Patricia Bedrosian, who provided editorial assistance. xxii Acronyms AB AVEKWA BMO CALFED CEQA CVP CVPIA DWR EIR EWA GCID IID KWCA MWDSC NHI PVID RCRC SB SDCWA SWP SWRCB TAC UC USBR WAC YCWA Assembly Bill Antelope Valley–East Kern Water Agency Basin management objective Multiagency state and federal program for the San Francisco Bay Delta California Environmental Quality Act Central Valley Project Central Valley Project Improvement Act Department of Water Resources Environmental impact review Environmental Water Account Glenn–Colusa Irrigation District Imperial Irrigation District Kern County Water Agency Metropolitan Water District of Southern California Natural Heritage Institute Palo Verde Irrigation District Regional Council of Rural Counties Senate Bill San Diego County Water Authority State Water Project State Water Resources Control Board Technical Advisory Committee University of California U.S. Bureau of Reclamation Water Advisory Committee Yuba County Water Agency xxiii 1. Water Marketing and Third Parties Over most of the past century, the water needs of California’s growing economy and population have been met by increasing water supplies. The result has been a complex mosaic of hydraulic investments in surface storage and conveyance undertaken at local, state, and federal levels. This “developed” surface water system provides the state’s agricultural, residential, and industrial customers with roughly two-thirds of the water they use, the remaining third coming from groundwater reserves. Until recently, the system generally has met the needs of these users, except in periods of severe drought. Concerns that insufficient water resources were being devoted to the environment—and in particular to the survival of endangered wildlife—have led to reductions, beginning in the early 1990s, in the amounts available to some agricultural and municipal users. With the environment now explicitly considered as a user with legitimate (and sometimes paramount) requirements, and with the promise of continued, rapid population growth, the state’s Department of Water Resources projects chronic water shortages before the year 2020 at current patterns of use (Department of Water Resources, 1998). Redressing supply and demand imbalances through additional surface reservoir development has become more difficult. Hydrological options are less favorable, cost considerations loom larger, and concerns about the potential environmental consequences of such investments play a much greater role in the public decisionmaking process. This difficulty has prompted considerations of alternative technologies to increase supply, through “groundwater banking”—or storing excess run-off in underground aquifers—and through recycling and desalination. It has 1 also heightened interest in making better use of the existing supply through conservation efforts and the development of a water market. Water Marketing as a Component of California’s Water Future A market permits the temporary, long-term, or permanent transfer of water from the existing rights-holders to other water users in exchange for payment. In California, these rights generally have been appropriated for many decades under the state’s “first in time, first in right” legal system. Water transfers are seen as a way of adding flexibility to the state’s water supply—both to address temporary drought conditions and to accommodate longer-term changes in the pattern of demand. Because water systems have been intimately linked to the development of California’s agricultural heartland, the bulk of use rights are held by farming interests in the Central Valley and the desert counties to the south that rely on Colorado River water (Hundley, 2001). In normal water years, agriculture uses about 34 million acre-feet of water, or 80 percent of the total used by Californians for their combined residential and business needs (Department of Water Resources, 1998). Agricultural users often pay significantly less for water than municipal and industrial users do in the coastal metropolitan areas, even allowing for differences in transportation and treatment costs. Ready availability of water has enabled California to become the nation’s largest agricultural economy, with one of the world’s most extensive irrigation systems. This can be seen as a successful outcome of past federal and state support to the development of western agriculture through large-scale hydraulic projects. As other demands for water continue to grow, however, there has been increasing pressure to weigh this policy goal against one emphasizing the scarcity of this natural resource. In an era when pricing has been advocated as a solution for a whole range of resource allocation issues, water transfers are seen as a way to accommodate the changing pattern of demand while compensating water-rights-holders for forgoing their own access. The discussions on water marketing have often emphasized the potential of the market to move water from agricultural to growing urban areas, many of which are 2 willing to pay more for the water than it is worth to the seller in the nextbest agricultural use. But a water market can also help balance supply and demand within the agricultural sector, as farmers with higher-value uses for the water purchase it from those with more-senior rights and less-productive farms. Finally, market-based transfers are seen as a potential tool for reallocating water from agriculture to the environment through voluntary means rather than rationing. State and Federal Support for Water Marketing California’s foray into water marketing began in 1977, a year of severe drought. Two reports commissioned at that time, one by the governor and one by the legislature,1 strongly endorsed water marketing as a component of the state’s water future. The governor’s commission also advocated a number of changes in the Water Code to facilitate transfers, notably provisions to ensure the security of water rights for transferring parties and access to the use of conveyance facilities. Although many of the recommendations were accomplished in the years that followed, the 1980s saw little uptake in market activity. In the early 1990s, several events significantly changed the trading climate. First, natural conditions provided the occasion for a large-scale experiment in water trading when a multiyear drought prompted the state to initiate an emergency water bank in 1991. The following year, in response to findings that the federally run Central Valley Project (CVP) was having deleterious effects on the indigenous wildlife of the San Francisco Bay-Delta water system, Congress passed the Central Valley Project Improvement Act (CVPIA). The CVPIA mandated that 800,000 acre-feet of project water (of a total of 7 million) be returned to instream uses to regenerate salmon runs and that another 400,000 acrefeet be allocated to wildlife refuges. The CVPIA also contained provisions to facilitate water marketing and introduced a mechanism for the project to purchase additional water for environmental purposes. In 1994, contractors of the State Water Project (SWP) concluded negotiations for the Monterey Agreement, a revision of project operating ____________ 1Governor’s Commission to Review California’s Water Rights Law (1978); Phelps et al. (1978). 3 rules that included measures to make it easier for contractors to transfer water to one another. At the close of the decade, two further state and federal actions were significant. Under instructions from the Secretary of the Interior in 1996 and 1997, California began to devise a plan to reduce its use of Colorado River water to the contractually allocated amount of 4.4 million acre-feet over a 15-year period. This 4.4 Plan created strong incentives for water transfers between agricultural and urban users of Colorado River water within California. In 2000, state and federal authorities launched the Environmental Water Account (EWA), a program of water purchases for the environment under CALFED, a multiagency state and federal program to restore health to the fisheries of the San Francisco Bay-Delta system while securing water supplies to agricultural and urban users. The Rise of Third-Party Concerns in the Selling Regions Although substantial trading resulted from these policy changes, those wishing to enter the water market—on either a short- or long-term basis—have encountered a number of obstacles. Central among these is the question of how to deal with the “third-party” or indirect effects of the transfers on other water users and the larger communities where water is being sold. Community resistance has soured a number of deals over the past decade and has likely prevented others from being proposed. Many of California’s rural counties have introduced ordinances that directly restrict groundwater exports and indirectly restrict the sale of surface water. In several counties, the ordinances circumscribe the rights of individuals and local water agencies to engage in groundwater banking. These measures respond to the potential consequences of the transfers on the local community. Short-term transfers during the drought water bank in the early 1990s raised two contentious issues. First, communities were concerned about local economic consequences when some farmers fallowed their land to sell water, simultaneously drying up demand for labor and other farm inputs and cutting off the 4 supply of raw materials to local processors. Second, farmers who sold their surface water and maintained their crops by pumping more groundwater than usual were in some cases seen as reducing the overall quantity and quality of supplies available to other water users. Community concerns can be exacerbated if the seller makes a multiyear commitment. For the proposed long-term transfer of water from the Imperial Irrigation District to San Diego, negotiated as part of California’s 4.4 Plan to reduce Colorado River use, leaders in the Imperial Valley have been reluctant to agree to multiyear land fallowing. Sending water toward urban coastal areas adds to the uneasiness of such long-term transfers: Will such areas take precedence in the future, regardless of the terms of the transfer? Some observers express a general skepticism regarding the market for water, given the sharp regional disparities between the wealthy metropolitan areas on the coast and California’s agricultural valleys, which lag far behind in terms of income, employment, and education levels. In this view, arguments about market efficiency are little more than a justification for draining the poorer hinterland of the resources it may need for more favorable growth and development in the years ahead. Although the tone of the debate is at times highly charged, local officials’ resistance to transfers can be seen as a precautionary approach to a potential irreversibility problem: Once the water is gone, they will lack the money and political influence to get it back. Reinforcing this concern is the specter of the Owens Valley affair—the notorious deal in the early 20th century wherein Los Angeles bought up so much of that region’s water that the local agricultural economy collapsed. Community opposition to water transfers challenges the notion that water rights belong to users and suggests instead that water is a community resource. The county ordinances are an attempt to impose a level of local oversight on water transfers not provided for in state law, which is not uniform in its treatment of transfers or in the assessment of effects on third parties. State approval is required only for transfers involving surface water rights acquired since 1914, certain types of groundwater banking, and any water that is “wheeled” (conveyed) through a publicly owned facility (e.g., the California Aqueduct). Only in the latter case must the state ensure against unreasonable negative 5 economic effects on the source county; for the other transfers, there is simply an obligation not to harm other legal users of the surface water (i.e., other rights-holders), fish, and wildlife. There is no provision to protect other groundwater users, as groundwater rights remain largely unregulated by the state. Although there is some question as to whether the county ordinances would stand up to a legal challenge, there is no doubt that they arise where communities are not confident about the existing legal and political process governing transfers. The question nevertheless remains as to whether counties are managing community interests effectively; specifically, an overly conservative policy on transfers can lead to missed opportunities for enhancing collective welfare. The Scope for Resolving Third-Party Issues This study examines the issue of third-party effects of water transfers in California from the economic, institutional, and legal perspectives. It evaluates potential mechanisms for resolving the conflicts between those wishing to trade in water and the wider community. By drawing on a range of data sources, including original survey work, the analysis aims to answer the following questions: How has local resistance to water transfers affected California’s water market to date and what are the likely future effects? What distinguishes the cases where conflicts have been successfully resolved and the stalled deals? Are revisions of state water law a necessary or desirable means for dealing with third-party issues, or should solutions be left to local institutions? The report begins by providing some background on the water market and on the extent of community response. Chapter 2 presents an overview of the statewide water market since the mid-1980s, drawing on a new dataset developed for this study. It shows the evolution of total volumes transferred and the composition of the market—by geography, types of water users, and shifts in end use—in response to the state and federal policy changes outlined above. The next two chapters document the rise in local restrictions on water marketing, with a specific focus on the adoption of county ordinances. Chapter 3 summarizes the statewide trends and discusses some of the legal, economic, and operational issues that the ordinances 6 raise. Drawing on county-level economic indicators and qualitative information gathered from interviews with county officials and water users, Chapter 4 explores why certain counties have adopted export restrictions and others have not. Chapter 5 examines whether and how local opposition affects the water market. This examination includes a review of the permitting process for exports in counties that have adopted restrictions and a statistical analysis of the factors influencing water transfer behavior at the county level. The study then turns to the policy questions that arise from the economic effects of land fallowing and the physical effects on water users arising from transfers involving groundwater. Both economic theory and institutional practice provide justifications for mitigating negative effects of water transfers. In theory, any transfer that truly raises efficiency by moving a resource into a higher-value use will generate sufficient gains to enable the losers to be compensated, such that the well-being of all parties is enhanced or maintained. For transfers that affect the availability of the physical resource to other water users, compensation is justified on grounds of efficiency as well as equity (Coase, 1960). This principle already underlies the state’s policy to “make whole” other surface water users affected by a transfer. More generally, it forms the basis for environmental mitigation of development projects. Measures might include providing alternative sources of water supply to the affected party or devising some other form of financial or in-kind compensation. Compensation mechanisms are not always easy to devise and reach agreement on, however, particularly when the damages are unclear or difficult to quantify. Chapter 6 examines these issues as they apply to the economic effects of land fallowing, and Chapter 7 investigates the physical effects on water users of groundwater transfers and the related practice of groundwater banking. Mitigating harm to groundwater users implies the existence of an effective system for managing groundwater resources more generally. Because these resources are almost always shared by multiple parties, groundwater management usually requires collective oversight. In California, there is a vigorous debate concerning the appropriate level of governance for groundwater. At one extreme are those who argue that 7 groundwater should come under the jurisdiction of the state alongside surface water (Sax, 2002). At the other extreme are those who argue against any restrictions on individuals’ rights to pump. The middle ground in this debate has sought solutions involving local institutions. Two potentially competing approaches have grown in popularity since the early 1990s—the establishment of groundwater management plans by local water entities overlying a shared basin and the coordination of groundwater management at the county level through groundwater protection ordinances. Chapter 7 examines the policy and institutional questions that arise from these approaches, with a particular focus on the role counties are and might be playing. Chapter 8 concludes by summarizing the policy issues that face state and local governments in addressing third-party issues and the role local entities can and should play in the statewide water market. 8 2. California’s Water Market, by the Numbers This chapter documents the evolution of the state’s water market from the mid-1980s to 2001, drawing on a new dataset on individual water transfers developed by the author from a variety of state, federal, and local sources. The data allow an analysis of volumes transferred by duration, region of origin and destination, initial and final use, type of transacting party, and affiliation (if any) with the large state and federal water projects. The discussion highlights two areas of particular interest. First, to what extent has the market responded to the policy measures to encourage trading over the past two decades? Second, does the trading that has occurred correspond to expectations? Who Can Sell Water and What Kinds Can They Sell? The state’s Water Code provides two basic guidelines on who can participate in the water market and what they can sell: Sellers must be rights-holders, and the water they sell must be “wet.” “Wet water” is the term commonly used to contrast with “paper water”—water rights held on paper for which actual water is not available. Under the appropriative rights doctrine governing most of the state’s surface water, the “use it or lose it” clause dictates that rights lapse for any water not used for five consecutive years. The Water Rights Division of the State Water Resources Control Board (SWRCB) estimates that there are roughly three times as many paper claims on surface water as water available in the system. Water-rights-holders must therefore demonstrate that the water they propose to part with is indeed water that they would otherwise use in some way. Without this safeguard, the seller would end up transferring “paper” water that someone else is already legally using. 9 Sources of wet water are of four basic types: surface storage, excess surface water, conserved surface water, and groundwater. The first two sources are not widely available. Only a limited number of rights-holders have surpluses available in surface storage, and the use-it-or-lose-it principle limits conditions under which excess surface supplies are actually considered wet. Contractors with the CVP and SWP projects are the only ones who have been able to sell regular excesses of surface supplies, on grounds that their rights are determined by contract rather than the appropriative doctrine.1 Otherwise, rights-holders may sell the excess surface water generated in very wet years. These are times when overall market demand is more limited. Conservation and groundwater are more generally available options. Conservation can be achieved through investments to improve the efficiency of the conveyance and use systems (e.g., canal lining, installation of drip irrigation, recycling), or through land fallowing. Groundwater can be transferred directly or, more commonly, can be used on-site in lieu of surface water transferred to another party. This latter practice, known as “groundwater substitution” or “groundwater exchange,” is a form of “conjunctive use” of groundwater and surface water. With conjunctive use, the groundwater aquifer serves, in effect, as an underground reserve that can be drawn on to a greater or lesser degree as the quantity of available surface water varies. In some places where prior pumping activity has generated unused space in the aquifer, active storage or banking projects are increasingly popular. As a consequence, storage in underground banks is becoming a new source of water for the market, much like surface storage. Although the initial appropriation of water rights in the second half of the 19th century primarily involved individuals and private companies, most surface water rights today are held by local public agencies: special water and irrigation districts and some municipalities. Legally, some of these parties actually hold long-term “contract entitlements” rather than “rights” to surface water; the large state and federal projects they contract with hold the actual water rights. In some ____________ 1A loosening of restrictions on such trades among Central Valley Project contractors is arguably the main effect of the CVPIA on the water market to date. 10 water districts, individual farmers have specific contractual amounts of water (or “allocations”) assigned to them and are therefore in a position to sell their surface water. Generally, the right to pump groundwater (and hence, potentially, to sell it) is available to all private individuals overlying the aquifer. As we will see in greater detail in subsequent chapters, there are few places where rights are “quantified” (i.e., where users have rights to a specific quantity of water) or where local agencies, rather than individuals, effectively control access to the aquifer. Current groundwater banking operations typically involve local agencies (for instance, the Semitropic Water Storage District and the Arvin-Edison Water Storage District in Kern County each run programs) or consortia of private and public agencies (for example, the Kern Water Bank). Such programs could conceivably be run by a single private entity if it were sufficiently large to cover the banking area. A case in point is the recently abandoned Cadiz groundwater storage and dry-year supply project. In this project, a private agribusiness firm with land holdings in eastern San Bernardino County would have banked and sold water to the Metropolitan Water District of Southern California. The approval process for transfers varies according to the nature of the water right as well as the source of water. The SWRCB must approve transfers (changes in purpose or place of use) involving surface water rights established from 1914 onward, the year the state’s “modern” Water Code became effective. Transfers of surface water among contractors of the federal and state water projects generally do not require SWRCB approval because they do not involve a change in the purpose and place of use assigned to the overall water right. The projects themselves must authorize these sales. The general practice is for farmers to go through the water district, which negotiates the transaction, and for the state or federal project to decide whether the district may make the sale. Transfers of groundwater and of surface water held in pre-1914 appropriative rights do not require SWRCB approval. However, they do come under state or federal jurisdiction if government-owned conveyance facilities are involved, which is likely to be the case in most parts of the state. 11 It was not possible to track systematically the type of water being sold in the data presented below. On average, local public agencies (water districts, irrigation districts) account for over 90 percent of the volumes sold since the mid-1980s, with private parties and mutual water companies making up the balance. The share of direct private activity was highest during the early 1990s, when the state contracted directly with individual farmers for a substantial portion of the water acquired under the drought water bank. The data presented focus on annual flows of water resulting from temporary transfers (under one year), long-term transfers (over one year), and what we have termed “deferred exchanges,” which involve a promise that the buyer repays the water (along with a cash payment in some cases) to the seller at a later date. The data on a fourth category, the permanent transfer of water rights or contract entitlements, are presented separately. Such transfers amount to an outright sale of the rights to use the specified amount of water in perpetuity or for the remaining duration of the contract in question. Because the actual amount accessible to the buyer can vary with the conditions of the water year, it is not strictly appropriate to consider an annual flow of water transferred. For more details on the sources and methods used to construct the water transfers database, the reader is referred to Appendix A. Overall Market Trends The statewide water market got a jumpstart during the last multiyear drought, which began in 1988 (Figure 2.1). Market growth was largely driven by direct interventions of the Department of Water Resources. DWR began making dry-year purchases to offset lower deliveries to State Water Project contractors and wildlife refuges in the first year of the drought. These early operations, which involved only a handful of Sacramento Valley water districts (most notably the Yuba County Water Agency), quickly brought the total volume traded to over 500,000 acrefeet, three times the pre-drought levels. In 1991, when the dry-year market was opened up to any willing buyers and sellers, DWR purchased 821,000 acre-feet of water for resale, bringing the overall market volume to over 1.1 million acre-feet. Water banks and other dry-year purchases were also operated in 1992 and 1994. From 1988 to the end of the 12 Acre-feet traded (thousands) Rainfall index 1,400 1,200 1,000 800 600 400 200 0 Rainfall Acre-feet traded 18 17 16 15 14 13 12 11 Wet 10 9 8 Normal 7 6 5 Dry 4 3 11111111111122111999999999999009999999888889990099932109876556410987 NOTES: For details, see Table A.1. Rainfall is measured by the Sacramento Valley 40-30-30 index, an indicator of water supply conditions for the state’s primary river system (see Appendix D). Figure 2.1—Short- and Long-Term Water Transfers in California Since 1985 drought in 1994, state and federal dry-year purchases for resale and environmental uses accounted for over 40 percent of a market that had jumped from an average of 150,000 acre-feet to over 600,000 acre-feet per year. 2 Although the second half of the 1990s saw a succession of wet years, market activity remained strong, with volumes typically exceeding the drought-year levels, especially by the end of the decade. The only dips in ____________ 2The average market volume as measured by end-user purchases is just over 500,000 acre-feet for the 1988–94 period. This discrepancy arises mainly because DWR’s purchases were higher than its resales to other users in some years. The gap was most notable in 1991, the first year of the water bank, when purchases exceeded resale volume by over 400,000 acre-feet. In several years in the 1990s, there were also smaller discrepancies between purchases and sales of water user pools, notably the SWP “turnback” pool, described in the text, and the Sacramento River Water Contractors’ Association, a pool run by some CVP contractors. 13 a generally upward trend in purchases occurred in the exceptionally wet years of 1995 and 1998, when many areas of the state experienced flooding. With annual trades now exceeding 1.2 million acre-feet— eight to ten times the levels of the mid-1980s—the market appears firmly established as a component of California’s water allocation process. It is also useful to consider the size and scope of the market from three other perspectives. First, the statewide market at current levels represents only 3 percent of all water used by Californians for municipal, industrial, and agricultural purposes (Department of Water Resources, 1998). Second, although there has been an increase in the number of long-term transfer agreements, the market continues to be dominated by short-term transactions, negotiated on an annual basis, which account for about 80 percent of the total volume transferred. Finally, the size and scope of the market are strongly influenced by the intervention of state and federal authorities. This influence stems not only from their important direct role in purchases but also from the relative ease water users have in gaining approval for transfers within the confines of the state and federal projects. Since 1988, direct government purchases have accounted for nearly one-third of the total volume traded. Transfers among contractors within the same projects (SWP, CVP, and the Colorado River Project) account for more than half of all water sold (Figure 2.2). Transfers within the CVP and the SWP have generally increased in response to the more-flexible trading rules adopted by those projects in the early 1990s. The growth has been most pronounced within the SWP, for which internal transfers were rare before the Monterey Agreement. By contrast, the “open market,” a residual category defined broadly to include any transfers between water users not associated with the same project, accounts for only 15 percent of the water transfers recorded over the 14-year period. This share initially increased immediately after the drought, but it has been on the decline again recently, as direct government purchases for environmental programs have been on the rise. In this context, it is noteworthy that a provision in the Central Valley Project Improvement Act to allow project water to be sold to noncontractors, considered a major innovation at the time, has gone virtually unused. Such transfers invoke a $25 per acre-foot 14 Within Colorado River Project, 16% Within SWP, 10% Direct government purchases, 31% Within CVP, 28% Open market, 15% Figure 2.2—Share of Total Water Transfers, by Type of Market, 1988–2001 environmental surcharge, a fee that appears to have proven prohibitive until now.3 Water for the Environment: A Key Factor in Market Growth Market growth in the aftermath of the drought has been largely driven by environmental concerns. The influence of environmental policy is most readily seen by comparing the patterns of water purchases during the multiyear drought (1988–1994) to those in the more recent period (1995–2001), when rainfall has generally been above normal (Figures 2.3a and 2.3b). The most obvious element of the new role for the environment is the rise in direct purchases for instream uses to support aquatic life and for wildlife refuges, through federal and state programs and most recently ____________ 3With the aid of a simulation model, Loomis (1994) predicted that this surcharge would make it unlikely for out-of-project transfers to occur except under severe drought conditions, when buyers would be willing to pay enough to cover it. 15 Acre-feet per year (thousansds) 1,000 900 800 700 600 500 400 300 200 100 0 1988–1994 NOTE: For details, see Table A.2. 1995–2001 Mixed uses Other farmers San Joaquin Valley farmers Environment Cities Figure 2.3a—Annual Volumes Purchased, by Type of End User 100 90 80 70 60 50 40 30 20 10 0 1988–1994 NOTE: For details, see Table A.2. 1995–2001 Mixed uses Other farmers San Joaquin Valley farmers Environment Cities Figure 2.3b—Market Share of End Users Percentage 16 the CALFED EWA. As one of the beneficiaries of DWR’s drought purchases, the environment already accounted for 12 percent of demand during the 1988–94 drought. Since 1995, this share has doubled on average; it reached a third of total demand in 2001, the first full-fledged year of the EWA. On an average yearly basis, environmental purchases have increased by 155,000 acre-feet, out of a total market gain of 430,000 acre-feet. The less obvious component of environmentally related demand is the rise in purchases by San Joaquin Valley farmers. Although this group’s change in market share is less dramatic (moving from 37 percent to 44 percent over the two periods), its increase in volume (228,000 acrefeet per year) accounts for over half of total market growth. Much of this growth can be linked to the changes introduced under the CVPIA in 1992, which mandated that a portion of project water be returned to instream uses. Since then, the CVP’s agricultural service contractors located south of the Delta received full project deliveries in only the two very wet years, 1995 and 1998. One outcome has been the development of an active water market, as certain contractors (most notably Westlands Water District) sought to offset reductions in deliveries via purchases. Within the San Joaquin Valley, CVP service contractors’ market share of purchases has risen from 63 percent to 87 percent across the two time periods. The corollary of this growth has been a decline in the relative importance of municipal and industrial water users. Whereas cities were the main recipients of traded water during the 1988–1994 drought with 42 percent of all purchases, their share since 1995 has been halved. With the exception of 1991, when purchases nearly reached 500,000 acre-feet, volumes have remained relatively flat, at around 200,000 acre-feet per year. The modest role of urban demand contrasts with the frequently voiced assumption that a water market would develop primarily as a response to population growth and the ability of urban dwellers to pay more than agricultural users for water.4 As we will see below, cities are significant players in defining the market for long-term and permanent ____________ 4See Phelps et al. (1978) and Vaux and Howitt (1984) for early economic arguments in favor of a water market in California. 17 transfers of water. This pattern suggests that their role may increase as this part of the market develops. Agriculture’s Leading Role in Market Supply Economists have typically assumed that agriculture would be the leading source of water supply, and this assumption is confirmed by the data. In most years, agricultural water users provide at least 90 percent of supply. It is nevertheless interesting to note that the introduction of “turn-back” pools within the State Water Project has made a significant amount of urban agency water available to agricultural users in wet years. Under the program, introduced as one of the reforms in the 1994 Monterey Agreement noted in Chapter 1, SWP contractors are able to sell back amounts of project water they will not need if there are willing SWP buyers.5 From 1998 to 2000, the turn-back pool generated 200,000 acre-feet or more of water per year, or roughly one-fifth of total market supply. Given the primacy of agricultural supply, it is not surprising to find that the main source regions are the Central Valley (served by the CVP, the SWP, and several large, autonomous, surface water projects) and the agricultural valleys to the south served by the Colorado River Project (Table 2.1). In most years, the Central Valley has furnished about threequarters of the total volume transferred. Within this region, there are pronounced shifts, depending on the nature of the water year. From 1988 to 1994, Sacramento Valley farmers and districts supplied about 45 percent of all water sold. Although this share fell considerably after 1994 (in some years, to under 10 percent of the statewide market), the region again provided over 40 percent of the water in 2001, the first dry year experienced since the previous drought. In the interim, the market shifted to the San Joaquin Valley, which established a vibrant intraregional market to supply water-short districts within the region. In every normal to wet year since 1993, the first year of CVPIA ____________ 5Because the sale price is fixed below the level of per-acre-foot charges incurred by the seller for project water, the pool’s attractiveness to sellers is relative, not absolute: It enables them to pay less than full price for contract amounts in excess of current needs. 18 Table 2.1 Regional Sources and Destinations of Water Sales by water users Sacramento Valley San Joaquin Valley Southern California San Francisco Bay Area Other 1988–1994 1995–2001 1,924,937 1,363,037 970,942 87,195 3,055 1,057,064 3,715,039 1,577,597 82,575 88,694 Total sales 4,349,166 6,520,969 Purchases by water users Sacramento Valley San Joaquin Valley Southern California San Francisco Bay Area Other 135,079 1,450,917 1,187,157 313,197 6,152 515,509 3,253,292 1,234,555 43,505 14,993 Purchases of environmental water % from Sacramento Valley % from San Joaquin Valley 408,672 74 25 1,484,255 17 81 Total purchases 3,501,174 6,546,109 Exports (imports) of nonenvironmental water Sacramento Valley San Joaquin Valley Southern California San Francisco Bay Area 1,488,725 (190,683) (216,215) (230,738) 288,383 (739,455) 343,042 9,070 NOTES: For details, see Tables A.3 and A.4. The bulk of the difference between total purchases and total sales in the first period is the surplus purchased by DWR and distributed through means other than sales to other end users. There were also some small discrepancies in both periods between purchases and sales of user pools. These discrepancies account for the fact that nonenvironmental exports and imports do not sum to zero. See footnote 2 in this chapter. All water measurements are in acre-feet. implementation, San Joaquin Valley farmers and districts have furnished at least half of the total amount transferred. It is common for agricultural districts in this region to restrict out-of-district transfers to cases where land in the recipient district is owned or leased by the same farmer. 19 The two Central Valley regions are the principal suppliers of environmental water (Table 2.1). There has been a major source shift from north to south of the Delta since the early 1990s drought, with the institution of restoration programs along the San Joaquin River system, supplied by area water districts. Kern County water users have also been major suppliers of the EWA. Most environmental water is used in the region of purchase; the main exceptions are EWA purchases of water north of the Delta, which facilitate the delivery of project water to users in the south while protecting Delta fisheries. The Southern California region, defined broadly to include both the desert counties and the coast, generates most of the remaining quarter of total supply. The largest single source has been the Imperial Irrigation District, which has operated a 110,000 acre-feet per year long-term transfer with the Metropolitan Water District of Southern California (MWDSC) since 1988. In the mid-1990s, there was also a large twoyear transfer to the MWDSC by the Palo Verde Irrigation District, another agricultural contractor on the Colorado River Project. With the exception of the early 1990s drought, when several Bay Area cities made substantial purchases, no other region has played a significant role in the market on either the supply or demand side. The Bay Area’s share in demand has dropped from 9 percent of the total in the drought years to only 1 percent in the more recent period. This region’s share in supply is about 2 percent. Most Transfers Are Local or Regional We have already seen that the San Joaquin Valley is both a major supplier and a purchaser of water. The market in Southern California is also primarily regional in nature (Table 2.1). The only exports leaving the region are the transfers by municipal agencies to the SWP turn-back pool, purchased by San Joaquin Valley agricultural districts. In the Sacramento Valley, the only region in the state where water users purchase significantly less than the volume sold, exports are concentrated in dry years. In years with normal to high rainfall, two-thirds of the water is transferred to other water users within the region. Another way of seeing the shares of local and regional markets is to look directly at the source of water obtained by users. Figure 2.4 shows 20 Percentage 100 90 Seller in other region 80 (through bank or pool) Seller in other region 70 (direct) 60 Seller in same region Seller in same county 50 40 30 20 10 0 1990–1994 1995–2001 (525,000 acre-feet/year) (720,000 acre-feet/year) Figure 2.4—Nonenvironmental Water Purchases, by Location of Selling Party this breakdown, for nonenvironmental water only, according to the location of the selling party in relation to the buyer. Nearly one-quarter of total volume is purchased from parties in the same county and another half from parties in the region. Interregional transfers account for the remaining 25–30 percent of the market.6 Only a relatively small fraction of these transfers (one-fifth) are negotiated directly between parties in different regions; the lion’s share moves through banks and pools run by DWR. This pattern highlights again the important role played by the government agencies in California’s water market. Given the need for federal and state agency approval for the use of conveyance facilities, this role is indispensable for any interregional transfers of water physically moving across the Delta. Approval decisions are complicated because pumping additional water through the Delta can affect water quality standards and put protected fish species in danger. When an agency acts as an intermediary, it is able to facilitate the movement of water across the Delta. Agency input has also been crucial ____________ 6Because this graph presents data from the standpoint of end-user purchases, it includes only the water bank volumes that were actually resold in the first period, not the full amounts acquired by DWR. 21 in successful district-to-district transfers—for instance, the 2001 transfer of 160,000 acre-feet from Sacramento Valley CVP contractors to the Westlands Water District. Many observers believe that the absence of state or federal agency sponsorship significantly complicates the approval process for cross-Delta transfers. From Farms to Cities: A Key Element of LongTerm and Permanent Transfers Our records show 15 approved long-term transfers and 14 permanent transfers of surface water rights or entitlements from 1985 to 2002.7 At least ten additional transfers were pending approval in late 2002. As noted, long-term transfers have generally accounted for about one-fifth of all trades since the late 1980s. Volumes traded surpassed the 250,000 acre-feet mark for the first time in 2001. Contract duration runs from a low of two years to a high of 35, with an average of 15 years. The permanent transfers, bunched at the end of the decade, total another 175,000 acre-feet. These mainly concern the transfer of SWP contract entitlements under the Monterey Agreement (representing over 110,000 acre-feet) and the transfer of pre-1914 water rights among parties within Kern County. Almost all these transfers involve shifts of water from agricultural to urban uses. The handful of exceptions includes one long-term deferred exchange and one long-term transfer between districts with a large municipal customer base (Solano County Water Agency and the Mojave Water Agency; Placer County Water Agency and Northridge Water District, respectively) and several long-term transfers from agricultural districts to the environment (most notably, as part of the negotiated agreement for the restoration of the San Joaquin River). Only two longterm transfers between agricultural districts appear, one in the Sacramento Valley and one in San Joaquin, and two permanent transfers of contract entitlement among agricultural users (both in the San Joaquin Valley). ____________ 7For details on the transactions discussed in this section, see Tables A.5 and A.6. 22 This trend holds for transfers pending approval in 2002. Key among these are two large long-term agriculture-urban transfers within the Colorado River Project (Imperial Irrigation District to San Diego, Palo Verde Irrigation District to MWDSC) and several permanent transfers of CVP contract entitlement from agricultural water districts to towns in the San Joaquin Valley. Thus, agricultural users remain the largest single source of demand for water in today’s market, but they conduct their purchases almost entirely through temporary arrangements. Also noteworthy is the highly local character of much of the longterm and permanent transfer market. Transfers involving CVP contractors and water agencies with their own surface supplies have essentially taken place in the neighborhood, between districts and municipalities in the same or adjacent counties. The only cross-regional movements of long-term or permanent water involve SWP contractors and Colorado River contractors.8 In both cases, the transfers came about mainly as part of global renegotiations of project operating rules. Under the Monterey Agreement, Kern County agriculturalists negotiated the sale of contract entitlement to municipal users in Southern California and the San Francisco Bay Area in exchange for greater certainty of supply for their remaining SWP entitlement. Contractors within the Colorado River group have been under intense pressure to set up transfers from the agricultural valleys of the desert counties to coastal municipal agencies as part of California’s obligation to reduce its use of Colorado River water under the 4.4 Plan referred to in Chapter 1. The stronger presence of municipal agencies in the market for longterm and permanent water contracts is logical, given their need to ensure reliability of supplies for growing populations. Looking ahead, we should expect residential demand to increase, as a consequence of demographic trends themselves and recent legislation that makes a tighter connection between water supplies and land-use planning. In 2001, the legislature passed the “show me the water” bills, SB 610 (Costa) and ____________ 8Although technically within the same region (Southern California), most of the Colorado River transfers involve buyers and sellers at quite some distance from one another. The one deal involving close neighbors is the pending transfer from Imperial Irrigation District to Coachella Valley Water District. 23 SB 221 (Kuehl), which require that local governments demonstrate the adequacy of water supplies for growth.9 Long-term transfers are among the measures considered adequate for this purpose. Summing Up Jumpstarted by a prolonged drought in the late 1980s and early 1990s, California’s water market has now become a firmly established feature of the state’s water allocation process. The market remains largely intraregional in nature, with the state’s Department of Water Resources directly mediating most transfers across regions. The market is also highly segmented, with over half of the volumes traded among contractors of the large state and federal water projects and another third involving direct purchases by state and federal agencies for drought relief and environmental mitigation. Outside drought years, when urban agencies have been important buyers, the main sources of demand have been directly and indirectly linked to environmental concerns. Direct purchases for instream uses and wildlife reserves have accounted for over one-third of the increase in purchases since 1995. The other growth sector, accounting for over half of market expansion, has been agriculture in the San Joaquin Valley, as farmers whose contractual water deliveries have been cut because of environmental mitigation programs have turned to the market for replacement water. Municipal agencies are the principal buyers of long-term and permanent contracts. In light of the state’s rapid population growth, it is not surprising that municipal agencies are taking the lead in negotiating long-term and permanent arrangements for water supply. Legislation passed in 2001 requiring that local governments demonstrate adequate water supplies for development should increase urban demand for longterm water transfers. As we shall see, municipalities’ success in forging these deals and assuring new supplies will depend on their ability to smooth the waters of community resistance in the source regions. ____________ 9See Association of California Water Agencies (2002) and Department of Water Resources (2002b). 24 3. The Rise of Local Restrictions on Water Marketing As state and federal authorities have taken steps to facilitate water marketing in California over the past two decades, concerns over potential negative effects of the market on the source regions have prompted many county governments to erect new barriers to trade. This chapter describes the statewide trends in county adoption of groundwater protection ordinances and raises some of the key legal, economic, and operational questions concerning their implementation. The Mobilization of Rural Counties By the end of 2002, 22 of California’s 58 counties had adopted ordinances that restrict the export of groundwater. Although the specific language of the ordinances varies, one common thread is their focus on the regulation of exports, as distinct from groundwater uses on-site. In most ordinances, “exports” are defined as shipments of water beyond the county’s administrative boundaries. Although several counties apply instead an “out-of-basin” definition of exports, and several others an “offparcel” definition, a review of the implementation record suggests that these nonadministrative boundaries reflect an intent to protect the ordinance against potential legal challenges (discussed below) rather than to regulate groundwater use within the county.1 The precursor to this movement was the adoption by three northern counties (Butte, Glenn, and Sierra) of urgency ordinances prohibiting ____________ 1Counties with “out-of-basin” restrictions include Inyo, Kern, Mono, and Siskiyou. Counties with “off-parcel” restrictions include Tehama, San Benito, and Sierra. Imperial County’s ordinance has separate restrictions on exports leaving the county and on groundwater extractions for within-county use. As discussed in Chapter 5, only three counties within this group—Imperial, San Benito, and Sierra—have enforced a permitting process for within-county uses. For details, see Hanak and Dyckman (2003). 25 the “mining” of groundwater in 1977, a year of severe drought. Modoc County followed suit early in the following year, with an ordinance limiting transfers outside the groundwater basin. Over the next 15 years, a handful of Sacramento Valley and mountain counties introduced ordinances with explicit export restrictions. The slow pace may be explained in part by the fact that two counties, Inyo and Nevada, saw their ordinances successfully challenged at the trial court level during the 1980s.2 The floodgates opened once a third county, Tehama, won an appellate court victory in 1994, upholding its authority to regulate groundwater. Since the Tehama decision, which was widely publicized in water law and county government circles,3 14 counties adopted explicit export restrictions for the first time, and three counties regularized urgency ordinances adopted earlier (Figure 3.1).4 Geographically, the group is concentrated in rural California: the mountain counties to the north and east, the Sacramento and San Joaquin Valleys, and Imperial County to the south (Figure 3.2). By and large, these counties fall into the group traditionally considered “source” regions for the state’s water supply; many have relied heavily on groundwater for agriculture. Over this period, five other counties adopted groundwater protection ordinances that focus on management of groundwater resources within the county or in a particular geographic subarea.5 The regulations include various types of restrictions on extraction for on-site use (e.g., well permitting, flow monitoring, pump taxes). In effect, the county ____________ 2Inyo County’s ordinance was successfully challenged by the City of Los Angeles in 1983 (City of Los Angeles, Department of Water and Power v. County of Inyo, case no. 12,908, July 8, 1983). Nevada County’s ordinance, introduced in 1986, was successfully challenged in a suit brought by the Truckee-Donner Public Utility District in 1988. Inyo pursued the matter through negotiations with the City of Los Angeles and reintroduced a revised ordinance in 1998. Nevada County did not seek review and has not reintroduced an ordinance subsequent to the Tehama holding. For this reason, we have not counted Nevada among the 22 counties with export restrictions. 3See Goldsmith (1995a, 1995b) and Bunn (1997). 4In addition, Modoc County introduced a more stringent ordinance restricting exports out of the county rather than the basin in 2000, and Inyo reintroduced an ordinance in 1998 to replace the one invalidated by its court proceedings in the early 1980s. 5For details, see Table B.1, middle column. 26 Number of ordinances per year 5 Export restrictions 4 Tehama County court victory 3 2 1 0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 NOTES: For details, see Table B.1. The dates refer to the adoption of regular ordinances restricting exports. Figure 3.1—Adoption of County Ordinances Restricting Exports assumes an authority resembling that exercised by other specialized groundwater institutions present in the state: basins adjudicated by the courts, special groundwater management districts created through acts of the legislature, and certain special water districts that exercise full control over access to the resource.6 In adjudicated basins, use rights are attributed to individual users in much the same way as surface water rights. In the special districts with groundwater authority, the districts have the authority to regulate individual water use, typically through pump taxes. Counties with these types of groundwater management institutions tend to be located along the coast and in Southern California; most have significant urban populations. The on-site groundwater protection ordinances appear to substitute for or complement the activities of special districts and adjudicated basins. Ordinances substituting for other measures include those of San Diego and Napa Counties; in Napa, ____________ 6For maps showing the location of these institutions, see Figures B.2 and B.3. 27 Export restrictions NOTE: Kern County’s ordinance is limited to the southeast portion of the county within the Lahontan hydrologic region. Figure 3.2—California Counties with Groundwater Export Restrictions, 2002 officials view the ordinance as an alternative to a costly adjudication process. Counties where the ordinances complement other institutions include Monterey, Mendocino, and San Bernardino. In Monterey, the ordinance applies to several zones not already covered by a special groundwater district. Mendocino’s ordinance applies only to the town of 28 Mendocino, where a special district has groundwater authority. The San Bernardino ordinance applies to desert regions of the county not already under a local management system and notably excludes the Mojave Basin, which is adjudicated. At first glance, several of the mountain counties (Lassen, Modoc, Mono, and Sierra) seem to be exceptions to this geographical division because they have adopted both special groundwater management districts and countywide ordinances restricting exports. However, all but one of the six special groundwater districts in these counties were set up with the primary purpose of controlling exports rather than for local use management.7 Siskiyou County also appears as a partial exception, by virtue of the presence of one adjudicated basin. As noted above, several of the counties with export restrictions have ordinances that would, in principle, provide authority to regulate in-county uses as well. However, this authority appears to be exercised actively in only three counties— Imperial, San Benito, and Sierra. An Overview of Export Restrictions All 22 ordinances restrict the direct export of groundwater; those adopted since 1996 also implicitly restrict exports of surface water by regulating the extraction of groundwater used to replace exported surface supplies. Sacramento’s ordinance overtly restricts surface water exports of any kind. A handful of counties (Madera, San Joaquin, and Fresno) have specific restrictions on the use of county groundwater basins as storage sites for groundwater banking projects. San Joaquin County supervisors recently introduced a separate ordinance restricting the location of aboveground storage as well, in response to a proposed project to convert two islands in the Delta into surface storage areas. In most counties, restricted activities require a county permit, which invokes a review under the California Environmental Quality Act (CEQA).8 Permit applicants are expected to conduct from one to several ____________ 7The exception is the Sierra Valley Groundwater Management District, which was initially intended to regulate agricultural pumping in response to drought conditions. 8In four of the five counties with ordinances oriented toward local groundwater management (San Diego, Mendocino, Monterey, and Napa), the process involves 29 studies. Applications go through a multilayered review by county departments and commissions, with the final decision most often in the hands of a political body (the board of supervisors or a body appointed by the board). If approved, permits generally run from one to three years. Most ordinances provide certain categorical and conditional exemptions to the permitting process. Water districts or landowners with service areas or holdings overlying adjacent counties typically do not need permits for water use on those lands. Most counties also provide a blanket exemption to permitting as long as quantities remain within historical use levels. A number of the mountain counties exempt bottlers of spring water, as long as the bottling is done within the county. Finally, various counties exempt specific types of local entities—such as incorporated cities and water districts—from permitting altogether, either because the counties have assurances that the entities already engage in sound groundwater management practices or because there are questions of regulatory authority and an interest in avoiding jurisdictional conflicts. Legal Issues The question of jurisdictional authority is at the heart of the legal issues raised by the county ordinances.9 In California, municipalities may invoke police powers to protect the public welfare in areas not regulated by the state. This is the basis for the groundwater protection ordinances, because groundwater falls outside state jurisdiction. The lawsuit filed in 1992 against Tehama County by two landowners, Baldwin and Myers, challenged this position, arguing that some provisions of the California Water Code already dealt with local groundwater protection, thereby “preempting the field.” Three areas of state intervention were cited: the specially enacted groundwater management districts (noted above); Section 1220 of the Water Code ________________________________________________________ incorporation of a groundwater review or overlay in a regular ministerial process, rather than application for a discretionary permit with CEQA review. 9For a detailed discussion of the legal issues raised by the ordinances, see Hanak and Dyckman (2003). 30 (enacted in 1984), which restricts the direct export of groundwater out of the Sacramento River Basin; and the recently enacted AB 3030 (Water Code Sections 10750–10753.9), which authorizes existing water agencies to create groundwater management districts. The county lost in trial court but prevailed at the appellate level, with a holding that the cited state legislation did not preempt the county’s ability to regulate an arguably open field by establishing a review process in the interest of public health and safety.10 The Tehama case was the first appellate decision to address the issue of county authority to regulate groundwater extraction. As noted, it followed at least two earlier Superior Court rulings that took the opposite position, holding that the state had preempted the power of cities and counties in this domain. Although there have been no further proceedings against county groundwater ordinances, several potential legal concerns were not addressed by the Baldwin v. County of Tehama case. The first of these relates to provisions of the Water Code governing surface water transfers, where it would be difficult to argue that the state has not occupied the field. Sacramento’s ordinance, which openly restricts any surface water exports, would suffer under such a legal challenge. County ordinances that indirectly restrict surface water exports by restricting groundwater substitution may also conflict with the state’s authority in this area. Some ordinances may also be open to legal challenge on the grounds that they exceed the county’s police power. Court decisions have tended to take the view that governments, in exercising this power, should take the minimum steps needed to protect public health, safety, and welfare. In counties where it is possible to demonstrate that there is already a significant problem of overdraft, imposing export restrictions is a classic use of police power, as it protects residents’ ability to exercise their property rights. However, a number of counties are not in a position to justify the export restrictions on the basis of current needs. The widespread exemptions for historic use levels are, in effect, an admission that there is ____________ 10Baldwin v. County of Tehama, 31 Cal. App. 4th 166, 173-74 (1994, 3rd Dist.); review denied, Cal. Sup. Ct., March 17, 1995. 31 no current threat to public health, welfare, and safety. In such cases, applying restrictions could amount to barring present use somewhere in the state to preserve future use in the areas of origin. In addition to potentially exceeding the extent of police powers, an overly protective ordinance may violate Article X, Section 2 of the California Constitution, which requires that “the water resources of the state be put to beneficial use to the fullest extent of which they are capable.”11 Many counties are in the process of incorporating the ordinances into their General Plans as a way of shoring up the basis for the restrictions. In a different vein, there is some debate over whether the ordinances could be subject to a federal Commerce Clause challenge. As noted above, the language in most of the ordinances is jurisdictionally based, restricting exports beyond the administrative boundaries of counties rather than hydrologic basins or some other distinction that reflects the physical links between groundwater extraction and harm to adjacent users. This language raises the potential for a challenge of discrimination based on arbitrary distinctions.12 Although this issue has not been raised in any of the court challenges to date, it has been of some concern for ordinance drafters in several counties, as reflected in the move to an “outof-basin” or “off-parcel” permitting system.13 A final, and arguably more significant, legal concern is that the ordinances generally do not distinguish between native groundwater and imported surface water banked underground. As a result, even ____________ 11In the Tehama case, the 4th Appellate Court rejected this argument, observing in a footnote that the issue “is not so simple as plaintiff’s cursory argument supposes.” Counties may nevertheless have reason for some concern over a potential future legal challenge against “hoarding water.” Note too that the source of county regulatory authority—police power—relates to current threats and not anticipated future threats. 12The case law precedent is Sporhase v. Nebraska ex rel. Douglas (1982), in which the U.S. Supreme Court held that discriminatory groundwater export regulation interferes with interstate commerce (Getches, 1997). 13Counties that moved from a county to a basin restriction include Inyo, Mono, and Siskiyou. Discussions with officials in Inyo and Mono revealed that compliance with the Commerce Clause was a factor in this decision. Kern County drafted its ordinance with basin restrictions for similar reasons, following discussions with Inyo County officials. Constitutionality issues were also a factor in the drafting of the Tehama County ordinance, which restricts use off-parcel rather than out-of-county. 32 ordinances that do not directly address groundwater-banking projects do so implicitly because the county might claim authority to restrict the reexport of water brought in for temporary storage. The legal ambiguity on this question arises from overlapping and potentially conflicting jurisdictions, as the state governs the surface water brought into the county, whereas counties may seek to govern its reexport. The Water Transfer Workgroup convened by the SWRCB in 2000 recommended that county authority over imported banked water be limited to preventing injury to local users, which could arise if the banking project negatively affected either quality or quantity of water available locally.14 Economic Issues The legal basis for county oversight of groundwater reflects the underlying economic justification for regulation of a collective resource. In an unregulated situation, a collective resource such as groundwater risks mismanagement. In general, individual users or water districts will not have the incentive to prevent overuse, with negative consequences for both quantity and quality of the water available in the basin. In the absence of state regulation, there is a strong case to be made for local oversight mechanisms that encourage sound management. Against this background, California’s county groundwater protection ordinances raise two questions from an economic standpoint. The first relates to the scope of the ordinances themselves; the second to whether counties are the appropriate level of local jurisdiction. Concerning scope, the question is whether ordinances whose sole focus is restricting exports can contribute to better management of the groundwater resource. Export restrictions could be a sensible management tool, from an economic standpoint, in two situations. First, when there is uncertainty about the characteristics of the groundwater basin (size of the aquifer, recharge rate, etc.), restricting exports could be a reasonable precautionary measure. Second, in the absence of consensus on mechanisms to manage groundwater use by ____________ 14See the discussion of this issue in the report of the Water Transfer Workgroup (2002). 33 those within the basin, restricting exports at least ensures some upper bound on extraction rates. Even where this precautionary strategy is a sensible first step in groundwater protection, however, it can quickly reach it limits as a management tool. A policy focusing on export restrictions can discourage active management of groundwater basins, a practice that can benefit local water users financially and enhance the regularity of supply in overdrafted basins. Active management can include various programs involving the conjunctive use of groundwater and surface water, such as the intentional recharge of aquifers that have been drawn down and the intentional drawdown of aquifers that are full.15 In most places, active management programs require interactions with water users outside the area—notably through transfers of surface water in lieu of groundwater and through banking of imported surface water. Prerequisites include setting up information systems to better understand the aquifer and establishing some form of local oversight on groundwater use. All of this implies moving beyond export restrictions to a more comprehensive system of groundwater management. In areas where the conjunctive use of surface and groundwater is not feasible because of limited surface supplies, the case for moving beyond export restrictions to active management is admittedly weaker. Even here, however, there can be benefits to understanding the limits and opportunities of the resource base. When groundwater supplies and recharge rates are more than adequate, a well-structured export program could be a boon to the local economy. Are counties the appropriate level of jurisdiction to provide this leadership? The economic literature on the management of collective resources shows that success is greatest when local oversight institutions reflect a high degree of alignment of interests among concerned parties (Ostrom, 1990). The alignment can be geographical, increasing in the extent to which the physical boundaries of the resource management problem coincide with the jurisdiction of the local governing institution. The alignment can also be relational, increasing with the capacity of ____________ 15For a description of conjunctive use programs used in California, see Purkey et al. (1998). 34 concerned parties to participate in or affect the deliberations of the governing institution. In the case of groundwater protection and management, a natural point of geographical organization might be the aquifer—or the larger watershed draining into it. In all but a few cases, these physical boundaries correspond neither to those of the county nor to those of other local institutions, such as water districts. Recognition of this “misalignment” has led to the formation of multiparty institutions to address groundwater and watershed issues in many parts of the state. For groundwater, the most common institutional forms are groundwater management programs created under the AB 3030 legislation noted earlier or under joint powers authority. According to Department of Water Resources records, roughly 20 multiparty programs of this type currently exist. As we will discuss below, some of these programs show promise as a structure for local resource management. Creating new institutions is not without difficulties, however, and there is some merit to considering counties as a potential rallying point for local resource management, even if their administrative borders do not coincide with the limits of the aquifer or watershed. Counties have the merit of having well-established representative institutions and public consultation mechanisms and can provide a convening point for parties. As such, they offer the potential for a high degree of relational alignment. Together with cities, they are also the only local institutions with the authority to invoke police powers, which could be a useful component of a local resource management program. Operational Issues The extent to which the ordinances can play a beneficial role in local water management also depends on the way certain operational issues are handled. Foremost among these is the review process for permitting. In many ordinances, this process is open-ended and involves significant upfront costs. A strong case can be made that those wishing to transfer water should be responsible for conducting the necessary environmental reviews (as is the practice for surface water transfer projects at the state level). In a number of counties, however, the initial requirements appear disproportionately high—effectively calling for a full-fledged study of the 35 aquifer before allowing any transfer project to go through, no matter how small. Whether intentional or not, this requirement can deter those seeking permits. It raises a potential legal issue of undue burden because the first applicant bears the cost of the background studies for all those who follow. In many counties, questions can also be raised about the transparency of the review process and the extent to which the technical characteristics of the project will be considered by the political body empowered to grant permits. Finally, the number of reviewing entities and effective amount of review time create other deterrents. Summing Up In the 25 years since the state adopted a policy in favor of water transfers, a movement to regulate the water market has gained momentum in California’s rural counties. As of late 2002, 22 of the state’s 58 counties had adopted groundwater protection ordinances requiring a permit to export groundwater or to extract groundwater used in substitution for transferred surface water. Although this policy can be justified on economic grounds as a firststep precautionary measure in the face of uncertainty about the resource base, it is harder to justify in the longer run in the absence of a broader water management program. A strictly precautionary policy prevents the water users in the county from actively managing their groundwater resources, a practice that can reap financial and water supply benefits. It also makes counties susceptible to charges of “hoarding water,” which is disallowed by the California Constitution. Because groundwater is a shared resource, active water management requires some form of local oversight. Whether counties or other local institutions are better suited to the task is an open question to which we will return below. 36 4. Why Do Some Counties Adopt Export Restrictions? As we have seen, counties with ordinances restricting water exports are concentrated in California’s inland rural regions—the mountain counties to the north and east and the Sacramento and San Joaquin Valleys in the center. Despite this common denominator, however, there are some clear distinctions within the group. Some counties are agricultural powerhouses; in others, farming is a marginal activity. Some are sparsely populated; others contain large and fast-growing metropolitan areas. Nor is geography a complete determinant: Over a third of the counties in these regions have not adopted ordinances. This diversity suggests that the rural county ordinance movement is not monolithic; a range of factors may influence individual counties’ choices. This chapter explores some of these factors, using two approaches. First, we examine whether there is a statistical basis for predicting which counties adopt ordinances, taking into account characteristics about county economic and water conditions and institutional factors. Second, we use a more qualitative lens, drawing on information obtained from interviews with county officials and water users. This discussion highlights the specific dynamics at play in each of the three regions noted above. Factors That Make a Difference Because the timeline of county adoption is idiosyncratic, we will confine the statistical analysis to a simple “yes” or “no” prediction of ordinance adoption and not attempt to model the factors determining the year of adoption. This leads us to focus on characteristics that vary across counties, measured, when possible, in the mid-1990s, when the adoption movement was fully under way. 37 Data on County Water Economy and Institutions We could expect two structural factors to be important in the decision to adopt water export restrictions: the local economy’s dependency on water and the extent to which exports might compromise the groundwater resource base. To measure the first of these, we have identified four indicators. Two capture the role of agriculture, the primary water-dependent economic activity: the share of farm jobs in total employment and the share of agriculturally related jobs, a category including agricultural services and agro-processing.1 A third indicator, the share of irrigated acreage in total farm acreage, reflects the intensity of water use as an input. Counties with higher proportions of dry-land farming or rangeland will have lower agricultural water needs. The fourth indicator, the share of residential population dependent on groundwater, captures the relative importance of groundwater as a water source. Ideally, we would have included a comparable measure for agriculture, but the share of groundwater in farming is not known in many counties. We might expect all four of these measures to be positively related to the decision to adopt an ordinance restricting water exports. We might also expect counties whose groundwater basins are experiencing serious overdraft to be more inclined to adopt precautionary measures than those whose aquifers are full. Identifying these locations proved difficult. A good measure of overdraft would be the extent to which the water table in an aquifer was declining on a long-term basis. There are no statewide measures of overdraft so defined, in part because the Department of Water Resources does not have access to well data in many areas, in part (and relatedly) because the issue raises thorny legal questions. Designation of overdraft has implications for the rights of groundwater appropriators and can serve as a basis for adjudication proceedings. The last time DWR ventured an official designation of groundwater basins in difficulty was in the 1980 Bulletin 118-80 (Department of Water Resources, 1980). The bulletin provides a list, established ____________ 1See Appendix C for a description of data sources. 38 through a process of data analysis and public review, of 11 basins that are “critically overdrafted” and four basins “with special problems.”2 Although conditions of individual groundwater basins in the state may have changed for better and for worse in the intervening years, this list remains the best indicator of counties with an at-risk groundwater basin. We would expect that counties overlying such basins would be more likely to have adopted some type of restrictions. It is also likely that institutions play a role in the decision to restrict exports. One unquantifiable but integral aspect is the degree of harmony among the county’s institutions. Counties where the water agencies are not in conflict with each other and with the municipal and county governments will be less likely, all else equal, to have the motivation to impose export restrictions as a controlling mechanism. This institutional concordance can result from a range of factors: limited number or low diversity of institutions, overlap of county and water agency governing bodies, or good working relationships borne of individual initiative. Another important institutional factor is an affiliation with the Regional Council of Rural Counties (RCRC), an association of counties with populations generally below 300,000. It began in the early 1970s as an association of the mountain counties and subsequently expanded to cover most of the small counties in the state.3 In 1995, on the heels of the Tehama County court victory, RCRC established a water program to promote the protection of member counties’ water rights. The program encourages members to adopt groundwater ordinances and to reflect these in their general plans. RCRC influence is transmitted in two ways. For some counties, the secretariat has provided technical and legal advice; more generally, the board of directors, made up of a supervisor from each member county, serves as an important forum for information exchange. ____________ 2See Figure B.4 for a map showing the location of these basins. 3As of 2002, RCRC had 30 member counties: Alpine, Amador, Butte, Calaveras, Colusa, Del Norte, El Dorado, Glenn, Imperial, Inyo, Lake, Lassen, Madera, Mariposa, Merced, Modoc, Mono, Napa, Nevada, Placer, Plumas, San Benito, Shasta, Sierra, Siskiyou, Sutter, Tehama, Trinity, Tuolumne, and Yuba. Imperial County officially joined in October 2002 but was informally affiliated with the group for some time before membership. 39 Cross-County Results For several of these indicators, there is a considerable and statistically significant difference in the mean values between counties with export ordinances and those without (Table 4.1). On average, counties with export restrictions have twice the share of farm employment and 40 percent higher residential dependence on groundwater. They are also twice as likely to be members of RCRC. The ordinance adopters also register a third more irrigated farmland and appear nearly twice as likely to overlie a critical or specially designated groundwater basin, although the differences are not statistically significant at the 90 percent level of confidence. The one area where the groups do appear alike is in their share of agriculturally related employment, at roughly 2 percent of all jobs. Several of these factors contribute to the probability of a county adopting an export restriction in the expected ways.4 The most sizable effects are associated with membership in RCRC and the presence of an at-risk groundwater basin. These factors increase the likelihood of ordinance adoption by 30 and 26 percent, respectively. Farm employment also raises the likelihood of adoption. A county with 10 Table 4.1 Average Characteristics of Counties With and Without Export Restrictions Counties With Restrictions Farm employment (%) 8.0*** Agriculture-related employment (%) 2.0 Irrigated farmland (%) 34 Residents using groundwater (%) 75*** Counties overlying critical/special basin (1980) (% of counties in group) 45 RCRC membership (% of counties in group) 77*** Sample size 22 Counties Without Restrictions 3.8 1.9 24 54 All Counties 5.4 1.9 27 62 25 33 36 52 36 58 ***Indicates significant difference of group means at the 99 percent level of confidence. ____________ 4See Appendix C for a more detailed presentation of the statistical results discussed here. 40 percent of its workforce in agriculture is a third more likely to adopt an ordinance than a county with only 2 percent of farm jobs.5 Somewhat surprisingly, agriculturally related employment appears to have the opposite effect. For a given level of on-farm employment, counties with a higher share of value-added activities related to agriculture are less likely to adopt export restrictions. In effect, the geographical link between agriculturally related jobs and farming is less direct than is commonly thought. Many counties with a relatively low share of farming have significant processing activities. The statewide county average ratio of off-farm to on-farm agricultural jobs is 73 percent. Only one county with an export ordinance (Sacramento) exceeds that level. Thirteen counties without ordinances do so— including virtually all of the San Francisco Bay Area and Southern California. Many of these counties have more comprehensive groundwater management systems, which do not discriminate against transfers. Regional Issues To understand the reasons for adopting export restrictions, we interviewed county and water district officials and other resource persons in 36 counties, including most having some type of groundwater ordinance and the key agricultural counties without one.6 The interviews sought background on the following types of questions: Was the ordinance a response to a specific local event or a general precautionary measure? Is the main concern with activities of private individuals or water districts? Which types of prospective buyers raise red flags—farmers in neighboring counties or distant municipal water suppliers? Is the objective mainly to control water exports or groundwater banking projects? Within the county, how contentious was ____________ 5Neither the share of irrigated farmland nor the share of residential dependence on groundwater has statistically significant effects on the adoption of export restrictions. 6The only counties with ordinances for which interviews could not be conducted are Siskiyou and Monterey. Counties without ordinances for which interviews were conducted include Kings, Merced, Placer, Plumas, Solano, Sonoma, Stanislaus, Sutter, Trinity, Tulare, and Yuba. All 58 counties were contacted to verify whether an ordinance was in place or under consideration. 41 the adoption process? Were exemptions granted because of jurisdictional issues or because the exempted parties were considered to have an adequate groundwater management system in place? Although each county has a unique story in some respects, we have opted to highlight the results of these interviews from a regional perspective.7 The three regions singled out have different water supply conditions and a different set of demands for agricultural and residential uses. These factors lend a specific regional character to the nature of the perceived threat the water market brings with it. So, too, do historical events relating to water transfers. Within this context, the weight of the ordinance in a given county will depend on the local dynamics of the adoption process: whether it came in response to a specific event, the degree of internal conflict, and the extent of public involvement preceding adoption. The Mountain Counties: The Legacy of Owens Valley Counties in this region are sparsely populated and have limited agricultural production (Table 4.2).8 Many rely heavily—some exclusively—on groundwater, although rivers and streams provide a source of surface water in others. An indicator of the extent to which groundwater is a local concern is the fact that six of the 15 counties overlie basins identified in Bulletin 118-80 as having “special problems.”9 The history of one of these basins, Owens Valley, has indelibly marked the regional consciousness on water issues. In the early part of the 20th century, the City of Los Angeles bought up vast tracts of land in the valley and began exporting large quantities of groundwater. The transfer had immediate consequences for the local economy, where agriculture became unviable, and has proven over time to be deleterious to the environmental health of the valley. The expansion of pumping capacity in 1970 with the addition of a second aqueduct exacerbated the ____________ 7For a list of counties by region and a map showing county locations, see Table B.1 and Figure B.1, respectively. 8The only county with over $100 million in gross agricultural product in 1999 was Siskiyou, which is part of the Klamath Project. Neighboring Modoc, also in the project, had output valued at just over $60 million in that year. 9Only Calaveras overlies a basin identified as subject to critical overdraft. 42 Table 4.2 Regional Characteristics (County Averages) Population Urban population (%) Agricultural output per capita ($) Residents using groundwater (%) Counties overlying critical/special basin RCRC membership Export ordinances Counties in group Mountain Counties 37,870 0 882 67 7 15 8 15 Sacramento Valley 234,335 (118,161) 41 (35) 3,530 78 San Joaquin Valley 430,075 39 4,366 88 08 82 74 10 8 SOURCES: Population figures for 2002 are from the California Department of Finance. Urban population share is from the 1990 Census, the most recent year for which this estimate is available. Per capita agricultural output is from the 1997 Agricultural Census. See also Appendix C. NOTES: Numbers are county average values for each region. For a list of counties in each region, see Table B.1. For the Sacramento Valley, numbers in parentheses indicate values excluding Sacramento County. environmental problems, both in Mono Lake and in the Owens Lake bed. Following years of difficult and acrimonious legal proceedings, Inyo County and a coalition of environmental groups reached mitigation agreements with the City of Los Angeles in the early 1990s. These have resulted in a considerable reduction in the flow of water out of Mono and Inyo Counties through the Los Angeles Aqueduct.10 The mountain counties also have an acute awareness of their status as source regions for the federal and state water projects and the projects developed by the City of San Francisco and East Bay Municipal Utilities District. Although these projects have not necessarily affected the volumes of water available to local water users, whose needs are limited, ____________ 10The settlement between the City of Los Angeles and the County of Inyo over Owens Valley was reached in October 1991, and the Mono Lake settlement was reached in September 1994 (Hundley, 2001). Together, these agreements have reduced the City of Los Angeles’s exports by an annual average level of 150,000 acre-feet, or one-third of the conveyance capacity of the Los Angeles Aqueduct (personal communication, Jerry Gewe, Los Angeles Department of Water and Power, December 2002). 43 there is a tradition of vigilance in defending the area of origin status, by which counties can reclaim water they may need for future growth.11 Moreover, contention over the regulation of instream flows can at times be significant. One case in point is the current dispute over Central Valley Project appropriations from the Trinity River. The federal government’s decision to reduce off-take, in response to concerns by Native American tribes and environmentalists, has become the subject of legal proceedings with project contractors in the Sacramento and San Joaquin Valleys. A second is the dispute on the allocation of water between instream and agricultural uses on the Klamath River, which has pitted agricultural water districts in Siskiyou, Modoc, and neighboring Oregon counties against environmental and tribal advocates within the same region. Against this backdrop, one might expect these communities to be wary of transferring water. Several of the basins bordering the state of Nevada have been the subjects of intense controversy for just this reason. A project to market groundwater to Nevada from Long Valley, one of the “special problem” basins underlying Lassen and Sierra Counties and neighboring Washoe County in Nevada, prompted the introduction of the state’s first special groundwater management district act in 1980. Since then, potential export projects to Nevada from the Honey Lake Basin (Lassen) and the Surprise Valley Basin (another “special problem” basin underlying Modoc) led county officials to obtain special district status for these basins as well. In addition to the ability to limit exports, the special groundwater management districts have the authority to negotiate directly with the state of Nevada concerning joint basin matters.12 County-level export restrictions have been adopted as an additional safeguard, with little substantive consequence for the management of these basins. ____________ 11This includes the “County-of-Origin” statute (1927; Cal. Water Code Section 10505), the “Watershed Protection Act” (1933, Cal Water Code Sections 11460– 11464), the “Delta Protection Act” (1959), and the “Protected Areas” statute (1984; Cal Water Code Sections 1215–1222). See Hundley (2001), pp. 531–533. 12This authority has been essential, but the border counties nevertheless find themselves at a disadvantage in cross-border basin management negotiations, because groundwater extraction in Nevada is regulated at the state level. 44 Although the potential for groundwater exports to Nevada has also been an issue in Mono and Inyo Counties (as well as San Bernardino, further south), the more-pressing concerns there have arisen from modern-day projects to export groundwater to Los Angeles. Inyo County’s initial ordinance, passed by a ballot referendum in 1980, was part of the county’s attempt to seek mitigation for the ongoing transfers to the City of Los Angeles begun decades earlier. But mitigation of both Mono Lake and the Owens Valley has spawned a new set of concerns. The freed-up capacity in the Los Angeles Aqueduct has sparked the interest of several private firms, which see the potential to use it for conveying water to the coast. The Inyo County ordinance was revised in 1998 to expressly limit transfers through the aqueduct and any sales to Los Angeles. By contrast, in Calaveras and Tuolumne Counties, ordinance adoption appears to have been largely preventive in nature, sparked more by the Tehama ruling and RCRC encouragement than by any specific threat to the groundwater resources. On the whole, the introduction of ordinances in this region has been spearheaded by county officials and has proceeded with little local dissent. The one exception is Calaveras County, where there were protracted negotiations on the terms of the ordinance because of the concerns of local farmers and the main surface water district. The challenges to Inyo’s initial ordinance were raised by landowning “outsiders”—a private company, Anheuser-Busch, which hoped to export water to use in its plants in Los Angeles, and the City of Los Angeles itself, which sued the county and prevailed in the Superior Court.13 Subsequent versions of the ordinance exempt the City of Los Angeles from the permitting process, as its water use is regulated by a separate agreement with the county. Among the counties without ordinances, at least two have alternative institutional mechanisms in place for management. Plumas County overlies the Sierra Basin, for which a special groundwater management district was set up in 1980 to deal with in-basin supply problems, rather ____________ 13City of Los Angeles, Department of Water and Power v. County of Inyo, case no. 12,908, July 8, 1983. 45 than in response to export threats. One commentator assured us, “Were there a need to, the Board of Supervisors could vote in an ordinance in no time.” El Dorado’s only groundwater supplies come from fractured rock—a geological formation that does not lend itself to recharge in the way alluvial basins do. Well permitting has been strictly controlled in the county for many years. Sacramento Valley: A Balancing Act Between the Surface Water “Haves” and “Have-Nots” The Sacramento Valley is an important agricultural region, producing rice, tomatoes, and various fruits and nuts (Table 4.2). It also contains one of the fastest-growing metropolitan areas, centered on the city of Sacramento. Unlike the mountain counties, the valley is simultaneously a source region for the large surface water projects and a major surface water user. The region is also relatively rich in groundwater, which is a primary supply for residential uses and for some farmers. In 1980, not a single aquifer in the valley was on the critical basin list. Although several counties are now concerned about cones of depression, or pockets of overdraft, the valley would still be absent from that list if it were drawn up today. Debates on water marketing are colored by two characteristics of the region’s water supply: the uneven distribution of surface water rights within the valley and the valley’s overall abundance of supplies relative to other parts of the state. The distribution of surface water rights within the valley reflects the historical patterns of water district formation and construction of storage and conveyance facilities. A number of water districts and mutual water companies have senior rights to Central Valley Project and State Water Project deliveries, by virtue of having laid claim to the water before project inception. Most of these senior contractholders are agricultural water users in the south-central and eastern parts of the valley, in Glenn, Colusa, Butte, and Sutter Counties. To the east, Yuba and Placer Counties are also rich in surface water, thanks to autonomous projects with aboveground reservoirs. Districts along the west side of the valley, served by the Tehama-Colusa Canal, have junior rights to CVP water, typically involving both lower per acre allocations and less-reliable supply from one year to the next. In several counties, 46 independent groundwater pumpers—i.e., those not affiliated with water districts—constitute a significant portion of the farming population. “Area of origin” concerns—notably the perception that Southern California’s thirst for the north state’s water would never be quelled— are a longstanding component of the valley’s water lexicon. A landmark event was the 1982 defeat at the ballot box of the project to build the Peripheral Canal, which would have facilitated the movement of water past the San Francisco Bay Delta to southern contractors of the federal and state water projects (Hundley, 2001). Voters in this region (as elsewhere in Northern California) were overwhelmingly opposed to the project, which still looms as a symbol of the need for vigilance on water rights. Soon after this defeat, at the urging of valley and mountain counties, the legislature added Sections 1215-22 to the Water Code to firm up the area-of-origin protections. Under Section 1220, it is illegal to directly export native groundwater appropriated after 1975 outside the Sacramento River Basin without the authorization of the overlying county. To authorize, the county first has to establish a groundwater management plan to ensure that the export does not compromise supplies for local needs. Water purveyors south of the Delta were the intended targets of the section, because earlier area-of-origin statutes provided safeguards only with respect to the state and federal projects.14 It appears that the measure was largely preventive in nature, rather than a response to specific transfer projects under consideration. The multiyear drought of the late 1980s and early 1990s prompted renewed fears about the need to protect native groundwater. One source of the problem was actual transfer activity that occurred under the stateoperated drought water bank. Various water districts and individual farmers in the region participated in the bank in all three years of its operation (1991, 1992, and 1994). In 1994, bank purchases sparked a well-publicized controversy in Butte County, where water districts with senior rights sold some of their surface water and engaged in additional ____________ 14Personal communication, Clyde McDonald, September 2002. Mr. McDonald was a member of the staff of the bill’s author, Senator Norman Walters, at the time of passage. 47 groundwater pumping for irrigation. When wells in parts of the county ran dry, some independent pumpers linked this development to the transfer activity. Because there was no mechanism in place to monitor the effects of the pumping or to mitigate third-party effects, the problem festered and a heated conflict erupted among local water users.15 Although the Butte conflict may well have had repercussions beyond county lines, the state’s considerations about using the region’s groundwater as a longer-term source of supplemental supply for users south of the Delta were probably at least as significant in generating concerns about groundwater protection. In the wake of the water bank experience, the Department of Water Resources began development of a supplemental water purchase program, intended to be a more systematic approach to future water transfers. Initially, the program envisaged annual purchases of up to 400,000 acre-feet of groundwater from Sacramento Valley basins.16 Upon release of the draft document in 1996, local reaction was immediate and sufficiently spirited to condemn the proposal to mothballs. In our discussions with water users and officials throughout the region, the imprint of this proposal remains strong, although the details have faded into the haze. In explaining the reasons for county mobilization, a typical recollection runs something like this: “Back in the early 1990s, DWR decided to put a straw into the aquifer, so we knew we had to do something.” The other well-publicized groundwater transfer controversy, which provoked the drafting of the Tehama County ordinance, demonstrates that exports to points south of the Delta are not the only concern, however. This case involved the direct pumping of groundwater off a small parcel in proximity of the Tehama Colusa Canal, which the landowner planned to use to irrigate his farmland in Colusa County, farther south along the canal. Section 1220 of the Water Code would not be applicable in this type of situation, as it only limits the direct ____________ 15Notably, DWR did not reserve the right in its purchase contracts to require cessation of pumping in the event that effects were encountered. A detailed case study is provided in Thomas (2001). 16Department of Water Resources (1996b), as discussed in Thomas (2001). 48 exportation of groundwater out of the valley, not transfers within the region.17 Thus, Sacramento Valley counties have put ordinances in place to control two types of behavior, depending on the specific water supply issues they face. In several of the west side counties with limited surface water supplies—Tehama, Shasta, and Yolo—the primary target has been direct groundwater exports by private landholders located alongside conveyance facilities, whatever the ultimate destination. In a larger set of counties, the key issue is the dichotomy between water districts with senior surface water rights and other users who rely on groundwater. Concerns have focused on the potential for surface water districts to engage in indirect groundwater exports, by pumping more groundwater and selling their surface water. This problem has played out in quite different ways across the region. In Butte, the controversy went to the ballot box. A coalition of citizens’ groups angry with the water transfers sponsored an initiative to give the county fairly sweeping control over groundwater management, including the potential to introduce pump taxes. In reaction, the county farm bureau and local water districts proposed an alternative ballot measure, limiting the scope of county oversight to exports. It was this second measure, supported by greater campaign funding, that voters adopted in the fall 1996 elections. Although the air has cleared in the six years since its passage, the ordinance-by-initiative has left its mark on local water politics. At the opposite end of the spectrum, Yuba County has avoided the controversy over groundwater substitution altogether thanks to a particular set of institutional conditions. The Yuba County Water Agency (YCWA), the wholesale purveyor to most of the county’s agricultural water districts, has the entire county as its service area. All five county supervisors also serve as directors on the agency’s board. The agency has adopted a policy of immediate mitigation in the event of any ____________ 17The common interpretation has also been that Section 1220 does not apply to indirect groundwater exports via groundwater substitution, hence another argument for county ordinances. 49 well problems that could be linked to groundwater pumping for transfers. In the three other counties with significant potential for groundwater and surface water interaction—Colusa, Sutter, and Glenn—this high degree of overlap between county and water district jurisdictions does not exist. Ordinances were proposed as preventive measures to protect local users from harm. In Colusa, the county took the lead with RCRC support, and the ordinance appears to have been passed with little dissension by water districts. In Sutter, county officials proposed an ordinance as early as 1995 or 1996 but have held off on adopting it in response to water district opposition. Glenn is the only one of the three counties where, as in Butte, the ordinance adoption process reflected a serious disagreement between water users in the county. Although the supervisors unanimously adopted an export ordinance in 1990, a group of groundwater users considered that it did not have the teeth necessary to protect them from potential exports by the large surface water districts on the east side of the county. The group sought a legislatively sanctioned ordinance and succeeded in getting both houses to pass one during the 1992 legislative session.18 The governor vetoed the bill on the advice of the county’s water districts. It took a dramatic change in course by the largest water district, Glenn-Colusa Irrigation District (GCID), to enable a new set of working relations to develop among the county’s water users. Soon after the veto event, an internal management crisis within the district provoked the recall of the entire board and the hiring of new management. With the blessing of the district’s new board and the county board of supervisors, one of the new GCID directors took the initiative to organize an ad hoc, countywide water users group. The point of departure for the group was to find ways for the county’s different types of water users to coexist in harmony. GCID and some of the other senior surface rights districts recognized that there would be situations where they would be solicited and would want to transfer water south, as during the recent drought. The question was how to ____________ 18SB 867, the “Glenn County Groundwater Management Act,” introduced by Senator Mike Thompson in March 1991 and vetoed by the governor in September 1992. 50 engage in transfers without harming other local users dependent on groundwater. After several years of informal meetings, the county board formalized the process and nominated water users representing all the subbasins in the county to a Water Advisory Committee. In the discussions on management alternatives to export restrictions that followed, observers identified two key turning points. First, the irrigation districts came forward with a proposal for a new ordinance emphasizing “safe yield” of the aquifer. This proposal allayed the fears of some parties that there would not be safeguards for protecting the aquifer in the event of groundwater substitution-based transfers. The question then became how to operationalize the concept of safe yield, which is difficult, if not impossible, to measure directly.19 With input from DWR’s Northern Office, the committee developed the ordinance to follow the concept of “basin management objectives” (BMOs). Under this system, water users in the different subbasins would be responsible for establishing a monitoring system and determining target levels for the aquifer under different water conditions. The target levels (or BMOs) are, in effect, a surrogate measure of safe yield. Exports will be restricted only if they lead to unacceptable groundwater levels as determined by the targets. Significantly, pumping for in-county use can also be limited if problems persist after exports are cut back. The Glenn County supervisors passed the new ordinance in 2000, and the first set of basin management objectives was adopted in 2001. These are subject to revision annually as data-gathering and knowledge improve.20 Over this same period, Sacramento County was involved in a similar process, on a much larger scale, to find consensual solutions to its water problems through the Sacramento Water Forum. One outcome has been the creation of a Regional Groundwater Authority to facilitate ____________ 19“Safe yield” is generally used to mean a level of groundwater use that will not cause long-term decline of groundwater levels. Its estimation is complicated because records of groundwater extraction are limited and recharge rates are highly variable from one year to the next. See Dudley (2000). 20For a discussion of the background leading up to the passage of the new ordinance, see Brown et al. (2001). For details on the basin management objective philosophy and process, see Glenn County (2001) and Fulton and Dudley (n.d.). 51 groundwater management across a set of water entities.21 The export ordinance, adopted in 1980, was a response to the 1976–1977 drought and to more general concerns about protecting water rights prevalent in that period. County water officials indicate that for local water management purposes, the Water Forum process has largely supplanted the ordinance. The ordinance, recently transferred from the county code to the statutes of the county water agency, may still be invoked for water exports, however. Following Glenn’s experience, a number of other northern counties are in the early stages of adopting a BMO-based system: Tehama, Modoc, Siskiyou, Lassen, Butte, and Yolo. As with Glenn, this represents the potential for a significant departure from the exportoriented approach of the past. We will return to a discussion of BMOs as a groundwater management alternative in Chapter 7. San Joaquin Valley: Coping with Overdraft and Surface Water Scarcity Favorable soils and climate and the availability of water for irrigation have enabled the San Joaquin Valley to become California’s leading agricultural region. Historically, irrigation was developed with a heavy reliance on groundwater, as surface supplies were less plentiful than in the north. The availability of imported supplies through the federal and state water projects was a major boon to the valley’s agriculture. Part of the motivation for these infrastructure investments was to remediate the problem of groundwater overdraft in the valley. In 1980, most of the valley’s groundwater basins were classified as “critical,” a result of decades of agricultural pumping in excess of natural recharge rates. All eight counties overlie at least one of these basins (Table 4.2, Figure B.4). Concern with overdraft has spawned active conjunctive use programs in various parts of the region, whereby water districts take advantage of the higher levels of surface supplies in wet years to recharge the aquifers. The longest-running program of this nature is conducted by the members of the Friant Water Users Association, along the east side of the valley. Overdraft has also provided the opportunity for groundwater ____________ 21Thomas (2001) and McClurg (2002) provide detailed case studies. 52 banking. Kern County water districts have been the most active in this area. Although water tables have improved in some places since 1980 as a result of natural and artificial recharge,22 the region has been under new pressures because of environmental restrictions on Delta pumping. The reduced supply of imported project water, particularly to the CVP agricultural contractors on the west side of the valley, has been an important factor in the growth of the state’s water market. Meanwhile, ensuring adequate water supply for municipal and industrial uses is becoming a bigger issue, as the valley is now one of California’s fastestgrowing regions. Over the next two decades, its population is expected to grow by 51 percent, or 1.7 million inhabitants.23 The only region growing faster is the Inland Empire (Riverside and San Bernardino Counties). Higher populations have also meant that most of the region’s counties do not qualify for RCRC; only Madera and Merced are members. Ordinances restricting exports have nevertheless been proposed in six counties and adopted in four. The debates on adoption have been flavored by the problems of overdraft and surface water scarcity. San Joaquin County put in place an ordinance in 1996 over concerns that groundwater transfers by farmers along the Delta Mendota Canal, in the southwestern portion of the county, had contributed to overdraft of that basin during the early 1990s drought. In the years that followed, ordinances motivated by similar concerns were proposed in Stanislaus, Tulare, and Fresno. In both Stanislaus and Tulare, where the proposals do not appear to have come in reaction to specific transfer activity, water districts successfully argued that they were already engaged in adequate groundwater management practices. In Tulare, most districts are members of the Friant Unit, and there is an active multiparty groundwater management plan linking a number of these.24 In ____________ 22This is the case in Kern County, for instance (Kern County Water Agency, 2002). 23Projections are for the period 2000 to 2020. See Hans Johnson (2002). 24Deer Creek and Tule River Authority, whose groundwater management plan was introduced in 1995 53 Stanislaus, several multiparty groundwater management plans and associations were in place at the time of the proposal.25 In Fresno, the impetus for an ordinance was an impending permanent transfer of surface water. After prolonged public discussions, the districts managed to gain concessions but not to forestall its passage. At stake was a transfer of CVP contract entitlement from a small water district on the west side of the valley. The Widren Water District had agreed to sell its entire entitlement (just under 3,000 acre-feet) to a real estate developer for a new residential development near the town of Tracy, in San Joaquin County.26 County officials raised concerns that this agreement might lead to increased groundwater pumping to replace the transferred surface water. Although the amount of water was small, county officials feared that the transfer could set a bad precedent. The Widren deal was blocked, following litigation under CEQA and negotiation, but it convinced the board of supervisors to press for an ordinance to prevent further transfers out of the county. The Fresno ordinance went through over two dozen formal drafts before adoption. In particular, water districts did not want restrictions on their ability to engage actively in the water market, on which many farmers depended as a supplemental source of supply. The version ultimately adopted in 2000 reflects many of these concerns and makes the Fresno ordinance unique. In addition to the standard exemptions for use by districts with lands in bordering counties, it categorically exempts both water exchanges and temporary transfers. It also provides for a tenyear permit, whereas the other ordinances require renewal every one to three years. Although the initial impetus for the Fresno case was the potential for groundwater substitution, a number of observers share the view that the ultimate target is to retain existing surface supplies within the county, whether or not substitution is involved. While the Fresno ordinance was being debated, another permanent transfer of entitlement was proposed ____________ 25On the east side of the county, two local groundwater associations, including most water districts and municipalities; on the west side, the Northern Delta Mendota Groundwater Basin management plan, including five CVP contractors. 26Campbell (2000) provides a detailed discussion of the background to this ordinance. 54 involving Mercy Springs Water District, a neighbor to Widren. Proposals have also been aired concerning the sale of entitlement from Broadview Water District, in the same area. All three of these districts overlie lands affected by drainage problems, which severely reduce their agricultural productivity and the likelihood of groundwater pumping. If the county attempts to block future transfers of surface entitlement, districts may argue in court that the county is overstepping its authority. However, the net effect of county opposition to transfers so far has been to keep most of the water within the county.27 In Madera and Kern, the two remaining counties with ordinances, the reasons for adoption were atypical for this region. The Madera ordinance was introduced to provide the county with oversight for groundwater banking projects. The controversial “Madera Ranch” project that mobilized county residents and officials involved the ranch owner, the U.S. Bureau of Reclamation (USBR), and several Central Valley Project contractors.28 The project generated widespread county opposition over potential negative effects to local water quality and supply. Some parties also believed that the primary project participants were intentionally withholding information from local stakeholders. Although the Madera ordinance also regulates both direct and indirect groundwater transfers, these are generally not perceived to be major issues in the county. The two surface water districts that are members of the Friant Water Users Association were granted exemptions, as was the City of Madera. At first glance, the appearance of an export ordinance in Kern County may seem surprising, given the major role water agencies there have played in the development of the state’s water market over the past decade. The Kern County Water Agency and districts such as ArvinEdison Water Storage District, a Friant Unit member, have actively facilitated water exchanges and transfers and have been leaders in the ____________ 27There is still no long-term deal concluded for Widren’s water, which is transferred annually to other agricultural users in Fresno. Part of the Mercy Springs entitlement will pass over to users in Santa Clara and Santa Cruz Counties, and part will stay in Fresno (see Table A.6). The county is also encouraging prospective Monterey County purchasers of Broadview’s water to take only part of the supply. 28For case studies, see Campbell (2000) and Thomas (2001). 55 groundwater banking movement that has already helped ensure dry-year supplies. Closer examination reveals that the ordinance has nothing to do with water management in the San Joaquin Valley part of the county. Rather, the reasons for Kern County’s ordinance hearken back to the water marketing concerns in Inyo and Mono. The southeastern part of Kern overlies the Lahontan Basin, a hydrologic area that also encompasses Inyo and Mono Counties. The ordinance was developed in response to concerns over a proposal by a private water marketing company to buy a former alfalfa ranch overlying the basin and transfer 10,000 acre-feet per year over a 40-year period to the City of Los Angeles through the Los Angeles Aqueduct. In light of the low rainfall and lack of surface water in this desert and foothill area, people worried that Los Angeles would “recreate an Owens Valley in Kern County.” Following a stopgap emergency ordinance, a permanent ordinance was rapidly put in place to require environmental review of any groundwater transfer out of the basin. This was passed with the approval of the Kern County Water Agency, and there was never any serious consideration of extending the ordinance to cover the San Joaquin Valley portion of the county. Only two valley counties, Kings and Merced, have never had formal consideration of export ordinances. One observer suggested that Kings County has not felt the need for one because there is a relatively harmonious atmosphere among water districts, most of which operate with a fairly high level of groundwater monitoring and management. There is, notably, an active multiparty groundwater management plan for users overlying the Tulare Lakebed. Another observer noted that one local agency, the Kings County Water District, has had its own ordinance to prohibit groundwater mining since the 1950s. Like the Yuba County Water Agency, this district has board members who are also county supervisors. Merced County has benefited from a highly effective water planning approach, spearheaded by the county’s two main purveyors—the City of Merced and Merced Irrigation District. In the mid-1980s, dissatisfied with the ten-year planning horizon required by CEQA, the city decided to do a 40-year general plan. The plan raised questions over whether the groundwater system in place would be capable of sustaining the much 56 larger city of the future. The city began discussions with the irrigation district, and in 1991 the two parties embarked on a regional water supply planning exercise, with considerable input from outside consultants for the modeling work. A Technical Advisory Committee with representatives of county departments, other cities, the large water-using industries, USBR, the regional water quality board, and the state and federal environmental agencies provided regular input into the planning process. The Merced Basin regional water supply plan was completed in the mid-1990s and provided the basis for a long-term regional conjunctive use program that could meet the basin’s needs.29 Following its release, the two lead agencies spearheaded the development of an AB 3030 groundwater management plan, involving 15 water purveyors in the basin. Founded in 1997, the Merced Area Groundwater Pool Interests, or MAGPI, has a board of directors that meets quarterly and a Basin Advisory Panel involving a wider range of stakeholders that meets monthly. Initially, the group focused on establishing an effective monitoring system for the basin. It has also begun investigating conjunctive use projects in association with the Department of Water Resource’s Integrated Storage Investigation program. This process, both highly participatory and involving a great deal of technical input, has preempted the need for alternative protective mechanisms in Merced County. Summing Up A statistical analysis of California’s counties suggests that their likelihood of adopting export restrictions varies with economic, water supply, and institutional characteristics. Counties with a higher share of the workforce in agriculture are more likely to impose restrictions. Other key factors are the presence of a groundwater basin designated as “critical” or having “special problems” by the Department of Water Resources in 1980 and membership in the Regional Council of Rural ____________ 29The water plan was updated in the late 1990s in collaboration with the University of California, in connection with the plans to establish UC Merced. 57 Counties, an association that has promoted groundwater ordinances since the mid-1990s. A qualitative review of the reasons for ordinance adoption suggests a strongly regional character to the patterns observed. Many mountain counties have responded to specific threats of long-term groundwater exports from their basins to Nevada and to the Los Angeles region. Ironically, the mitigation settlements for Owens Valley and Mono Lake have kept the threat of new groundwater exports very much alive because of freed-up conveyance capacity in the Los Angeles Aqueduct. The menace of uncontrolled private groundwater pumping for sale to Los Angeles also motivated Kern County’s ordinance, which covers only the southeast portion of the county adjacent to Inyo. Elsewhere in the San Joaquin Valley, ordinances have responded to concerns about transfer and banking operations in overdrafted basins and in an overall context of surface water reductions. In the Sacramento Valley, a major issue has been local disparities in the distribution of water rights—notably the ability of senior surface water-rights-holders to engage in groundwater substitution while others risk shortages. In general, Central Valley counties without ordinances have benefited from greater institutional cohesiveness on water matters. Countywide, umbrella water agencies have played key leadership roles in both Yuba and the San Joaquin Valley portion of Kern County. Elsewhere, substantial efforts have been undertaken to create new institutions. In Merced, an initiative begun by the city and the largest water district in the early 1990s has led to a regional water planning exercise and an active groundwater management plan involving all players overlying the Merced Basin. In Glenn and Sacramento, stakeholder consultation has produced programs to actively manage the groundwater basin. In Glenn’s case, this process culminated in the adoption in 2000 of a new county ordinance based on basin management objectives, which has effectively supplanted the exportpermitting ordinance on the books since 1990. A number of other counties are now following Glenn’s lead. 58 5. Water Market Effects: Do County Restrictions Have Teeth? What effects, if any, are county policies having on the state’s water market? Because most ordinances address both direct groundwater exports and surface water exports that result in additional groundwater extraction, we might expect significant overall market effects. In counties with restrictions, the only types of transfers not subject to local approval are those involving water held in surface storage (available to few water purveyors), excess surface water (available mainly in very wet years), or water conserved through efficiency gains or land fallowing (available only through investment outlays or forgone crop income). If the ordinances reflect a public view that out-of-county sales from any source should be discouraged, even these types of transfers could be affected. Of course, it is also possible that county restrictions have little practical consequence if enforcement efforts are ineffective. This chapter assesses the market effects of export restrictions from two angles. First, it draws on information from county-level interviews to evaluate ordinance implementation. This discussion focuses on how the permitting process has worked for out-of-county transfers. Second, it looks at the effects of the restrictions on county water marketing activity, drawing on the database on water transfers presented in Chapter 2. Export Permitting: A Largely Uncharted Territory In counties with ordinances restricting exports, those wishing to engage in the restricted activity must obtain a county permit, which invokes review under CEQA. The permit-granting authority—in this case, the county—determines the extent of CEQA review. CEQA obligations can be met with a relatively simple environmental assessment, 59 if the county is satisfied that the permitted activity is likely to have little or no negative effect, or if adequate provision is made for mitigation. The county can also require a full-scale environmental impact review (EIR), in addition to preliminary studies. Moving from an environmental assessment to an EIR easily increases costs for the applicant by a factor of ten or more, with EIRs typically costing $300,000 or more. In the years since counties introduced a review process for water exports, there have been few requests for permits and even fewer permits granted. Among the 22 counties with export ordinances, a total of 16 export permit applications have been tendered in four counties: Inyo (12), Mono (2), Tehama (1), and San Joaquin (1).1 Of the 16 applications, only one has been granted, in Tehama, and four remain actively on file, in Inyo. Most of the applications from Inyo and Mono have concerned private sector transfers to Los Angeles. The permit granted in Tehama was for the movement of a small quantity of groundwater to land owned by the same farmer in another county. The application in San Joaquin was for the reexport of imported banked groundwater by the East-Bay Municipal Utilities District in 1997.2 After a protracted review process, the permit was denied in response to concerns about granting a noncounty entity access to the aquifer. A permit for groundwater banking was nearly filed in Madera by a private water firm, Azurix, a partial subsidiary of the Enron Corporation, which intended to pursue the Madera Ranch project that was at the origin of that county’s ordinance. Whether Azurix’s sale of the property in 2001 without filing ____________ 1As noted in Chapter 3, three counties in this group have actively required permitting for some in-county water movements. In San Benito, there were five applications for small residential subdivisions, of which three were approved and two withdrawn. In Sierra County, one permit was granted for a transfer of treated wastewater from an industrial property to adjacent property in agricultural use. In Imperial County, 19 conditional use permits have been granted since 1994 under the well-permitting process instituted in 1972. Recently, Imperial County’s Board of Supervisors rejected a groundwater use petition that had been pending for eight years for use on a farm near the San Diego County border. Among reasons for the rejection were apparent concerns that the applicants might subsequently transfer some of the water to the Borrego Water District in the neighboring county (Mitchell, 2003). 2For a detailed case study, see Thomas (2001). 60 was more a result of its own financial problems or the continued local controversy remains a matter of speculation. A broad local coalition of interests was against both the project and the involvement of an outside firm, and popular sentiment was prominently displayed on billboards along Route 99 admonishing Enron/Azurix to stay out of the valley. In two other counties, those interviewed indicated that permit applications might be on the horizon. In Calaveras, where the ordinance is quite recent, the Calaveras County Water District is investigating a potential conjunctive use project and intends to launch an application if the study confirms the project’s viability. In Sacramento, officials anticipate that a groundwater export being considered as part of a water quality mitigation program may trigger a permitting requirement. The limited permitting experience in counties with export restrictions stands in stark contrast to that of counties whose groundwater protection ordinances were put in place to regulate withincounty uses. In San Diego, Napa, and Mendocino, the review process has been active, with numerous approvals granted.3 When asked to account for the relative dearth of permitting activity in counties with export restrictions, observers offer three explanations. The first holds that there is little local interest in transferring water outside the county anyway; the ordinance is merely an expression of the popular consensus. The second explanation is that the ordinance may be ineffectual in screening exports that continue to occur. This could arise either through lack of public awareness of the permitting requirement or lack of good surveillance to ensure compliance of those inclined to avoid permitting. The third and most common view holds that the permitting process itself discourages transfers. Up-front costs of environmental review, the likelihood of rejection in places hostile to transfers, and the limited length of time for permits once granted are all cited as deterrents. In a number of counties, officials admit that this “discouragement factor” is at least partially intentional. Legally, counties cannot prohibit exports, but they can make it difficult for potential sellers to obtain approval. ____________ 3San Bernardino’s ordinance was approved late in 2002, too recently to have established a record on permitting. We were unable to obtain information on the implementation of Monterey’s ordinance. 61 In principle, the information requirements and up-front study costs depend on the hydrological characteristics of the basin and the extent to which a knowledge base has already been established through prior studies. In practice, another consideration is the degree of public confidence in the process. In places where prior experiences have engendered a distrust of water transfers, the up-front requirements are bound to be higher than where there is no particular local history. A case in point is Butte County, where water officials would like to see a permitting process that effectively screens transfers for third-party effects without blocking the possibility of water marketing. In light of the heated controversy surrounding the adoption of the ordinance, however, it is likely that the first test case would need to go through a full-scale environmental impact review. The cost might be worth it for a long-term transfer program, but it virtually precludes anyone wishing to engage in short-term transfers from applying. Faced with the opportunity to sell water under DWR’s dry-year purchase program in 2001, farmers in the Western Canal Water District elected to generate surface water through land fallowing rather than applying for a permit to engage in groundwater substitution. As it happened, rice prices were sufficiently low that year that some farmers found it worthwhile to forgo crop revenues. More generally, it might have been better for all parties to consider groundwater substitution, but the district was effectively prohibited from pursuing that option. Effects on the Water Market The Butte example highlights the fact that the ordinances will not always limit trade in water if alternative means such as fallowing are available and acceptable to farmers. Because this will not always be the case, however, ordinances limiting the use of groundwater for transfers could have aggregate effects on the market. These effects would be compounded if the ordinances have a more general dissuasive effect on water exports from any source, which appears to be the intent in some counties. Aggregate market effects of two types might be expected. In counties where there are willing buyers locally, the ordinances might be expected to shift supply from exports to in-county uses. In counties without local 62 demand, or where the local demand is more limited than the potential supply, there would be an overall reduction in the volume traded. The alternative prediction—of no effects on the volumes of total transfers and transfers leaving the county—would correspond to a scenario where the ordinances are generally ineffectual. Data Sources To examine whether there is evidence of these effects, we will employ the database on water transfers in California presented in Chapter 2.4 The analysis covers the 12-year period beginning in 1990, the point at which data on counties of origin and destination become more precise. The aim will be to determine the effect of export ordinances on two county-level measures of water trading: annual sales and annual “exports,” defined as transfers going to users outside of the county. We will focus the analysis on two geographical groups. The first is the set of 34 “water trading” counties—counties that appear at least once in the transactions database over the period under review (Figure 5.1). This includes all 18 Central Valley counties, the Southern California region, the inland portion of the San Francisco Bay Area, and San Francisco itself. It excludes the mountain counties and counties along the north and central coast. These nontrading counties are excluded for statistical reasons; the key econometric models cannot be estimated when they are included. Moreover, there may be structural reasons for the lack of trading activity in many of these counties: Counties along the coast and to the far north rely on local river and groundwater sources and are not hydraulically connected to the state’s main water arteries. It is possible that some local trades occur in these regions that we have not been able to trace with our sources. The second geographical group is the set of 18 Central Valley counties. It is interesting to look separately at this region for two reasons. First, Central Valley counties have been the major source of water for the market since the early 1990s. Second, this region has the greatest potential for groundwater-related transfer activity, both through ____________ 4For a detailed description of data sources and methods used in the analysis, see Appendix D. 63 Transfer activity Figure 5.1—California’s Water-Trading Counties, 1990–2001 groundwater substitution transfers and through groundwater banking.5 The potential role of groundwater in this region stands in stark contrast to the situation in Imperial County, the other major water supplier. Although Imperial has imposed groundwater export restrictions since ____________ 5See Purkey et al. (1998) and related work from the Natural Heritage Institute on the potential for groundwater banking in the Central Valley. 64 1996, these restrictions are unlikely to have much practical influence on the overall volume of water sales. Imperial County has vast quantities of surface water rights from the Colorado River and few areas with usable groundwater.6 To see whether the presence of an ordinance affects county trading behavior, it is necessary to control for other factors that might also be important. Multiple regression techniques allow us to isolate the effect of the export restrictions while holding the influence of other factors constant. We have selected a set of seven variables—described in Appendix D—to account for agricultural and residential demand and water supply conditions. In addition, we will test for the effects of state and federal policy changes on the trading environment by including a time trend in the regressions. If the new operating rules are effectively increasing transfer activity, above and beyond what would occur because of changes in the other variables noted above, this variable should be positively related to sales. A time trend also captures the effect of “learning-by-doing” by water users as they gain familiarity with the market. Results County regulations have noticeably restricted sales in the statewide market. In any given year, the typical county with an export restriction sold 14,308 acre-feet less than a county without one. The estimated effect on exports is larger—at 16,948 acre-feet—although the difference is not statistically significant. These results suggest that the county restrictions have not only reduced sales but have also resulted in some shifting of water from external to within-county buyers, by an average of 2,640 acre-feet per year. The quantitative effect of export restrictions is even more pronounced when the analysis is restricted to the 18-county Central Valley sample. Holding other factors constant, export ordinances reduce overall sales by 20,789 acre-feet and out-of-county ____________ 6In large parts of Imperial County, the groundwater is too saline for agricultural uses. Salinity is also a limitation for groundwater use in San Benito County, the other non–Central Valley county within the “trading counties” group that has an export ordinance. 65 sales by 26,245 acre-feet. The corresponding increase in local sales induced by ordinances is nearly 5,500 acre-feet per county per year. Since 1996, the point at which a number of counties began to adopt ordinances, the restrictions have reduced exports by 932,000 acre-feet, or 19 percent of all out-of-county sales (Figure 5.2). Of this total, 145,000 acre-feet that would otherwise have been exported have been sold locally. The lion’s share (787,000 acre-feet) has simply been kept off the market. In all, this represents a 14 percent reduction compared to the level of predicted sales in the absence of county restrictions. For the Central Valley counties, these effects are even larger: a 39 percent reduction in exports and a 25 percent reduction in overall sales because of restrictive ordinances. The resulting shift from exports to the local market appears to have increased within-county sales by nearly 50 percent. Meanwhile, state and federal measures to improve the trading environment, as measured by the time trend, have had a substantial positive effect on water sales. The typical county was likely to sell 42,000 acre-feet per year more in 2001 than in 1990, under identical conditions of water supply and agricultural demand. During the first six years analyzed here, the positive effect of state and federal support far outweighed the negative effect of county restrictions (Figure 5.3). As the 7,000 6,000 5,000 Actual sales Sales if no export restrictions Acre-feet (thousands) 4,000 3,000 2,000 1,000 0 All sales Exports Within-county sales Figure 5.2—Market Effects of Export Restrictions Since 1996 66 Acre-feet (thousands) 900 800 700 600 500 400 300 200 100 0 1990–1995 Increases in sales from state and federal policy Reductions in sales from county restrictions 1996–2001 Figure 5.3—Market Effects of State and County Policy Environments number of counties with restrictions has grown, this has ceased to be the case. From 1996 to 2001, county restrictions cancelled out the positive effect of state and federal policies to encourage trade. Counties introducing export restrictions may have reduced their trading activity for reasons other than the ones captured in our model. In this case, the strong correlation we observe between ordinances and the reduction in sales and exports would not imply causality. In light of the reasons given for ordinance adoption, however, we regard this as unlikely. The bigger question is whether the ordinances will continue to wield the same effect in the years ahead. The pending long-term transfers from the agricultural to the municipal contractors of the Colorado River Project, negotiated as part of California’s program to reduce its overall use of project water, will substantially increase overall volumes traded statewide. The combined transfers from Imperial Irrigation District to San Diego (200,000 acre-feet), Imperial to Metropolitan Water District of Southern California and Coachella Valley Water District (100,000 acre-feet), and Palo Verde Irrigation District to MWDSC (111,000 acrefeet)—achieved through a combination of land fallowing and efficiency gains—represent over 30 percent of the amount traded in 2001. This 67 shift of activity by water users in Imperial County, which has an export ordinance, is bound to alter the statistical relationship between trades and county restrictions in the statewide market.7 For counties in the Central Valley, the other main source of market supply, nothing on the immediate horizon suggests a weakening of the effect of export restrictions. Some Sacramento Valley observers have suggested, however, that the restrictions could loosen under another major and prolonged drought. With external pressure to make water available, counties may have little choice. A more positive impetus for change lies in the possibility that counties will move beyond export restrictions to a wider groundwater management system, thereby ensuring protection of local users without discouraging market activity. The Glenn County experience in 2001 indicates how this might work. That year, several of the county’s CVP settlement contractors participated in a program to send water to Westlands Water District, using a combination of fallowing and groundwater substitution to free up surface water supplies. The potential groundwater effects of the transfer were monitored through the county’s new basin management objectives system as part of the more general monitoring program being established by the county’s water users. In conjunction with the export, Glenn-Colusa Irrigation District also made surface water available to some water-short entities within the area, at a lower price. Summing Up Have county-level groundwater export restrictions had an effect on the water market? The record on permitting suggests that they have. The very low number of permit applications supports the view that this process is more useful as a deterrent than as a screening mechanism. ____________ 7As an indication, we incorporated all pending long-term transfers listed in Table A.5 to the sales and export data for 2001 and reran the regressions for the 34-county sample. With just one year of the new Colorado River Project trades (assumed at full volume), the estimated effect of county restrictions diminishes in size and statistical significance. 68 High up-front costs and the likelihood of negative public opinion guiding the decision process are factors discouraging parties from filing. As the recent experience in Butte County demonstrates, a lack of groundwater permits will not necessarily block transfers if alternatives such as fallowing are available and acceptable to farmers. In the aggregate, however, there is likely to be a market effect, both in reducing total sales and in shifting some water to in-county users (who will typically be willing to pay less than outsiders). We find evidence of both effects in a statistical analysis of county trading behavior from 1990 to 2001. In any given year, the presence of an export restriction reduced a county’s trades by 14,300 acre-feet and shifted 2,640 acre-feet to incounty buyers. In aggregate terms, this reduced out-of-county sales by 932,000 acre-feet, or 19 percent, and total sales by 787,000 acre-feet, or 14 percent, since 1996. Overall, the negative effect of county restrictions cancelled out the positive effect of a generally improved trading environment resulting from state and federal regulatory changes. Looking ahead, the key question is whether communities can move beyond an export restriction mode to one combining protection of local users with the flexibility to allow water trading where feasible. In areas where the groundwater effects of trade are the concern, this means establishing systems for monitoring, mitigation, and actively managing the aquifer. Land fallowing, a key alternative source of water for the market, also raises questions of harm to local communities. As the economics of the water market make this an increasingly attractive option for some farmers, there is a need to consider how and when economic mitigation programs will be a necessary component of water transfers. The following chapter addresses this issue. 69 6. Mitigating the Economic Effects of Land Fallowing Land fallowing has been the basis for several major water transfer programs in California since the early 1990s and is an integral feature of the active agricultural water market within the San Joaquin Valley. Noncontroversial in some situations, the concept of idling farmland to sell water has generated tremendous conflict in others. At issue are both equity questions and the potential aggregate consequences of fallowing. In this chapter, we address the economic, legal, and institutional issues that fallowing for the water market raises, with a focus on the following questions. Under what circumstances is it important to address the distributional consequences of the transaction? What do we know about the conditions under which fallowing would cause harm to the local economy, and what policy alternatives are available to minimize negative effects? We begin with a review of California’s recent experiences with land fallowing for the water market. California’s Recent Experiences with Land Fallowing DWR’s Dry-Year Programs Fallowing was a major component of the 1991 drought water bank, accounting for 415,000 acre-feet of the 821,000 acre-feet purchased. Most of the contracts were negotiated directly with individual farmers, who were paid not to irrigate and were compensated on the basis of the imputed water savings. In Yolo County, where a substantial part of the fallowing occurred, the board of supervisors took the view that the state, as purchaser, should indemnify the county for the losses it incurred as a result of the decreased economic activity. Notably, the county observed an increase in demands for unemployment-related social services as a consequence of lower farm labor employment. Challenging both the 71 legality of the claim and the facts on which it was premised, DWR declined to pay the county the modest amount requested.1 It also cut the fallowing program short. In the 1992 and 1994 banks, water was purchased entirely from surface storage and groundwater exchange. Over time, however, the Yolo episode has led DWR to approach the fallowing question somewhat differently. When it launched a new trial run with fallowing during its 2001 dry-year program, DWR made arrangements to pay a 5 percent fee ($3.75 per acre-foot) to Butte County to handle the associated mitigation costs of the transfer by Western Canal Water District. As we will discuss below, this raised both practical and legal issues at the county level that are still being worked through. It nevertheless appears that the state’s current default position is to compensate for third-party effects of fallowing for its own water purchases. Mitigating the effects of fallowing is also a central focus of the programmatic environmental impact review now under way for the Environmental Water Account, a joint state-federal program. Long-Term Fallowing Along the Colorado River The year following the drought water bank experience, MWDSC launched a two-year fallowing program with the Palo Verde Irrigation District (PVID). Under the trial program, the district idled land to free up 93,000 acre-feet of water per year for MWDSC. Encouraged by the outcome, the two parties then developed a long-term transfer arrangement, under which PVID farmers will fallow up to 29 percent of their 91,000 irrigated acres of land, to send up to 111,000 acre-feet annually to the coast. The 35-year transfer was in the final stages of approval by the boards of both agencies in late 2002 and should get under way in 2003. Both the PVID board and area farmers regard the transfer as an opportunity for the area’s economy on the grounds that it helps stabilize farm incomes.2 A $6 million fund has been set aside to ____________ 1The bill submitted by the Yolo County Board of Supervisors was for $129,305 to cover general assistance and Aid to Families with Dependent Children entitlements for 450 persons deemed to have lost their employment because of the water transfers (Gray, 1994b). 2See “Rural Palo Verde Valley Agrees to Colorado River Pact” (2002) and Lyn Johnson (2002). 72 compensate the community for potential income losses from lower agricultural activity. At the same time that Palo Verde farmers and the PVID board were lauding the forthcoming fallowing program in the local press, representatives of the neighboring Imperial Irrigation District (IID) were going on record with their doubts over whether they would vote for a similar program for sending Colorado River water to San Diego. Unlike the Palo Verde deal, the Imperial deal with San Diego has been one of the biggest water controversies in recent California history. Palo Verde’s transfer was premised on land fallowing, but this method was essentially imposed on Imperial because environmental constraints precluded the preferred option of conserving water through more efficient irrigation practices. Like some other local agencies, IID had a policy disallowing fallowing as a source for water transfers. Accordingly, the initial transfer deal, agreed to in 1998 by the IID and San Diego County Water Authority boards, explicitly ruled out fallowing. It was not until early 2002, during the environmental review phase of the transfer, that objections were raised to the efficiency-based method. By reducing the district’s agricultural runoff, the irrigation improvements would hasten the increase in salinity levels in the Salton Sea, a major aquatic bird sanctuary along the Pacific flyway. By this time, the transfer had become a linchpin of California’s 4.4 Plan to reduce its use of Colorado River water over a 15-year period. Under the threat that the U.S. Secretary of the Interior would immediately reduce California’s annual supplies from 5.2 to 4.4 million acre-feet if parties within the state did not reach an agreement by the end of 2002, IID was pressured to consider land fallowing as an alternative means of conserving water for the transfer.3 ____________ 3Technically, the agreement that needed to be reached among California’s Colorado River contractors by the end of 2002 is the “Quantification Settlement Agreement,” under which the senior agricultural contractors—Palo Verde, Imperial, and Coachella Valley Water District—agree to “quantify” their water rights to a specified amount. Up to now, in the order of seniority, these districts have had the right to any amount of water they can put to beneficial use, within the overall allocation available to the state. The lack of a firm upper limit on use has meant that any transfer deals concluded between any one of the parties and a more junior rights-holder (notably Metropolitan Water District of Southern California and San Diego) would not necessarily result in a reduction of water 73 Negotiations over this transfer have been extremely complex, in part because the fate of the Salton Sea is uncertain, even without the loss of runoff from Imperial’s farms. Created in 1905 when massive river flooding broke through a canal erected by area farmers, the sea has relied on agricultural runoff from the district ever since as its primary source of replenishment. Even without the transfer, the sea is predicted to become too saline to support the fish and other marine life on which the birds feed within one to two decades unless extraordinary actions are taken.4 So far, the science of the problem has proven elusive, with uncertain proposals involving price tags of about $1 billion or more. As a consequence, one of IID’s major concerns in the transfer talks has been to bind its liability limits with respect to the sea’s future health. The other major sticking point has been the fallowing question. Once fallowing became the suggested means of achieving the water savings for the transfer, the deal switched from one that was “win-win” for the district to one involving winners and losers. An efficiency-based transfer program would involve keeping all the land in production while creating local jobs to carry out conservation investments. A fallowing program implies some job losses. The debate over just how many and over what time horizon has been central to the negotiation process. Under the terms of a proposed deal negotiated in October 2002, the new per-acre-foot price to be paid by San Diego includes enough additional money to cover up to $20 million in mitigation funds over 15 years. The maximum annual acreage to be fallowed is 30,000 (of a total of 450,000 irrigable acres in the district). Imperial would be free to switch from fallowing to efficiency-based methods of water savings beginning in year 16. San Diego would agree to cover any excess costs of third-party effects beyond those provided for in the $20 million.5 ________________________________________________________ use by the agricultural contractors. This problem arose with the initial transfer of 110,000 acre-feet from IID to MWDSC begun in 1988 and noted in Chapter 2. The transfer was based on efficiency gains in IID’s network, financed by MWDSC. Although the gains were realized (and allocated to MWDSC), IID actually increased its water use in the subsequent period. 4See Imperial Irrigation District and U.S. Bureau of Reclamation (2002). 5For details of the proposal as per the October 15, 2002, agreement, which has remained the basis of the proposals on fallowing, see “Summary of Water Agreement,” 74 In sharp contrast to the Palo Verde case, no one in the Imperial Valley has gone on record to praise the deal, except to say that it was the best the negotiators could do in a difficult situation. In the months leading up to the October negotiations, both federal and state authorities had made it clear that the district risked having its water rights challenged in the absence of a successful transfer agreement.6 In December 2002, IID’s board voted 3-2 against the terms of the deal brokered in October. Although the effects of fallowing were cited as one of Imperial’s Board’s concerns, the “deal-breaker” in the end appeared to be the lack of adequate guarantees against possible lawsuits over environmental effects to the Salton Sea.7 The fate of the transfer remains uncertain at the time of this writing (May 2003), but the latest proposal brokered by the state involves an additional $200 million in state funds directed to mitigating effects to the Salton Sea. Fallowing in the San Joaquin Valley The planned Palo Verde transfer and the possible sale by Imperial are the state’s first large long-term water transfers based on land fallowing, but they are not the only places where this is taking place. Fallowing has been a regular feature of the temporary agricultural water market within the San Joaquin Valley since the early 1990s. The district-to-district transfers of this type mainly involve moving water from land owned or leased by the same farmer to more productive, water-short land elsewhere in the valley, notably within Westlands Water District and several neighboring CVP contracting districts.8 Fallowing, in these contexts, is accepted as part of the farmers’ overall land management plans. Water districts allowing this activity do vary, however, in the extent to which they impose conditions on the transfer. In some districts, once the land is leased there are no restrictions on transfers to other lands farmed by the ________________________________________________________ October 17, 2002. The maximum amount to be fallowed is not explicitly mentioned in the agreement but has been cited elsewhere by IID directors and others (Vogel, 2002). 6See statements by U.S. Interior Department Assistant Secretary Bennett Raley in Kasindorf (2002). For a discussion of state policy, see Kasler (2002). 7See Conaughton (2002). 8Drainage problems, which are reducing the productivity of some areas within Westlands, have also encouraged land fallowing for water sales within the district. 75 lessee. For farmers within the Kern County Water Agency service area, a casual lease is not sufficient for gaining the right to transfer the water. Land retirement was also the basis for the permanent transfer of up to 130,000 acre-feet of State Water Project entitlement from Kern County farmers to municipal users under the Monterey Agreement. Importantly, however, this proposal came about as a way of reallocating water from land that had already come out of production because of marginal economic conditions; it did not precipitate crop idling. Two recent sales of SWP entitlement by Kings County farmers, one to other farmers and one for municipal uses, were also predicated on removing water rights from lands that are becoming unprofitable to farm.9 Recent proposals by some water users in Kern County would involve a multiyear, rotational fallowing scheme to free up water for municipal users outside the county.10 Whether and under what conditions such programs are acceptable to the wider community is one of the subjects currently on the table in a countywide review of water transfer policy launched by the Kern County Water Agency. In the eastern part of the San Francisco Bay Area and in San Joaquin County, several long-term, local agriculture-to-urban transfers also derive some of the water savings from land fallowing or land retirement.11 Such ____________ 9Sales by Tulare Lake Basin Water Storage District to Dudley Ridge Water District (for agriculture) and Antelope Valley–East Kern Water Agency (AVEKWA) (for municipal uses). For details, see Table A.6. Technically, there is no explicit fallowing requirement under the terms of the transfer; rather, lands for which SWP entitlement is sold are permanently disallowed from receiving future SWP entitlements. The transfer to Dudley Ridge involved lands owned by the same farmer; the water will be used in Dudley Ridge where the land is more productive. The transfer to AVEKWA involved lands being taken out of production by a large ranch, which is scaling back operations. 10See Semitropic Water Storage District (2002). 11This includes the ongoing transfer from Byron-Bethany Irrigation District to Alameda County Flood Control and Water Conservation District, Zone 7 (for which water is made available through temporary fallowing of approximately 600 acres and water savings from 300 acres of land already retired) and the proposed transfer of contract entitlement from the Westside Irrigation District and Banta Carbona Irrigation District to the City of Tracy (made possible by preexisting land retirement). See Tables A.5 and A.6 and the environmental documentation for these transfers (Alameda Flood Control and Water Conservation District Zone-7, 1994, and CH2MHill, 2002a, 2002b). 76 transfers raise somewhat different issues from the cases where fallowing involves sending water to a distant destination. The selling districts defend the action on the grounds that residential development is in all events encroaching on the area’s farmland; the transfers provide a way of lessening costs for the remaining farmers while moving water locally to new uses. Objections, when raised, relate to concerns over the consequences of development within the area, not to an outflow of economic opportunity.12 Future Trends? Recent developments in the Sacramento Valley suggest that fallowing for transfers to points south of the Delta will be an increasingly important component of the water market. As noted, DWR purchased water conserved through fallowing from farmers in Butte County for its 2001 dry-year program. Fallowing was also the primary source of the water for the transfer from CVP settlement contractors in the Sacramento Valley to Westlands in that year.13 As the fate of the Imperial–San Diego transfer remained uncertain in late 2002, officials of Metropolitan Water District of Southern California—the agency that stands to lose the most water if the state fails to meet the deadline for the 4.4 Agreement—unveiled a plan to purchase 205,000 acre-feet from this same group of senior water-rights-holders in 2003. For this transfer, again primarily based on land fallowing, Metropolitan has proposed to include a $5 per acre into a mitigation fund (5 percent of the $100 per acre-foot price negotiated for the water). Once the details of this oneyear deal are worked out, the water districts intend to discuss possible longer-term arrangements combining fallowing and groundwater ____________ 12For instance, the Sierra Club has taken issue with the pending sales to Tracy on grounds that they encourage sprawl (Cooper, 2002). 13Participants in the program included Glenn-Colusa Irrigation District, PrincetonCodora-Glenn Irrigation District, Provident Irrigation District, Reclamation Districts 108 and 1004, several mutual water companies, and private farms with individual CVP contracts. Of a total of 160,000 acre-feet, 91,000 acre-feet were made available through fallowing, corresponding to approximately 27,500 acres. 77 substitution. The fallowing is concentrated on acreage planted to rice, for which prices are at historical lows.14 Thus, we see diverse local reactions to the prospect of fallowing and to mitigation. So far, the only substantial mitigation programs envisaged are for the two large long-term transfers of Colorado River water and for temporary purchases by two big buyers—the state itself and the largest urban water agency. There is, as yet, no real track record on either the consequences of a long-term fallowing program or the “how-to” of mitigation. Before turning to some of the practical questions of designing programs to limit negative effects, we examine briefly the economic and legal issues of fallowing for water transfers. Economic and Legal Issues of Land Fallowing for Water Sales Economic Incentives and Third-Party Effects From the seller’s standpoint, fallowing to free up water for the market is likely to be most attractive, the higher the sale price of water and the lower the value of the water in agricultural uses. This is why short-term land fallowing is often seen as flexible tool for coping with drought conditions. At such times, water will fetch a better market price, thus compensating farmers for forgoing their own use. It is also why longer-term fallowing programs for sale to municipal users may be appealing to some farmers, because the municipal market can generally afford to pay a higher price than can other agricultural users. The incentive structure for individual farmers will depend critically on the rules established by the local water district. In some situations, the bulk of the proceeds will go directly to the farmer who is idling land. In a surprising number of cases, however, the proceeds remain at the district level and go toward keeping down costs to other users. This was ____________ 14See Economic Research Service (2002). Changes in the U.S. farm support programs since the second half of the 1990s have also made it more attractive to fallow when farm prices are low. Under current programs, farmers of commodities eligible for price support (notably rice and cotton in California) are paid on the basis of historical acreage rather than current acreage. As a result, they are not penalized for idling the land to make the water available for the market. 78 the procedure for the Monterey Agreement transfers in Kern County, as well as the transfer in the Bay Area noted above. Without a change in operating rules, this redistribution would also occur in the many districts that do not attribute specific contract amounts to individual landowners but rather deliver water on the basis of annual requests.15 The attraction of “farming water” instead of farming land is clearly greater when individuals, and not districts, stand to benefit from the sale. From the standpoint of the surrounding community, the effects of a fallowing operation will depend on the interaction of two types of effects. On the one hand, there are the effects of the changes in land use on onfarm and farm-related employment, tax revenues, and, indirectly, the wider economic activity of the area. On the other hand, there are the proceeds of the water sale, which, if spent locally, contribute to job and revenue creation. Thus the aggregate effect of a water sale achieved through a reduction in cropland is not necessarily negative, even in the short run. If, as is typically assumed, farmers elect to fallow the lowvalue crops—which produce less profit per unit of water used and which require relatively lower labor inputs—the associated employment and revenue reductions will be limited. If, at the same time, they reinvest proceeds of the water sales into farming operations, for instance, by releveling the land or making upgrades in equipment, this reinvestment can have a positive effect on employment and revenues. Typically, however, there are some losers from land fallowing, even in a generally positive scenario such as the one described. The new investments in land leveling or other farm improvements will give a boost to those sectors, whereas the fall in crop output will reduce the demand for specialized services such as harvesting and processing and, in all likelihood, some farm labor. This is why the notion of mitigation enters the equation. Mitigation is envisaged both as a short-term compensation for income losses and as a means of assisting those who may be permanently affected by a long-term water transfer to adjust to the new economic circumstances. ____________ 15For instance, Yolo County Flood Control and Water Conservation District and Madera Irrigation District. 79 Nevertheless, mitigation for land fallowing poses some distinct economic, legal, and policy questions. The negative effect on those who lose out is an economic spillover effect, or what economists sometimes refer to as a “pecuniary externality”—the effect of one person’s business decision on someone else’s financial outcome. This stands in contrast to a “physical externality,” which occurs when a water transfer negatively affects the quantity or quality of the physical resource available to other users. From the standpoint of economic theory, a transfer that results in a negative physical externality requires compensation. Compensation helps to ensure a socially efficient outcome by preventing overuse of the resource. By contrast, there are no efficiency grounds for compensating those affected by a pecuniary externality. Rather, the issue is one of equitable distribution of the benefits (Howe et al., 1990, Howitt, 1994). Lack of Legal Provisions for Mitigation The legal issues raised by the two types of effects are also quite distinct. There is a legal tradition for protecting third parties from the negative physical externalities associated with business decisions. In California, the no-injury provisions of state water law, established in case law as early as 1862, specifically aim to prevent negative physical effects on other water users, including wildlife.16 However, there is no clear legal tradition for protecting individuals from the effects on their livelihood of a change in other people’s business decisions—and this is what compensation for the economic effects of land fallowing implies. At the federal level, exceptions arise when a region or sector is negatively affected by a policy change considered beneficial for society as a whole. Notably, national programs have been available since the early 1960s to assist workers in industries affected by trade liberalization, and special regional programs were introduced in the 1970s to assist forestry workers affected by the expansion of national park areas in the western states.17 More generally, federal, state, and local governments provide ____________ 16The no-injury rule was established in the court case Butte T. M. Co. v. Morgan, 19 Cal. 609 (Gray, 1994a). The environmental protections for fish, wildlife, and instream beneficial uses were codified in Cal. Water Code Section 1738 in 1980. 17See Appendix F in Illingworth et al. (2002). 80 transitional assistance to workers facing negative economic conditions through unemployment insurance and other social programs. In California, the only legal provision for protecting communities from the economic effects of water transfers is found in the wheeling statute of the Water Code (Sections 1810–12), introduced in 1986. Under this statute, the transport of water through public conveyance facilities must be done “without unreasonably affecting the overall economy of the environment of the county from which the water is being transferred.” To date, these protections have not been invoked. In effect, it is difficult to demonstrate that fallowing programs pass the (admittedly vague) unreasonable effect test. Available studies suggest that fallowing programs would need to be considerably more extensive than those in recent years to have significant negative county-level effects (Table 6.1). The studies, which examine the effects of fallowing anywhere from 6 to 25 percent of a county’s irrigated farmland, find that it is likely to have no more than a 1 percent effect on overall county economic activity, even when the payments to farmers for the water transfers are excluded. This level was exceeded in only two counties—Colusa and Glenn—in a study simulating the effects of a 25 percent cut in surface water supplies (with no revenues for water sold). A 22 percent reduction in land farmed in the Palo Verde area under the test program had insignificant effects on income within that limited region, a result that is also anticipated even if farmers engage in maximum fallowing (29 percent of total acreage) under the upcoming long-term program. However, the same studies also demonstrate that there can be significant localized negative effects on individual farm workers and businesses and on local public agencies such as school districts. Thus, there may be ethical grounds for devising mitigation programs, even when a transfer does not trigger the legal requirement to do so. The case for mitigation is stronger when the transfer has negative distributional implications—a concentration of losses to low-income farm workers and processing plant laborers and accrual of most benefits to the relatively wealthy members of the community (or, in the case of absentee landlords, nonmembers). Given the structure of California’s agricultural economy, where wages are low, unemployment rates high, and large 81 Study Area and Subject 1991 drought water bank (11 counties) Table 6.1 Study Estimates of the Overall Economic Effect of Land Fallowing Authors Dixon et al. (1993) Acres and % of County Farmland Fallowed County Job Losses Regional or Sectoral Income Losses 2–3% of agricultural income Overall County Income Losses <1% 1991 drought water Howitt (1994) bank (Yolo and Solano) Solano: 23,500 acres (13%) Yolo: 45,700 acres (13%) 4.7% (model) 1.5% (survey) (farm job loss, both counties) Solano: 3.2% gross agricultural income Yolo: 5% gross agricultural income <1%a 82 1992–1993 Palo Verde Test Fallowing Program M. Cubed (1994) 20,215 acres (22% of region’s farmland) 1.3% regional job Negligible regional Negligible loss income effects 2003–2038 Palo Verde Long-Term Program M. Cubed (2002) 26,500 acres (29% of region’s farmland (program maximum) <2% regional job <1% regional loss income Negligible Irrigation water cuts Lee et al. (1999) in Sacramento Valley (8 counties)b 25% surface supply cut, no replacement with groundwater (4.5% to 6.2% fall in acreage) 305 jobs (8 counties) Colusa: 5% Glenn: 2.5% 5 other counties: 0.5% Sacramento: negligible Proposed Westlands Illingworth et al. 100,000 acres (6%) <1% (Fresno and 5 M&I to M&I Bay Area So. Cal. SWP 1998 Westside WD Colusa County WD 25,000 25 Agr to Agr Sac Valley Sac Valley CVP 1999 Byron–Bethany ID Alameda County FCWCD 5,000 15 Agr to M&I Bay Area Bay Area No 1999 Oakdale ID Stockton East WD 15,000 10 Agr to M&I SJ Valley SJ Valley No 1999 South San Joaquin ID Stockton East WD 15,000 10 Agr to M&I SJ Valley SJ Valley No 2000 Mercy Springs WD Westlands WD, Santa Clara Valley WD, and Pajaro Valley WMAa 6,260 25 Agr to All SJ Valley SJ Valley and CVP Bay Area 2000 Placer County WA Northridge WD 29,000 25 All to M&I Sac Valley Sac Valley No 2001 Kern County WA Western Hills WD 8,000 35 Agr to M&I SJ Valley SJ Valley No 2001 Merced ID U.S. Fish and Wildlife 12,500 12 Agr to Env SJ Valley SJ Valley No 2001 Oakdale ID U.S. Fish and Wildlife 15,000 12 Agr to Env SJ Valley SJ Valley No 2001 SJ River Group Authority U.S. Fish and Wildlife 110,000 12 Agr to Env SJ Valley SJ Valley No 2002 San Bernardino Valley MWD MWDSC 20,000 2002b Imperial ID Coachella Valley WD and MWDSC 100,000 2002b Imperial ID San Diego County WA 200,000 2002b Butte WD Madera ID and Root Creek WD 15,000 2002b Merced ID U.S. Fish and Wildlife 47,000 2002b Palo Verde ID MWDSC 111,000 2002b South San Joaquin ID Cities of Tracy, Escalon, Manteca, Lathrop 75,000 10 M&I to M&I So. Cal. So. Cal. 75 Agr to All So. Cal. So. Cal. 75 Agr to M&I So. Cal. So. Cal. 25 Agr to M&I Sac Valley SJ Valley 10 Agr to Env SJ Valley SJ Valley 35 Agr to M&I So. Cal. So. Cal. 25 Agr to M&I SJ Valley SJ Valley SWP CO River CO River No No CO River No NOTES: Abbreviations: WA = water agency or water authority; WMA = water management agency; FCWDC = flood control and water conservation district; MWD= municipal water district; M&I = municipal and industrial; Agr = agriculture; Env = environment. “U.S. Fish and Wildlife” as buyer indicates purchase by USBR’s Water Acquisition Program for environmental uses; SJ River Group Authority includes Merced ID, Modesto ID, South San Joaquin ID, Oakdale ID, SJ River Exchange Contractors, and Friant Water Users Association. aWestlands WD and Santa Clara Valley WD will receive deliveries until Pajaro Valley WMA completes construction of facilities for the permanent transfer of Mercy Springs WD entitlement. bPending approval in 2002. Table A.6 Permanent Transfers of Surface Water Since 1985 137 Year Seller Buyer Annual Acre- Region of Region of Within Feet (Max) Purpose Origin Destination Project 1998 1998 1998 1998 2000 2000 2000 2000 2001 2001 2001 2001 2002 2002 2002a 2002a 2002a 2002a Corning WD Kern County WA Proberta WD Thomes Creek WD Kern County WA Kern County WA Kern County WA Kern County WA Kern County WA Kern County WA Kern County WA Olcese WD and private ranch Tulare Lake Basin WSD Tulare Lake Basin WSD Banta Carbona ID Lower Tule River ID Mercy Springs WD The Westside ID U.S. Fish and Wildlife Mojave WA U.S. Fish and Wildlife U.S. Fish and Wildlife Alameda County FCWCD Alameda County FCWCD Castaic Lake WA Palmdale WD Alameda County FCWCD Napa County FCWCD Solano County WA Kern County WA Antelope Valley–East Kern WA Dudley Ridge WD City of Tracy City of Orange Cove Westlands WD City of Tracy 2,300 25,000 2,000 2,000 15,000 7,000 41,000 4,000 10,000 4,025 5,756 50,000b 3,000 3,973 5,000 2,000 1,071 5,000 Agr to Env Agr to M&I Agr to Env Agr to Env Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to Agr Agr to M&I Agr to M&I Agr to Agr Agr to M&I Sac Valley SJ Valley Sac Valley Sac Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley Sac Valley So. Cal. Sac Valley Sac Valley Bay Area Bay Area So. Cal. So. Cal. Bay Area Bay Area Bay Area SJ Valley So. Cal. SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley CVP SWP CVP CVP SWP SWP SWP SWP SWP SWP SWP No SWP SWP CVP CVP CVP CVP NOTES: Abbreviations: WA = water agency or water authority; WMA = water management agency; FCWDC = flood control and water conservation district; MWD= municipal water district; M&I = municipal and industrial; Agr = agriculture; Env = environment. “U.S. Fish and Wildlife” as buyer indicates purchase by USBR’s Water Acquisition Program for environmental uses; SJ River Group Authority includes Merced ID, Modesto ID, South San Joaquin ID, Oakdale ID, SJ River Exchange Contractors, and Friant Water Users Association aPending approval in 2002. bAverage annual volumes. Appendix B Groundwater Institutions and Basins Table B.1 Counties with Groundwater Protection Ordinances, by Region Export Restrictions Mountain Region Calaveras (2002) Inyo (1980) Lassen (1999) Modoc (1978) Mono (1988; 1998) Nevada (1986–1988 only) Sierra (1977; 1997) Siskiyou (1998) Tuolumne (2001) Sacramento Valley Butte (1977; 1996) Colusa (1998) Glenn (1977; 1990) Sacramento (1980) Shasta (1997) Tehama (1992) Yolo (1996) San Joaquin Valley Fresno (2000) Kern (1998) Madera (1999) San Joaquin (1996) North Coast On-Site Groundwater Use Restrictions No Ordinance Alpine Amador El Dorado Mariposa Plumas Placer Sutter Yuba Mendocino (1995) Kings Merced Stanislaus Tulare Del Norte Humboldt 139 Table B.1 (continued) Export Restrictions San Francisco Bay and Central Coast Regions Lake (1999) San Benito (1995) On-Site Groundwater Use Restrictions No Ordinance Monterey (1993) Napa (1999) Alameda Contra Costa Marin Santa Barbara Santa Clara Santa Cruz San Francisco San Luis Obispo San Mateo Solano Sonoma Southern California Imperial (1996) San Diego (1991) San Bernardino (2002) Los Angeles Orange Riverside Ventura NOTE: Year in parentheses is year of adoption. If two dates are listed, the first refers to the adoption of an urgency ordinance and the second to the adoption of a regular ordinance. Many ordinances have been revised at least once subsequently. Imperial County adopted an ordinance requiring conditional use permits for some within-county groundwater uses in 1972 and added explicit export restrictions in 1996. 140 Del Norte Siskiyou Modoc Trinity Humboldt Shasta Lassen Tehama Plumas Yuba Mendocino Glenn Butte Sierra Sutter Colusa Lake Placer Nevada Sonoma Napa Yolo El Dorado Amador Alpine Sacramento Marin Solano Calaveras San Francisco Contra San Costa Joaquin Tuolumne Mono San Mateo Alameda Santa Stanislaus Mariposa Clara Merced Santa Cruz Madera San Benito Monterey Fresno Kings Tulare Inyo San Luis Obispo Kern Santa Barbara Ventura Los Angeles San Bernardino Orange Riverside San Diego Imperial Figure B.1—California’s Counties 141 1. Raymond Basin (1944) 2. Cucamonga Basin (1958, management 14 structure being updated) 3. West Coast Basin (1961) 4. Central Basin (1965) 5. Santa Margarita River Watershed (1966) 6. San Bernardino Basin Area (1969) 7. Brite Basin (1970) 8. Cummings Basin (1972) 9. Tehachapi Basin (1973) 10. Main San Gabriel Basin (1973) and Puente Narrows (1972) 11. Warren Valley Basin (1977) 12. Chino Basin (1978) 13. Upper Los Angeles River Area (1979) 14. Scott River Stream System (1980) 15. Puente Basin (1985) 16. Mojave Basin Area Adjudication (1996) 17. Santa Paula Basin (1996) 18. Six Basins (1998) 7, 8, 9 16 1 10,15 17 2, 6 13 3 4 12,18 5 11 SOURCES: Adjudication dates—Department of Water Resources (2001). Basin contours—Department of Water Resources (2002a). NOTES: The basins illustrated are the full basin boundaries as determined by DWR. Actual adjudicated basin boundaries are defined by the court and are often smaller. The dates indicate the year of final adjudication. Figure B.2—Adjudicated Groundwater Basins 142 Special Groundwater Management Districts 11 1. Monterey Peninsula Water Management District (1977) 2. Long Valley Groundwater Management District (1980) 3. Sierra Valley Groundwater Management District (1980) 10 4. Fox Canyon Groundwater Management Agency (1982) 7 5. Pajaro Valley Water Management Agency (1984) 6. Mendocino City Community Services District (1987) 7. Honey Lake Groundwater Management Agency (1989) 8. Tri-Valley Groundwater Management District (1989) 9. Ojai Groundwater Management Agency (1991) 10. Willow Creek Groundwater Management Agency 6 2 (1993; inactive) 3 11. Surprise Valley Groundwater Management Agency (1995; inactive) Special Water Districts with Groundwater Control A. Santa Clara Valley Water District B. Orange County Water District 8 C. Coachella Valley Water District A 5 1 9 4 B C SOURCES: Agency names and dates are from the Department of Water Resources (1996a); basin contours are from the Department of Water Resources (2002a); agency contours are from the U.S. Bureau of Reclamation (2002). NOTES: Special Water Districts and Pajaro Valley Water Management Agency are represented by administrative district boundaries (U.S. Bureau of Reclamation, 2002). Other district boundaries are represented by underlying groundwater basin boundaries (Department of Water Resources, 2002a). The following district boundaries were estimated using corresponding basins: Monterey Peninsula Water Management District (Seaside Area subbasin of Salinas Valley and Carmel Valley), Fox Canyon Groundwater Management Agency (Oxnard subbasin of Santa Clara River Valley, Pleasant Valley, Arroyo Santa Rosa Valley, and Las Posas Valley), Mendocino City Community Services District (Fort Bragg Terrace Area), and Tri-Valley Groundwater Management District (Eastern Mono County portion of Owens Valley). Figure B.3—Special Groundwater Management Districts 143 A B C Basins Subject to Critical Conditions of Overdraft 1. Santa Cruz Pajaro Basin 2. Cuyama Valley Basin 3. Ventura Central Basin 4. Eastern San Joaquin County Basin 5. Chowchilla Basin 6. Madera Basin 7. Kings Basin 8. Kaweah Basin 9. Tulare Lake Basin 10. Tule Basin 11. Kern County Basin Basins with Special Problems A. Surprise Valley Basin 4 B. Long Valley Basin C. Sierra Valley Basin D. Owens Valley Basin 5 6 17 8 9 10 D 11 2 3 NOTES: The basins illustrated are those identified in Bulletin 118-80 (Department of Water Resources,1980), mapped using boundaries appearing in the 2002 Draft Groundwater Map (Department of Water Resources, 2002a). Owing to name or boundary changes, the following 1980 basin boundaries were estimated using the 2002 definitions: Santa Cruz-Pajaro Basin (Santa Cruz Purisima Formation; Pajaro), Ventura Central Basin (Piru, Fillmore, Santa Paula, Mound, and Oxnard subbasins of Santa Clara River Valley; Pleasant Valley; Arroyo Santa Rosa Valley; and Las Posas Valley). Figure B.4—Critically Overdrafted and Special Problem Groundwater Basins Listed in Bulletin 118-80 144 Appendix C Predicting County Adoption of Export Restrictions This appendix describes the data sources, estimation methods, and results of the statistical analysis of the likelihood of county adoption of export restrictions reported in Chapter 4. Data Sources Farm and Agriculture-Related Jobs The source for farm jobs is the Bureau of Economic Analysis regional accounts data. The source for agriculture-related jobs is the Bureau of the Census “County Business Patterns.” Data are from 1995. Share of Irrigated Agriculture in Total Farmland This series is derived from results of the 1997 Agricultural Census (U.S. Department of Agriculture and the California Department of Food and Agriculture). 1997 data were preferred over those from 1992, a year with unusually low acreage figures as a consequence of the prolonged drought. Share of Residential Population Dependent on Groundwater This series was approximated using information from county environmental health officers and a 1992 survey by the Water Education Foundation (1994). Other Variables Membership in the Regional Council of Rural Counties and presence of a critical or specially designated groundwater basin are presented in Chapter 4. 145 Estimation Methods and Results Probit regressions were used to estimate the effect on a county’s likelihood of adopting an export ordinance of the variables presented above, for which sample characteristics are reported in Table 4.1. The statistical exercise involves considering the effect of each variable on ordinance adoption, while holding the other variables constant. Table C.1 reports the results of two regressions, one including the full set of variables and one excluding residential groundwater share. The variables used in each model are jointly significant at the 99 percent level of confidence. The second model produces slightly tighter coefficient estimates because groundwater share, itself insignificant, is correlated with the measure of critical basins. This exclusion does not affect overall model fit. Both models slightly underpredict adoption of export restrictions: compared to the 22 actual cases (38 percent), the models predict adoption by 20 counties (34 percent). Most of the variables are marginally significant (at the 90 percent level of confidence), with coefficients of the expected sign. The effects of each variable on the likelihood of ordinance adoption are presented in Chapter 4. Because some of the counties in the nonadoption group have other types of more comprehensive groundwater management systems in place—including adjudicated basins, special districts, or groundwater protection ordinances that control local groundwater use—we also performed two tests to see whether the results changed significantly when controlling for this factor: a multinomial logit regression distinguishing among three possible outcomes (export restrictions, comprehensive groundwater management systems, and no groundwater rules), and a binomial probit with a control variable for counties with these types of groundwater systems. The additional “groundwater management” group for the multinomial logit included El Dorado, Los Angeles, Mendocino, Monterey, Napa, Orange, Plumas, Riverside, San Bernardino, Santa Clara, Santa Cruz, San Diego, and Ventura. The groundwater management control variable in the binomial probit included this group plus counties with dual systems: Imperial, Lassen, Modoc, Mono, San Benito, Sierra, and Siskiyou. In neither case was there a substantial 146 Table C.1 Effects of County Characteristics on the Probability of Adopting an Export Restriction Farm employment (%) Agriculture-related employment (%) Irrigated farmland (%) Residents using groundwater (%) Counties overlying critical/special basin RCRC membership Chi-squared test of joint significance of variables in model Log-likelihood Observed probability Predicted probability Model Including Residential Groundwater 0.045* (0.026) –0.13* (0.077) 0.004 (0.003) 0.001 (0.003) 0.23 (0.17) 0.27* (0.16) Model Excluding Residential Groundwater 0.048* (0.025) –0.12* (0.075) 0.004 (0.003) — 0.26* (0.16) 0.30* (0.15) 20.88*** –28.06 38% 34% 20.72*** –28.13 38% 34% NOTES: Coefficients are reported as marginal effects. For the binary variables (critical groundwater basins and RCRC membership), the coefficient is the effect of a discrete change from 0 to 1. Standard errors are reported in parentheses. ***Indicates coefficient significantly different from zero at the 99 percent level of confidence. *Indicates coefficient significantly different from zero at the 90 percent level of confidence. change in coefficient estimates in relation to those reported in Table C.1, although the loss of degrees of freedom in the multinomial logit reduces levels of significance. 147 Appendix D Measuring the Effect of Export Restrictions on County Water Sales This appendix provides detailed information on the data sources, estimation methods, and results of the statistical analysis of the effect of export restrictions on county water sales and exports presented in Chapter 5. Data Sources Annual County Water Sales and Annual County Exports These series are developed from the water transfer database presented in Chapter 2 and Appendix A. Annual county sales are defined as the sum of all short- and long-term transfers by county water users in a given year. For sales by cross-jurisdictional water districts, the approximate share of the district in each county has been attributed to that county. Annual county exports are the sum of transfers not destined for other water users within the county. Environmental water sales were considered as exports from the county. Although this water is most often used for habitat or instream purposes within the region, it rarely is under control of users in the county of origin. For water districts with multiple jurisdictions, we considered the transaction to be “in-county” if the purchaser was in any of the district’s counties. As such, the exports category unambiguously includes only those transfers going from a user within the county to a user somewhere else in the region or state. Because the coverage of transfers within the CVP’s Friant group was not consistently available in all years, we have excluded internal Friant transactions from the sales data. This concerns five San Joaquin Valley counties: Merced, Madera, Fresno, Tulare, and Kern. Since members of the Friant group are effectively exempted in the counties with ordinances, this should not pose a problem for interpretation of the 149 results. Trades between Friant members and other water users are included. County Export Restrictions We consider that counties have an export ordinance in operation beginning in the year of adoption, as indicated in Table C.1. There are two exceptions. Because Kern County’s ordinance applies only to the relatively unimportant desert and foothill region in the southeast, but not to the San Joaquin valley portion of the county where population, agriculture, and surface water entitlements are all concentrated, we have considered Kern to have no ordinance for the purposes of this exercise. The second exception is made for Glenn County, whose export ordinance was effectively removed in 2000, when the new basin management ordinance was adopted.1 The 2001 season is the first during which Glenn water users worked under the new system. State and Federal Policy Environment The general effects of an improved trading environment arising from state and federal policies to facilitate transfers are captured by a time trend. Agricultural Water Demand The model uses three measures of agricultural water demand: average county-level prices for annual crops (defined as all field and horticultural crops), the acreage under annual crops, and the share of perennial crops in total nonrange acreage. All three series are constructed using county agricultural statistics from the California Agricultural Statistics Service databases.2 ____________ 1Although the 1990 ordinance remains on the books, the numerous persons interviewed in Glenn, including two county supervisors, considered that the new ordinance has supplanted it for operational purposes. 2The annual crop price is calculated using the county’s prior year output data, valued at the statewide average price for the current year. This captures the notion that the farmer has an idea of the average market price for the coming season and can calculate what he would earn by farming the same crop mix as in the preceding year. The series is deflated using the western states urban consumer price index, with 1992 as the base year. Ideally, we would measure the value of crops on a per-acre-foot basis to capture the water 150 In principle, we would expect water sales to be inversely related to the average level of crop prices, which reflect the value of using water in agriculture. On average, real price levels have been relatively flat over the period, hovering around $160 per ton in 1992 prices. The range across counties is quite large, however, with averages over $300 per ton along the south and central coast and $100 per ton or less in parts of the Central Valley. As with prices, there has been little movement over time in the average level of annual crop acreages, although the cross-county differences are huge, with at least several hundred thousand acres in Imperial County and most San Joaquin Valley counties (and close to 1 million acres in Fresno), and fewer than 50,000 acres along the coast.3 In part, this range reflects differences in the overall scale of agricultural operations across counties; in part, it reflects a much higher share of perennial crops (fruit trees, nut trees, and vineyards) in some counties. The values span a high of over 90 percent of all nonrange farmland in Napa and over 50 percent in San Diego, Ventura, and Madera, to only 1 percent of cropland in Imperial. Over time, there has been a mild upward trend in tree crops as a share of the total, moving from 22 to 26 percent on average. Because farmers can make adjustments in annual crop acreages fairly easily as a function of water availability, we would expect water sales to be positively related a county’s crop acreage. Conversely, because a higher share of tree crops in total acreage introduces less flexibility in water use, we would expect tree crop share to be negatively related to water sales. ________________________________________________________ intensity of the crop mix. This would require making assumptions about the irrigation technology used in each county for each crop, however. If anything, the use of a per-ton measure probably dampens the effect of this variable, since low-value crops also tend to be those with a relatively high level of water use. Annual crop acreage includes all farm acreage except perennials and rangeland (i.e., including irrigated pasture). The share of tree crops is calculated as the share of perennials in total nonrangeland farm acreage. Both acreage measures are valued at the prior year levels to account for the fact that decisions on water sales are generally made before final planting decisions. 3San Francisco, the only county in the sample with no commercial agriculture (albeit some fine gardens), has no acreage recorded and no positive crop prices. 151 Residential Water Demand County population levels are used to account for residential water demands. The source is the annual population series from the California Department of Finance, based on updates from the 2000 Census (Department of Finance, 2001). Other things equal, we should expect counties with higher populations to be less likely to sell water. Water Supply Conditions Annual deliveries of project water from the CVP, the SWP, and the Colorado River Project are captured in two measures: senior rights and junior rights.4 The senior rights category includes those deliveries with a high degree of reliability, by virtue of the seniority of the contractors. This includes the CVP settlement contractors in the Sacramento Valley and the exchange contractors in the San Joaquin Valley, the SWP’s Feather River contractors in Butte and Sutter Counties, and the Colorado River contractors in Southern California. On average, 8.3 million acre-feet are delivered annually to these contractors. Half of this volume is destined to Imperial and Riverside Counties and a quarter to the four main settlement-contracting counties in the Sacramento Valley (Colusa, Butte, Sutter, and Glenn). The only dips in supply occurred during the early 1990s drought, when CVP and SWP contractors’ deliveries were reduced by 25 to 50 percent in some years. The junior-rights category includes the ordinary project contractors of the CVP and the SWP. On average, these projects have delivered just over 6 million acre-feet annually over the 12-year period, to a much larger number of water users. Only two counties in the sample, San Francisco and Yuba, do not have project contractors. Project deliveries have generally been much more variable from one year to the next, particularly for contractors south of the Delta. ____________ 4For the Colorado River Project and the SWP, actual delivery data were used. For the CVP, we applied the annual allocation rules by type of project contractor (settlement contractors, north of Delta and south of Delta service contractors, and Friant class 1 and 2 contractors). Because California received surplus deliveries of Colorado River water for the entire time period under consideration, the Metropolitan Water District of Southern California is included as a senior rights-holder. Under California’s official allocation of 4.4 million acre-feet, this agency would have its supplies cut back, as the junior rightsholder. 152 In general, we would expect counties with higher water deliveries to be more active in the water market. By the same token, individual counties should be more likely to sell in years when their deliveries are higher. Unfortunately, detailed data on other water supplies—from autonomous projects and from groundwater—are not available. We do have a general indicator of the quality of the water year, however, in the form of the state’s most important rainfall measure—the Sacramento Valley 40-30-30 index.5 Since market demands and water prices are likely to be higher in dry years, we would expect this indicator to be negatively related to sales. The period under review contains an equal number of dry and wet years (Figure 2.1). Sample Characteristics Table D.1 provides summary statistics for the complete set of variables used for both geographical samples used in the estimations: the 34 water-trading counties6 and the 18 Central Valley counties.7 We have excluded 24 nontrading counties from the first group for statistical reasons. Two key econometric models cannot be estimated when these counties are included. The inclusion of counties that never trade adds no information to the estimation of the effects of an export ordinance on trading behavior in a fixed-effects mode. The presence of many counties with all zero trades also complicates the estimation of a random-effects Tobit model. For the state as a whole, the convergence properties of this ____________ 5The Sacramento Valley 40-30-30 Index is the main index used by DWR to measure water conditions in the Sacramento Valley, source region for both the CVP and the SWP as well as several large local projects. It is computed as a weighted average of the current water year’s April–July unimpaired runoff forecast (40 percent), the current water year’s October–March unimpaired runoff forecast (30 percent), and the previous water year’s index (30 percent). For details, see DWR’s website, http://watsup2.water.ca.gov/ hydrologic.cfm. 6The sample excludes the 24 counties for which there are no transactions records in any of the 12 years: Alpine, Amador, Calaveras, Del Norte, El Dorado, Humboldt, Inyo, Lake, Lassen, Marin, Mariposa, Mendocino, Modoc, Mono, Monterey, Nevada, Plumas, San Mateo, Santa Cruz, Sierra, Siskiyou, Sonoma, Trinity, and Tuolumne. 7Butte, Colusa, Fresno, Glenn, Kern, Kings, Madera, Merced, Placer, Sacramento, San Joaquin, Shasta, Stanislaus, Sutter, Tehama, Tulare, Yolo, and Yuba. 153 Table D.1 Summary Statistics for Annual County Water Sales and Water Exports, 1990–2001 All sales (acre-feet) Observed probability of sales Out-of-county exports (acre-feet) Observed probability of exports 34 Trading 18 Central Counties Valley Counties 22,734 31,461 (40,658) (45,842) 0.69 0.89 19,465 (39,075) 0.61 25,419 (44,140) 0.82 Agricultural and residential demand Annual crop prices ($/ton) 162 (126) Annual crop area (acres) 205,893 (226,721) Tree crop area in total (%) 23.2 (19.7) Population 856,177 (1,602,147) 117 (51) 294,918 (230,010) 22.6 (13.9) 286,226 (292,132) Water supply conditions Project deliveries (acre-feet) Senior rights Junior rights Rainfall index 243,826 (537,918) 179,474 (272,842) 8.18 (3.18) 178,751 (210,645) 247,412 (346,500) 8.18 (3.18) County export restrictions (1 = restriction) 0.19 (0.39) 0.30 (0.46) Number of observations 408 216 NOTE: The table reports mean values with standard deviations in parentheses. model are not stable. However, regression results on the full state sample are consistent with the findings we report below on the effects of the export ordinances and other key variables for the 34-county sample. 154 Regression Results A random-effects Tobit model is the main specification used to account for bunching at zero in the distribution of the dependent variable.8 Because it is not straightforward to test for fixed effects using this specification (Arellano and Honoré, 2001), we did so with a linear model. Tables D.2 through D.5 present the results for the randomeffects Tobit model and the corresponding random- and fixed-effects linear models for all sales and exports for the water-trading counties and Central Valley counties, respectively. Despite the censoring in the dependent variable, the results of the random effects linear model correspond closely to the Tobit model results, suggesting that reliance on the linear model for inferring properties about the Tobit model is reasonable. For all four models, Hausman specification tests of the linear model fail to reject the null hypothesis of no fixed effects at conventional levels of significance, suggesting the absence of crosssectional omitted variables that would bias the random-effects Tobit results. Consequently, the discussion in Chapter 5 focuses on the results of the random-effects Tobit model. Overall, the identified variables have the anticipated effects on both sales and exports. Among the control variables, the group capturing the effects of water supply is highly significant. Crop prices and annual crop acreage, two of the agricultural demand variables, are significant for the determination of sales within the full 34-county sample. These factors are not significant for the Central Valley sample, where there is less crosscounty variability. Results pertaining to the key variables of interest—county export restrictions and the time trend used to capture the effect of an improved trading environment—are presented in Chapter 5. The cumulative market effect of ordinances (Figure 5.2) was calculated by multiplying the number of counties affected by the per-county coefficients listed in ____________ 8For the 34-county sample of water trading counties, 31 percent of all sales and 39 percent of all exports are zero; for the 18-county Central Valley sample, the corresponding values are 11 percent of all sales and 18 percent of all exports. 155 the left-hand column of Tables D.2 through D.5. For the 34-county sample, the number of counties with ordinances ranges from two in the early 1990s to 12 in 2000. For the 18-county sample, the range is from two in 1990 to 10 in 2000. The cumulative market effect of state and federal policies (Figure 5.3) was calculated by multiplying the total number of counties in the sample (34) by the coefficient on the time trend reported in the left-hand column of Table D.2. 156 Table D.2 Determinants of Annual County Water Sales in 34 Water-Trading Counties, 1990–2001 Random-Effects Fixed-Effects Random-Effects Linear Linear Tobit Regression Regression Agricultural and residential demand Annual crop prices ($/ton) –75.8** –40.8 –55.6 (33.0) (25.9) (52.5) Annual crop area (1,000 acres) 54.6** 37.0* 143 (23.1) (19.7) (88.4) Tree crop area in total (%) –96.1 –10.2 213 (220) (176) (680) Population (1,000) –1.5 –1.1 –0.8 (2.6) (2.3) (32.4) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index 23.8*** (7.4) 49.2*** (13.1) –3,529*** (749) 21.5*** (7.2) 42*** (10.6) –2,160*** (549) –140** (56.6) 51*** (12.5) –2,099*** (575) State and local institutional factors County export restrictions State and federal policy (time trend) –14,308** (7,246) 3,828*** (681) –12,671** (5,563) 2,442*** (515) –10,770* (6,387) 2,616*** (720) Log-likelihood; overall R2 Hausman specification test –3,419 0.34 0.00 0.33 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 157 Table D.3 Determinants of Annual County Water Exports in 34 Water-Trading Counties, 1990–2001 Random-Effects Fixed-Effects Random-Effects Linear Linear Tobit Regression Regression Agricultural and residential demand Annual crop prices ($/ton) –88.9** –45.0* –59.2 (38.4) (27.0) (52.4) Annual crop area (1,000 acres) 37.6 15.7 97.1 (27.2) (20.6) (88.3) Tree crop area in total (%) –113 –17.9 215 (254) (185) (679) Population (1,000) –2.0 –1.2 8.0 (3.0) (2.5) (32.4) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index 28.3*** (8.0) 43.2*** (14.3) –3,235*** (831) 24.4*** (7.6) 34.5*** (11.0) –1,909*** (551) –160*** (56.6) 44.7*** (12.5) –1,796*** (575) State and local institutional factors County export restrictions State and federal policy (time trend) –16,948** (7,722) 3,729*** (761) –16,276*** (5,632) 2,242*** (517) –13,875** (6,378) 2,321*** (719) Log-likelihood; overall R2 Hausman specification test –3,062 0.26 0.05 0.15 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 158 Table D.4 Determinants of Annual County Water Sales in 18 Central Valley Counties, 1990–2001 Agricultural and residential demand Annual crop prices ($/ton) Annual crop area (1,000 acres) Tree crop area in total (%) Population (1,000) Random-Effects Fixed-Effects Random- Linear Linear Effects Tobit Regression Regression –71.8 (87.7) 11.0 (28.8) 289 (364) –18.5 (21.6) –57.9 (91.1) 1.0 (31.4) 118 (421) –7.4 (24.2) –87.0 (129.3) 224* (130) –278 (1,659) 163 (166) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index –7.2 (25.6) 64.0*** (15.4) –5,410*** (1,049) –16.2 (28.7) 60.0*** (15.0) –4,482*** (969) –116 (75.8) 65.4*** (17.3) –4,639*** (1,037) State and local institutional factors County export restrictions State and federal policy (time trend) –20,789** (8,713) 4,645*** (977) –19,038** (8,397) 4,094*** (934) –13,034 (9,309) 3,412** (1,341) Log-likelihood; overall R2 Hausman specification test –2,321 0.27 0.11 0.58 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 159 Table D.5 Determinants of Annual County Water Exports in 18 Central Valley Counties, 1990–2001 Random-Effects Fixed-Effects Random-Effects Linear Linear Tobit Regression Regression Agricultural and residential demand Annual crop prices ($/ton) –139 –77.7 –105.6 (102) (92.0) (130) Annual crop area (1,000 acres) –30.7 –26.6 166 (32.1) (31.7) (130) Tree crop area in total (%) 409 172 –559 (408) (427) (1,665) Population (1,000) –10.3 –3.5 245 (23.7) (24) (167) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index –10.9 (29.0) 58.6*** (16.4) –4,661*** (1,109) –26.4 (29) 51.5*** (15.1) –3,963*** (975) –126* (76) 54.7*** (17.4) –4,072*** (1,041) State and local institutional factors County export restrictions State and federal policy (time trend) –26,245*** (9,350) 4,220*** (1,039) –23.481*** (8,450) 3,659*** (940) –16,902* (9,343) 2,699** (1,346) Log-likelihood; overall R2 Hausman specification test –2,173 0.20 0.04 0.52 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 160 References Alameda Flood Control and Water Conservation District Zone-7, Negative Declaration for the Five Year Water Transfer Agreement with Byron-Bethany Irrigation District, Pleasanton, California, July 1994. Arellano, Manuel, and Bo Honoré, “Panel Data Models: Some Recent Developments,” in J. J. Heckman and E. Leamer (eds.), Handbook of Econometrics, Volume 5, Elsevier Science B.V., Amsterdam, 2001. Association of California Water Agencies, Water Supply and Development: A User’s Guide to California Statutes Including SB 221 (Kuehl) and SB 610 (Costa), prepared by McCormick, Kidman and Behrens, Sacramento, California, 2002. Brown, Judy, Toccoy Dudley, and Allan Fulton, Locally Governed Groundwater Management in Glenn County, California, Water Advisory Committee, Glenn County, California, October 2001. Bunn, Thomas S., III, “Counties Have a Stake in Managing Groundwater,” California County, July/August 1997, pp. 10–13. Campbell, Christopher L., “County Groundwater Regulation: A Case Study of Fresno and Madera Counties,” ACWA Continuing Legal Education Workshop for Water Attorneys, Costa Mesa, California, October 2000. CH2MHill, Final Initial Study/Environmental Assessment for the BantaCarbona Irrigation District/City of Tracy Water Assignment Project, prepared for the Banta-Carbona Irrigation District, City of Tracy, and U.S. Bureau of Reclamation, September 2002a. CH2MHill, Final Initial Study/Environmental Assessment for the West Side Irrigation District/City of Tracy Water Assignment Project, prepared for The West Side Irrigation District, City of Tracy, and U.S. Bureau of Reclamation, September 2002b. 161 Coase, R. H., “The Problem of Social Cost,” Journal of Law and Economics, Vol. 3, October 1960, pp. 1–44. Conaughton, Gig, “Water Transfer Still Breathing,” North County Times, December 13, 2002. Cooper, Audrey, “Tracy May Buy Farmers’ Water: Irrigation Districts OK Proposal to Sell to City,” San Joaquin Record, September 12, 2002. Department of Finance, Interim County Population Projections, Sacramento, California, June 2001. Department of Water Resources, Ground Water Basins in California (Bulletin 118-80), Sacramento, California, January 1980. Department of Water Resources, Water Facts: Groundwater Management Districts or Agencies in California, Sacramento, California, 1996a. Department of Water Resources, State Water Project Supplemental Purchase Program: Draft Environmental Impact Report, Sacramento, California, 1996b. Department of Water Resources, California Water Plan Update (Bulletin 160-98), Sacramento, California, November 1998. Department of Water Resources, Water Facts: Adjudicated Basins in California, No. 3, Sacramento, California, January 2001. Department of Water Resources, Bulletin 118 Statewide Groundwater Basin Map, Version 2, Sacramento, California, May 3, 2002a. Department of Water Resources, Draft Guidebook for Implementation of Senate Bill 610 and Senate Bill 221 of 2001, Sacramento, California, September 2002b. Department of Water Resources, “DWR Announces 2003 Dry Year Water Purchase Program,” news release, Sacramento, California, November 15, 2002c. Dixon, L. S., N. Y. Moore, and S. W. Schechter, California’s 1991 Drought Water Bank: Economic Impacts in the Selling Regions, RAND, MR-301-CDWR/RC, Santa Monica, California, 1993. 162 Dudley, Toccoy, “Basin Management Objective (BMO) Method of Groundwater Basin Management,” September 2000, reprinted in Glenn County, Basin Management Objective (BMO) for Groundwater Surface Elevations in Glenn County, California, Willows, California, August 2001. Economic Research Service, U.S. Department of Agriculture, Rice Outlook, Washington, D.C., December 11, 2002. Fulton, Allan, and Toccoy Dudley, Frequently Asked Questions about Glenn County Groundwater Ordinance No. 1115: Its Background and Intent, Organizational Structure, and Status, Water Advisory Committee, Glenn County, Willows, California, n.d. Getches, David, Water Law in a Nutshell, 3rd edition, West Publishing Co., Eagan, Minnesota, 1997. Glenn County, Basin Management Objective (BMO) for Groundwater Surface Elevations in Glenn County, California, Glenn County Board of Supervisors, Willows, California, August 21, 2001. Goldsmith, Janet K., “Counties, Cities May Regulate Groundwater, Appellate Court Rules,” California Water Law and Policy, February 1995a, pp. 42–43. Goldsmith, Janet K., “It’s Official—Counties Can Regulate Groundwater,” California Water Law and Policy, May 1995b, p. 163. Governor’s Commission to Review California’s Water Rights Law, Final Report, Sacramento, California, 1978. Gray, Brian, “The Modern Era in California Water Law,” Hastings Law Journal, Vol. 45, No. 249, January 1994a. Gray, Brian, “The Role of Laws and Institutions in California’s 1991 Water Bank,” in H. O. Carter, H. J. Vaux, Jr., and A. F. Scheuring (eds.), Sharing Scarcity: Gainers and Losers in Water Marketing, Agricultural Issues Center, University of California, Davis, June 1994b. Gray, Brian, “A Model Water Transfer Act for California,” Journal of Environmental Law and Policy, Vol. 4, Fall 1996, pp. 3–22. 163 Hanak, Ellen, and Caitlin Dyckman, “Counties Wresting Control: Local Responses to California’s Statewide Water Market,” University of Denver Water Law Review, Vol. 6, No. 2, Summer 2003. Howe, C. W., J. K. Lazo, and K. Weber, “The Economic Impacts of Agriculture-to-Urban Water Transfers on the Area of Origin: A Case Study of the Arkansas River Valley in Colorado,” American Journal of Agricultural Economics, Vol. 72, No. 5, December 1990, pp. 1200– 1204. Howe, C. W., “Protecting Public Values in a Water Market Setting: Improving Water Markets to Increase Economic Efficiency and Equity,” University of Denver Water Law Review, Vol. 3, No. 2, Spring 2000. Howitt, R., “Effects of Water Marketing on the Farm Economy,” in H. O. Carter, H. J. Vaux, Jr., and A. F. Scheuring (eds.), Sharing Scarcity: Gainers and Losers in Water Marketing, Agricultural Issues Center, University of California, Davis, June 1994. Hundley, N. H., The Great Thirst. Californians and Water: A History, revised edition, University of California Press, Berkeley, 2001. Illingworth, W., et al., “Analysis of Economic Impacts of Proposed Land Retirement in Westlands Water District,” draft prepared for Westlands Water District, September 2002. Imperial Irrigation District and U.S. Bureau of Reclamation, Water Conservation and Transfer Project Draft Habitat Conservation Plan and Draft Environmental Impact Report/Environmental Impact Statement, prepared by CH2MHill, El Centro, California, January 2002. Johnson, Hans P., “A State of Diversity: Demographic Trends in California’s Regions,” California Counts, Vol. 3, No. 5, Public Policy Institute of California, May 2002. Johnson, Lyn, “Water Deal One Step Closer, Maybe. . . . ,” Palo Verde Valley Times, November 1, 2002. Kasindorf, Martin, “War Over Water Splits California Cities, Farms,” USA Today, October 2, 2002. 164 Kasler, Dale, “Powerful Forces Are Bearing Down on Imperial Valley Farmers, Who No Longer Can Ignore. . . . The Call for Water,” Sacramento Bee, September 28, 2002. Kern County Water Agency, Water Supply Report 1998, March 2002. Knapp, Keith, Marca Weinberg, Richard Howitt, and Judith Posnikoff, “Water Transfers, Agriculture, and Groundwater Management: A Dynamic Economic Analysis,” Journal of Environmental Management, forthcoming. Lee, Hyunok, Daniel Sumner, and Richard Howitt, Economic Impacts of Irrigation Water Cuts in the Sacramento Valley, Agricultural Issues Center, University of California, Davis, 1999. Loomis, John, “Water Transfer and Major Environmental Provisions of the Central Valley Project Improvement Act: A Preliminary Economic Evaluation,” Water Resources Research, Vol. 30, No. 6, June 1994, pp. 1865–1871. Lund, Jay R., Morris Israel, and Richard Kanazawa, Recent California Water Transfers: Emerging Options in Water Management, Center for Environmental and Water Resources Engineering, Department of Civil and Environmental Engineering, University of California, Davis, Report 92-1, November 1992. M. Cubed, Regional Economic Impacts of the Palo Verde Test Land Fallowing Program, prepared for the Metropolitan Water District of Southern California, December 1994. M. Cubed, Socioeconomic Assessment of the Proposed Palo Verde ID Land Management, Crop Rotation and Water Supply Program, prepared for the Palo Verde Irrigation District, October 2002. McClurg, Sue, The Water Forum Agreement: A Model for Collaborative Problem Solving, Water Education Foundation, Sacramento, California, 2002. Mitchell, Laura, “Fallowing Debate at County Meeting,” Imperial Valley Press, September 4, 2002. 165 Mitchell, Laura, “Groundwater Permit Denied by Supervisors,” Imperial Valley Press, March 18, 2003. Ostrom, Elinor, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, New York, 1990. Phelps, Charles E., Nancy Y. Moore, and Morlie Hammer Graubard, Efficient Water Use in California: Water Rights, Water Districts, and Water Transfers, RAND, R-2386-CSA/RF, Santa Monica, California, 1978. Provencher, Bill, A Quantitative Analysis of Private Property Rights in Groundwater, Ph.D. dissertation, University of California, Berkeley, 1991. Provencher, Bill, “Issues in the Conjunctive Use of Surface Water and Groundwater,” in Daniel Bromley (ed.), Handbook of Environmental Economics, Blackwell Publishers, Cambridge, Massachusetts, 1995. Provencher, Bill, and Oscar Burt, “The Externalities Associated with the Common Property Exploitation of Groundwater,” Journal of Environmental Economics and Management, Vol. 24, 1993, pp. 139– 158. Purkey, D. R., G. A. Thomas, D. K. Fullerton, M. Moench, and L. Axelrad, Feasibility Study of a Maximal Program of Groundwater Banking, Natural Heritage Institute, San Francisco, California, December 1998. “Rural Palo Verde Valley Agrees to Colorado River Pact,” Associated Press, October 22, 2002. Sax, Joseph, “Understanding Transfers: Community Rights and the Privatization of Water,” Hastings West–Northwest Journal of Environmental Law, Policy, Thought, Vol. 1, No. 1, Spring 1994. Sax, Joseph, Review of the Laws Establishing the SWRCB’s Permitting Authority over Appropriations of Groundwater Classified as Subterranean Streams and the SWRCB’s Implementation of Those Laws, State Water Resources Control Board report number 0-076-300-0, Sacramento, California, January 2002. 166 Semitropic Water Storage District, Semitropic Water Storage District’s Policy for Fallowing Land to Make Surface Water Available for Other Purposes, Wasco, California, May 2002. “Summary of Water Agreement,” Desert Sun, Palm Springs, California, October 17, 2002. Sunding, David, “The Price of Water . . . Market-Based Strategies Are Needed to Cope with Scarcity,” California Agriculture, Vol. 54, No. 2, March–April 2000, pp. 56–63. Thomas, Gregory, Designing Successful Groundwater Banking Programs in the Central Valley: Lessons from Experience, The Natural Heritage Institute, Berkeley, California, August 2001. U.S. Bureau of Reclamation, Water Districts Maps: Federal, State, and Private, edition 1.6, Sacramento, California, October 2002. Vaux, H. J., and Richard E. Howitt, “Managing Water Scarcity: An Evaluation of Interregional Transfers,” Water Resources Research, Vol. 20, No. 7, July 1984, pp. 785–792. Vogel, Nancy, “Milestone Water Accord Ok’d: Southland Deal Furthers Shift of Rural Supplies to Cities,” Los Angeles Times, October 17, 2002. Water Education Foundation, The Water Awareness Guide: Where Your Water Comes From, Sacramento, California, 1994. Water Transfer Workgroup, Water Transfer Issues in California, final report to the California State Water Resources Control Board, June 2002. Yniguez, Rudy, “Frank Discussion Hoped For with State Officials on Third-Party Impacts,” Imperial Valley Press, July 25, 2002. 167 About the Author ELLEN HANAK Ellen Hanak is a research fellow at the Public Policy Institute of California. From 1992 to 2001, she was a research economist at France’s Center for Cooperation in International Agricultural Development. Before that, she held positions with the President’s Council of Economic Advisers and the World Bank. Her recent work has focused on the competitiveness of agricultural supply chains, food safety, and water policy. She holds a Ph.D. in economics from the University of Maryland. 169 Related PPIC Publications California’s Infrastructure Policy for the 21st Century: Issues and Opportunities David E. Dowall Making Room for the Future: Rebuilding California’s Infrastructure David E. Dowall and Jan Whittington Building California’s Future: Current Conditions in Infrastructure Planning, Budgeting, and Financing Michael Neuman and Jan Whittington The California Electricity Crisis: Causes and Policy Options Christopher Weare PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 171" } ["___content":protected]=> string(102) "

R 703EHR

" ["_permalink":protected]=> string(132) "https://www.ppic.org/publication/who-should-be-allowed-to-sell-water-in-california-third-party-issues-and-the-water-market/r_703ehr/" ["_next":protected]=> array(0) { } ["_prev":protected]=> array(0) { } ["_css_class":protected]=> NULL ["id"]=> int(8291) ["ID"]=> int(8291) ["post_author"]=> string(1) "1" ["post_content"]=> string(0) "" ["post_date"]=> string(19) "2017-05-20 02:36:18" ["post_excerpt"]=> string(0) "" ["post_parent"]=> int(3455) ["post_status"]=> string(7) "inherit" ["post_title"]=> string(8) "R 703EHR" ["post_type"]=> string(10) "attachment" ["slug"]=> string(8) "r_703ehr" ["__type":protected]=> NULL ["_wp_attached_file"]=> string(12) "R_703EHR.pdf" ["wpmf_size"]=> string(7) "1426074" ["wpmf_filetype"]=> string(3) "pdf" ["wpmf_order"]=> string(1) "0" ["searchwp_content"]=> string(256345) "Who Should Be Allowed to Sell Water in California? Third-Party Issues and the Water Market ••• Ellen Hanak 2003 PUBLIC POLICY INSTITUTE OF CALIFORNIA Library of Congress Cataloging-in-Publication Data Hanak, Ellen. Who should be allowed to sell water in California? : third-party issues and the water market / Ellen Hanak. p. cm. Includes bibliographical references. ISBN: 1-58213-075-2 1. Water transfer—California. 2. Water supply—California. I. Title. HD1694.C2H26 2003 333.91'009794—dc21 2003011570 Copyright © 2003 by Public Policy Institute of California All rights reserved San Francisco, CA Short sections of text, not to exceed three paragraphs, may be quoted without written permission provided that full attribution is given to the source and the above copyright notice is included. PPIC does not take or support positions on any ballot measure or state and federal legislation nor does it endorse or support any political parties or candidates for public office. Research publications reflect the views of the authors and do not necessarily reflect the views of the staff, officers, or Board of Directors of the Public Policy Institute of California. Foreword Over 40 years ago, Jack Hirshleifer, James C. DeHaven, and Jerome W. Milliman introduced the idea of California water markets in Water Supply: Economics, Technology, and Policy. Since then, numerous presentations have shown how California’s existing system of aqueducts could be used to move water to the highest bidder. Why then has the development of water markets—a policy that might provide enough water for decades of growth—taken so long to gain acceptance? Part of the answer is provided in Ellen Hanak’s report, Who Should Be Allowed to Sell Water in California? Third-Party Issues and the Water Market. The report notes many obstacles to the expansion of water markets; however, Hanak focuses on a key player—local governments. Local officials, especially in rural areas, are fearful of losing a resource that is a key component of future economic growth. And the specter of bone-dry Owens Valley haunts residents, officials, and investors alike. For these and other Californians, the problem can be put very simply: “No water, no life.” Although the amount of water sold through markets is only 3 percent of all water used in the state, 22 of the state’s 58 counties have adopted ordinances restricting groundwater exports. With the rise of groundwater transfers during the drought of the early 1990s, the fear of uncontrolled “mining” of the aquifers became widespread in many rural counties. In effect, the counties—through a burdensome review process and the prospect of negative public opinion—have discouraged potential sellers from seeking permits in the first place. Hanak finds that, controlling for other factors, these counties have been selling less water, and more of their sales have been to in-county buyers. The report also points to the need for effective policies pertaining to land fallowing, or idling crops to sell water. If fallowing affects other employment and business opportunities, a case can be made for economic mitigation. Some stakeholders are concerned, however, that iii direct compensation to those whose businesses are affected would establish a dangerous legal precedent, generate excessive claims, and create unrealistic expectations about the potential community benefits from water transfers. Given such concerns, communities may prefer the status quo to the risk of losing the benefits that flow from the control of this precious resource. Fair and sustainable rules for fallowing would go a long way toward balancing the needs of local users with the potential gains that result from water transfers. Finally, the report helps explain why it has taken so long to implement water markets. The concerns expressed at the local level— including the prospect of rapidly growing urban centers appropriating water without adequate compensation—are as real as ever. Nevertheless, Hanak shows that there are solutions in the making, and that with proper concern for users, local communities, and the environment, markets could play a key role in addressing California’s water supply problem for decades to come. David W. Lyon President and CEO Public Policy Institute of California iv Summary At current patterns of water use, California faces the prospect of chronic shortages of this vital resource before the year 2020. Among the measures that can alleviate supply and demand imbalances is the development of a water market. A market enables the historical holders of water rights—mainly farmers in the agricultural heartland—to transfer water to other users willing to pay more for it. Potential buyers include urban and industrial users, other farmers with higher-value crops and more limited supplies, and environmental programs to support fish and wildlife habitats. Although significant trading has occurred since the state began promoting this solution in the late 1970s, obstacles remain. In particular, communities in the source regions have raised concerns over the potential adverse effects of water sales on local groundwater users and the local economy. In the absence of clear state-level policy on these “third-party” effects, many counties are attempting to gain an oversight role through local ordinances. This study examines the issue of third-party effects of water transfers in California from the economic, institutional, and legal perspectives. It also evaluates potential mechanisms for resolving the conflicts between those wishing to trade in water and the wider community. Drawing on a range of data sources, including a new database on water transfers and an extensive set of interviews with water users and county officials, the analysis aims to answer the following questions: How has resistance to water transfers affected California’s water market to date, and what are the likely effects of that resistance? What distinguishes cases where conflicts have been successfully resolved from the stalled deals? Are revisions of state water law a necessary or desirable means for dealing with third-party issues, or should solutions be left to local institutions? We begin with some background on the water market and the rise in local resistance to it. v Water Market Trends Jumpstarted by a prolonged drought in the late 1980s and early 1990s, California’s water market is now a firmly established—if modest—feature of the state’s water allocation process, with annual trades accounting for roughly 3 percent of water use. The state has been a major player, notably by running drought year water banks and purchasing water for the environment. As expected, agricultural water districts are the main suppliers, with Central Valley farmers typically accounting for three-quarters of all sales and farmers in the desert valleys of Imperial and Riverside Counties furnishing the rest. Contrary to expectations, urban agencies have played a limited role in market growth. Instead, the main sources of demand have been directly and indirectly linked to new environmental regulations. Direct purchases for instream uses and wildlife reserves have accounted for over one-third of the increase in purchases since 1995. The other growth sector, accounting for over half of market expansion, has been agriculture in the San Joaquin Valley. Farmers there whose contractual water deliveries have been cut back by environmental mitigation programs have turned to the market for replacement water. However, municipal agencies are the principal buyers of long-term and permanent contracts, which account for roughly 20 percent of all sales. Legislation passed in 2001 requiring that local governments demonstrate adequate water supplies for development should increase urban demand for long-term water transfers. Municipalities’ success in forging these deals and ensuring new supplies will depend on their ability to smooth the waters of community resistance in the source regions. The Rise of Local Resistance to the Water Market Concerns in the source regions relate to two distinct types of negative effects of water marketing on third parties. When sales reduce the quantity or degrade the quality of water available to other users, this constitutes a physical externality. California law protects other surface water users, including fish and wildlife, from such effects under the “no injury” statutes of the Water Code. These protections do not extend to groundwater users, however, because groundwater—a major source of vi supply in many regions—is not regulated by the state. Once the state made it clear that the market was open for business during the early 1990s drought, the fear of uncontrolled “mining” of the aquifers became widespread in many rural counties. The other type of negative effect can occur when farmers idle cropland to sell water. Any resulting losses to the local economy—in jobs, sales, or local tax revenues—constitute an economic effect or pecuniary externality. There is no legal tradition in California or elsewhere in the United States for protecting third parties from this type of effect. The state’s widespread use of fallowing contracts to purchase water for the 1991 drought water bank generated considerable discord in some Sacramento Valley counties, where local businesses and farm workers were affected. Rural communities have responded to the lack of state-level, thirdparty protections by putting in place local restrictions on water marketing. By late 2002, 22 of the state’s 58 counties had adopted ordinances requiring a permit to export groundwater or to extract groundwater used in substitution for exported surface water. Counties’ right to invoke police powers to protect groundwater resources was upheld in a 1994 appellate court decision favoring Tehama County. In effect, the absence of state protections for groundwater users provides the legal justification for county-level action. In some counties, the ordinances reflect a broader intent to discourage any type of transfer—whether or not linked to groundwater—that might harm the local economy. Counties do not have the legal authority to ban crop idling for water sales, but some water districts have adopted policies to that effect. This appears mainly to be a practice of districts whose boards are elected by the community at large rather than districts where only landowners have a vote. The recent controversy over a proposed long-term transfer from the Imperial Irrigation District to San Diego erupted when Imperial—whose board is elected by popular vote—was pressured to fallow land despite district policy against the practice. Landowner-run districts have been more likely to fallow land for the water market, especially in periods of low crop prices when the water is less valuable in agricultural uses. vii Effects of County Restrictions on the Water Market To measure the effects of local resistance on the water market, the study assessed the role of county ordinances restricting exports. In counties with ordinances, those wishing to export groundwater or surface water that is replaced by additional groundwater pumping can go through an environmental review process to obtain a county permit. The very low number of permit applications suggests, however, that this process is more useful as a deterrent than as a screening mechanism. High up-front costs and the likelihood of negative public opinion guiding the decision process are both factors discouraging parties from filing. A lack of groundwater permits will not necessarily block transfers if alternatives such as land fallowing are available and acceptable to farmers and their water districts. In the aggregate, however, there is likely to be an effect on the market, both in reducing total sales and in shifting some water to in-county users, who will typically be willing to pay less than outsiders. A statistical analysis of county trading behavior from 1990 to 2001 provides evidence of both effects. In any given year, the presence of an export restriction reduced a typical county’s trades by 14,300 acrefeet and shifted 2,640 acre-feet to in-county buyers. Since 1996, total out-of-county sales, or “exports,” were reduced by 932,000 acre-feet, or 19 percent, and total sales by 787,000 acre-feet, or 14 percent. Overall, the negative market effect of county restrictions cancelled out the positive effect of a generally improved trading environment resulting from state and federal regulatory changes. The Scope for Resolving Third-Party Issues Local resistance is likely to remain a force to reckon with in market development, especially for the long-term, interregional transfers from agricultural users that municipalities will seek to support growth. Moving forward requires finding solutions that provide communities in source regions with adequate safeguards against the potential negative consequences to local water users and the local economy. What have the experiences to date taught us about the scope for positive resolution of viii these conflicts, and what role can policy play in this process? The responses are distinct for the two types of third-party effects. From Groundwater Protection to Groundwater Management Groundwater is a shared resource, with many users drawing from the same aquifer. In the absence of regulation, these users do not have clear incentives to avoid overexploiting the resource. Because the state does not exercise authority over groundwater, the onus for developing management systems falls on local users. Concerns over the groundwater effects of trade have arisen in California’s rural heartland, where local management systems are inadequate or altogether absent. In this context, county ordinances restricting exports can be justified as a firststep precautionary measure to protect local water users from the effects of an unbridled water market. This defensive strategy is nevertheless suboptimal from the standpoint of local as well as statewide interests. A policy limited to restricting exports does little to stabilize the aquifer in places subject to overdraft. It also makes it difficult, if not impossible, to make economic use of the underground storage space through groundwater substitution transfers and banking of imported surface water. Attaining these goals requires a more assertive, comprehensive strategy of groundwater management that protects local users while providing opportunities to address supply and quality problems and allowing those with sound transfer and banking projects to participate in the market. California’s rural areas have so far eschewed the more comprehensive management systems that govern groundwater in Southern California and in many coastal counties. In these regions, high population densities and special technical problems such as saltwater intrusion have led to the introduction of adjudicated basins and special districts with full regulatory authority over the resource. Nevertheless, there is a movement under way toward more active groundwater management in some of California’s rural counties. In some places, the county itself or a special district with countywide jurisdiction has played a convening role for county water users; in others, water districts overlying a shared basin have grouped together to develop a groundwater management plan. ix Key ingredients of active management include the establishment of effective basin monitoring systems and the development of guarantees to mitigate any harm to third parties from market-related activity. A question that remains on the table is whether a strictly voluntary management principle is adequate—a policy still favored by many rural Californians—or whether target levels and pumping restrictions need to be developed for the program to be effective. Developing effective local groundwater management systems places a central responsibility on local authorities—water districts and city and county governments. But the state also has a key role to play, given the statewide benefits of sound local management. Three current forms of state support are appropriate: providing technical assistance, making funds available to support system development, and encouraging the adoption of programs with sound content by attaching conditions to the release of state funds. Once systems are in place, there is also an opportunity for private funding of groundwater infrastructure, especially for municipal supply projects. Mitigating the Economic Effects of Land Fallowing For fallowing, the problem is one of determining the ground rules under which those with access to water rights may take land out of production and sell water to others. Available studies suggest that the aggregate local effects of fallowing have been quite small for programs idling anywhere from 6 to 29 percent of acreage, with local gains from the program largely balancing out local losses. But the modern track record is limited, and popular sentiment in rural areas tends to be shaped by the dire consequences of fallowing for the local economy in the Owens Valley almost a century ago. The key policy issues on the table concern the rules to limit negative community effects: rules on the scale and content of fallowing program design and rules concerning financial mitigation. Both state law and locally determined guidelines already address the first point. Section 1745.05 of the Water Code requires public review of fallowing that exceeds 20 percent of the local water supply. In designing fallowing programs, water districts increasingly include restrictions to maintain the viability of the idled land and to make sure that participating farmers are x not solely in the business of selling water. The economics of fallowing also plays a natural mitigating role. Farmers have incentives to fallow the crops that generate the least profit per acre-foot, and these tend to be the low-value, highly mechanized commodities that generate the lowest onfarm employment and the least value-added through further processing. Even with this combination of operating rules and incentives to limit negative effects to the local economy, there remains the question of whether the community should receive some sort of compensation. At the federal level, there are some precedents for mitigating economic effects when policy changes shift employment and business opportunities in some sectors or regions. With different degrees of success, federal mitigation programs have aimed to assist affected workers and businesses to make a transition to other economic activities. A parallel case could be made for mitigating the economic effects of sizable, long-term fallowing operations, especially if they generate systematic hardships for low-income groups or local governments. In two large long-term deals pending approval, a transfer from the Palo Verde Irrigation District to the Metropolitan Water District of Southern California and one from the Imperial Irrigation District to San Diego, funds have been earmarked for local communities. This will no doubt become a standard component of any future deals of this type, where large volumes of water are sold to distant urban agencies over more than a decade, with expectations of some systematic effects on local employment opportunities affecting low-income immigrant communities. For temporary or intermittent fallowing operations, such as those undertaken in the Sacramento Valley since 2001, there are larger questions about the appropriateness of mitigation. Two buyers, the Department of Water Resources and Metropolitan, have developed a policy to provide mitigation funds, but it remains unclear what damages, if any, merit mitigation. Many are uncomfortable with the term mitigation because it implies the direct compensation of affected parties. In part, this wariness stems from an expectation that the fallowing programs will generate little if any hardship to low-income workers, given the highly mechanized nature of production process for the rice crop being fallowed and the considerable workload generated by land xi maintenance and improvement activities on fallowed acreage. It also stems from a concern that a direct compensation program would establish a dangerous legal precedent, generate excessive claims, and ultimately create unrealistic expectations about the potential community benefits from water transfers. For these reasons, it may make more sense to think of such funds as providing opportunities for community development rather than mitigation. Since 1998, the legislature has considered three bills to institutionalize mitigation, but none has met with approval. Further legislative actions on the fallowing question should be avoided for the time being, for two reasons. First, there is a limited track record on fallowing and no experience with implementing mitigation funds. Second, in the major short- and long-term fallowing programs slated to occur, the transacting parties themselves have been adopting design measures to limit negative effects and setting up funds to benefit the community. These cases provide the opportunity both to assess the consequences of responsible fallowing and to experiment with use of funds for community benefit. If, as the farmers in the Sacramento Valley and Palo Verde argue, the overall effects are not harmful to the local economy, this may help build wider confidence in a new model for fallowing that can displace the ghost of Owens Valley. xii Contents Foreword..................................... iii Summary..................................... v Figures ...................................... xvii Tables ....................................... xix Acknowledgments ............................... xxi Acronyms .................................... xxiii 1. WATER MARKETING AND THIRD PARTIES ....... Water Marketing as a Component of California’s Water Future ................................ State and Federal Support for Water Marketing ......... The Rise of Third-Party Concerns in the Selling Regions ... The Scope for Resolving Third-Party Issues ............ 2. CALIFORNIA’S WATER MARKET, BY THE NUMBERS ................................ Who Can Sell Water and What Kinds Can They Sell? ..... Overall Market Trends ......................... Water for the Environment: A Key Factor in Market Growth ................................ Agriculture’s Leading Role in Market Supply ........... Most Transfers Are Local or Regional ................ From Farms to Cities: A Key Element of Long-Term and Permanent Transfers ....................... Summing Up ............................... 3. THE RISE OF LOCAL RESTRICTIONS ON WATER MARKETING .............................. The Mobilization of Rural Counties ................. An Overview of Export Restrictions ................. Legal Issues ................................ Economic Issues ............................. Operational Issues ............................ Summing Up ............................... 1 2 3 4 6 9 9 12 15 18 20 22 24 25 25 29 30 33 35 36 xiii 4. WHY DO SOME COUNTIES ADOPT EXPORT RESTRICTIONS? ............................ Factors That Make a Difference ................... Data on County Water Economy and Institutions ...... Cross-County Results ........................ Regional Issues .............................. The Mountain Counties: The Legacy of Owens Valley ... Sacramento Valley: A Balancing Act Between the Surface Water “Haves” and “Have-Nots” ............... San Joaquin Valley: Coping with Overdraft and Surface Water Scarcity ........................... Summing Up ............................... 5. WATER MARKET EFFECTS: DO COUNTY RESTRICTIONS HAVE TEETH? ................. Export Permitting: A Largely Uncharted Territory ....... Effects on the Water Market ...................... Data Sources .............................. Results .................................. Summing Up ............................... 6. MITIGATING THE ECONOMIC EFFECTS OF LAND FALLOWING .............................. California’s Recent Experiences with Land Fallowing ...... DWR’s Dry-Year Programs ..................... Long-Term Fallowing Along the Colorado River ....... Fallowing in the San Joaquin Valley ............... Future Trends? ............................. Economic and Legal Issues of Land Fallowing for Water Sales .................................. Economic Incentives and Third-Party Effects ......... Lack of Legal Provisions for Mitigation ............. Are Landowner-Run Water Districts More Likely to Fallow (and Less Likely to Propose Mitigation)? ...... Limiting the Aggregate Negative Effects of Land Fallowing .. Putting Together a Viable Mitigation Program .......... Determining the Scale of Losses .................. Program Content: Targeted or General? ............ 37 37 38 40 41 42 46 52 57 59 59 62 63 65 68 71 71 71 72 75 77 78 78 80 83 85 88 88 89 xiv Program Administration: Counties or Special Institutions? ............................. “Mitigation” or Community Development? .......... Summing Up ............................... 91 93 94 7. FROM GROUNDWATER PROTECTION TO GROUNDWATER MANAGEMENT .............. 97 Mitigating the Effects of Groundwater Transfers ......... 97 Economic and Legal Issues ..................... 97 Mitigation Options and Experiences ............... 99 Mitigation Design Issues ...................... 103 Ingredients of Effective Groundwater Management Systems.. 106 Role of Counties and Local Agencies ............... 109 Role of the State ............................ 111 Summing Up ............................... 114 8. THE SCOPE FOR RESOLVING THIRD-PARTY ISSUES ................................... 117 Local Groundwater Management as the Linchpin of Water Marketing and Water Banking ................. 119 Land Fallowing and Community Development.......... 122 Appendix A. Tracking the Water Market: Data Sources and Caveats .... 127 B. Groundwater Institutions and Basins ................ 139 C. Predicting County Adoption of Export Restrictions ....... 145 D. Measuring the Effect of Export Restrictions on County Water Sales ................................ 149 References .................................... 161 About the Author ............................... 169 Related PPIC Publications .......................... 171 xv Figures 2.1. Short- and Long-Term Water Transfers in California Since 1985 .............................. 13 2.2. Share of Total Water Transfers, by Type of Market, 1988–2001 .............................. 15 2.3a. Annual Volumes Purchased, by Type of End User ..... 16 2.3b. Market Share of End Users .................... 16 2.4. Nonenvironmental Water Purchases, by Location of Selling Party ............................. 21 3.1. Adoption of County Ordinances Restricting Exports ... 27 3.2. California Counties with Groundwater Export Restrictions, 2002 .......................... 28 5.1. California’s Water-Trading Counties, 1990–2001 ..... 64 5.2. Market Effects of Export Restrictions Since 1996 ...... 66 5.3. Market Effects of State and County Policy Environments ............................ 67 B.1. California’s Counties ........................ 141 B.2. Adjudicated Groundwater Basins ................ 142 B.3. Special Groundwater Management Districts ......... 143 B.4. Critically Overdrafted and Special Problem Groundwater Basins Listed in Bulletin 118-80................. 144 xvii Tables 2.1. Regional Sources and Destinations of Water ......... 19 4.1. Average Characteristics of Counties With and Without Export Restrictions ......................... 40 4.2. Regional Characteristics (County Averages) ......... 43 6.1. Study Estimates of the Overall Economic Effect of Land Fallowing ............................... 82 A.1. California Short- and Long-Term Water Transfers, by Type of Market ........................... 132 A.2. Water Purchases, by Type of End User ............ 133 A.3. Transfers, by Region of Origin and Region of Destination .............................. 134 A.4. Source Regions for Environmental Water Purchases .... 135 A.5. Long-Term Transfers Since 1985 ................ 136 A.6. Permanent Transfers of Surface Water Since 1985 ..... 137 B.1. Counties with Groundwater Protection Ordinances, by Region ................................. 139 C.1. Effects of County Characteristics on the Probability of Adopting an Export Restriction ................. 147 D.1. Summary Statistics for Annual County Water Sales and Water Exports, 1990–2001 .................... 154 D.2. Determinants of Annual County Water Sales in 34 Water-Trading Counties, 1990–2001 ............. 157 D.3. Determinants of Annual County Water Exports in 34 Water-Trading Counties, 1990–2001 ............. 158 D.4. Determinants of Annual County Water Sales in 18 Central Valley Counties, 1990–2001 ............. 159 D.5. Determinants of Annual County Water Exports in 18 Central Valley Counties, 1990–2001 ............. 160 xix Acknowledgments This study would not have been possible without the generous assistance of many members of California’s water community. Individuals in a wide range of institutions, from local water districts to state and federal agencies, provided the building blocks of the original database on California’s water market. County and water district officials and other water experts graciously agreed to be interviewed regarding local responses to the rise of the market. At various stages, the study also benefited from the input of an informal advisory panel. Special thanks go to the attendees of a December 2002 meeting in Sacramento, convened to review a draft of the study’s findings and to discuss the policy implications. Attendees included Curt Aikens of the Yuba County Water Agency; Will Boschman of the Semitropic Water Storage District; Chris Campbell of Baker, Manock and Jensen; Jim Easton of Easton Water Resources; David Guy of the Northern California Water Association; Carl Hauge, Jerry Johns, and Rob Swartz of the Department of Water Resources; Richard Howitt of UC Davis; Wendy Illingworth of Economic Insights; Dave Orth of the Kings River Conservation District; Bill Phillimore of Paramount Farms and the Kern Water Bank; Tim Quinn of the Metropolitan Water District of Southern California; Allen Short of the Modesto Irrigation District; Tracy Slavin and Donna Tegelman of the U.S. Bureau of Reclamation; Steve Stroud of the South San Joaquin Irrigation District; Van Tenney of Glenn-Colusa Irrigation District; and Greg Thomas of the Natural Heritage Institute. Although responsibility for the conclusions rests entirely with the author, this feedback was invaluable. I would also like to extend a special thanks to Dave Sunding of UC Berkeley, Mike Kahoe of the San Joaquin Valley Water Users Association, Carol Whiteside and Holly King of the Great Valley Center, xxi and Jim Ganulin of Baker, Manock and Jensen for help in defining the issues in the early stages of the work. Two very capable research associates were key to the research effort. Antonina Simeti did much of the painstaking work of creating the database on the water market and pulling together other datasets used in the study. Caitlin Dyckman, a graduate student in the UC Berkeley Department of City and Regional Planning, worked on the evolution of county groundwater policies and conducted many of the interviews with county officials. The study was finalized with the help of Greg Thomas, Richard Howitt, Howard Schatz, and Ricardo Ramirez, who provided written reviews, and Peter Richardson and Patricia Bedrosian, who provided editorial assistance. xxii Acronyms AB AVEKWA BMO CALFED CEQA CVP CVPIA DWR EIR EWA GCID IID KWCA MWDSC NHI PVID RCRC SB SDCWA SWP SWRCB TAC UC USBR WAC YCWA Assembly Bill Antelope Valley–East Kern Water Agency Basin management objective Multiagency state and federal program for the San Francisco Bay Delta California Environmental Quality Act Central Valley Project Central Valley Project Improvement Act Department of Water Resources Environmental impact review Environmental Water Account Glenn–Colusa Irrigation District Imperial Irrigation District Kern County Water Agency Metropolitan Water District of Southern California Natural Heritage Institute Palo Verde Irrigation District Regional Council of Rural Counties Senate Bill San Diego County Water Authority State Water Project State Water Resources Control Board Technical Advisory Committee University of California U.S. Bureau of Reclamation Water Advisory Committee Yuba County Water Agency xxiii 1. Water Marketing and Third Parties Over most of the past century, the water needs of California’s growing economy and population have been met by increasing water supplies. The result has been a complex mosaic of hydraulic investments in surface storage and conveyance undertaken at local, state, and federal levels. This “developed” surface water system provides the state’s agricultural, residential, and industrial customers with roughly two-thirds of the water they use, the remaining third coming from groundwater reserves. Until recently, the system generally has met the needs of these users, except in periods of severe drought. Concerns that insufficient water resources were being devoted to the environment—and in particular to the survival of endangered wildlife—have led to reductions, beginning in the early 1990s, in the amounts available to some agricultural and municipal users. With the environment now explicitly considered as a user with legitimate (and sometimes paramount) requirements, and with the promise of continued, rapid population growth, the state’s Department of Water Resources projects chronic water shortages before the year 2020 at current patterns of use (Department of Water Resources, 1998). Redressing supply and demand imbalances through additional surface reservoir development has become more difficult. Hydrological options are less favorable, cost considerations loom larger, and concerns about the potential environmental consequences of such investments play a much greater role in the public decisionmaking process. This difficulty has prompted considerations of alternative technologies to increase supply, through “groundwater banking”—or storing excess run-off in underground aquifers—and through recycling and desalination. It has 1 also heightened interest in making better use of the existing supply through conservation efforts and the development of a water market. Water Marketing as a Component of California’s Water Future A market permits the temporary, long-term, or permanent transfer of water from the existing rights-holders to other water users in exchange for payment. In California, these rights generally have been appropriated for many decades under the state’s “first in time, first in right” legal system. Water transfers are seen as a way of adding flexibility to the state’s water supply—both to address temporary drought conditions and to accommodate longer-term changes in the pattern of demand. Because water systems have been intimately linked to the development of California’s agricultural heartland, the bulk of use rights are held by farming interests in the Central Valley and the desert counties to the south that rely on Colorado River water (Hundley, 2001). In normal water years, agriculture uses about 34 million acre-feet of water, or 80 percent of the total used by Californians for their combined residential and business needs (Department of Water Resources, 1998). Agricultural users often pay significantly less for water than municipal and industrial users do in the coastal metropolitan areas, even allowing for differences in transportation and treatment costs. Ready availability of water has enabled California to become the nation’s largest agricultural economy, with one of the world’s most extensive irrigation systems. This can be seen as a successful outcome of past federal and state support to the development of western agriculture through large-scale hydraulic projects. As other demands for water continue to grow, however, there has been increasing pressure to weigh this policy goal against one emphasizing the scarcity of this natural resource. In an era when pricing has been advocated as a solution for a whole range of resource allocation issues, water transfers are seen as a way to accommodate the changing pattern of demand while compensating water-rights-holders for forgoing their own access. The discussions on water marketing have often emphasized the potential of the market to move water from agricultural to growing urban areas, many of which are 2 willing to pay more for the water than it is worth to the seller in the nextbest agricultural use. But a water market can also help balance supply and demand within the agricultural sector, as farmers with higher-value uses for the water purchase it from those with more-senior rights and less-productive farms. Finally, market-based transfers are seen as a potential tool for reallocating water from agriculture to the environment through voluntary means rather than rationing. State and Federal Support for Water Marketing California’s foray into water marketing began in 1977, a year of severe drought. Two reports commissioned at that time, one by the governor and one by the legislature,1 strongly endorsed water marketing as a component of the state’s water future. The governor’s commission also advocated a number of changes in the Water Code to facilitate transfers, notably provisions to ensure the security of water rights for transferring parties and access to the use of conveyance facilities. Although many of the recommendations were accomplished in the years that followed, the 1980s saw little uptake in market activity. In the early 1990s, several events significantly changed the trading climate. First, natural conditions provided the occasion for a large-scale experiment in water trading when a multiyear drought prompted the state to initiate an emergency water bank in 1991. The following year, in response to findings that the federally run Central Valley Project (CVP) was having deleterious effects on the indigenous wildlife of the San Francisco Bay-Delta water system, Congress passed the Central Valley Project Improvement Act (CVPIA). The CVPIA mandated that 800,000 acre-feet of project water (of a total of 7 million) be returned to instream uses to regenerate salmon runs and that another 400,000 acrefeet be allocated to wildlife refuges. The CVPIA also contained provisions to facilitate water marketing and introduced a mechanism for the project to purchase additional water for environmental purposes. In 1994, contractors of the State Water Project (SWP) concluded negotiations for the Monterey Agreement, a revision of project operating ____________ 1Governor’s Commission to Review California’s Water Rights Law (1978); Phelps et al. (1978). 3 rules that included measures to make it easier for contractors to transfer water to one another. At the close of the decade, two further state and federal actions were significant. Under instructions from the Secretary of the Interior in 1996 and 1997, California began to devise a plan to reduce its use of Colorado River water to the contractually allocated amount of 4.4 million acre-feet over a 15-year period. This 4.4 Plan created strong incentives for water transfers between agricultural and urban users of Colorado River water within California. In 2000, state and federal authorities launched the Environmental Water Account (EWA), a program of water purchases for the environment under CALFED, a multiagency state and federal program to restore health to the fisheries of the San Francisco Bay-Delta system while securing water supplies to agricultural and urban users. The Rise of Third-Party Concerns in the Selling Regions Although substantial trading resulted from these policy changes, those wishing to enter the water market—on either a short- or long-term basis—have encountered a number of obstacles. Central among these is the question of how to deal with the “third-party” or indirect effects of the transfers on other water users and the larger communities where water is being sold. Community resistance has soured a number of deals over the past decade and has likely prevented others from being proposed. Many of California’s rural counties have introduced ordinances that directly restrict groundwater exports and indirectly restrict the sale of surface water. In several counties, the ordinances circumscribe the rights of individuals and local water agencies to engage in groundwater banking. These measures respond to the potential consequences of the transfers on the local community. Short-term transfers during the drought water bank in the early 1990s raised two contentious issues. First, communities were concerned about local economic consequences when some farmers fallowed their land to sell water, simultaneously drying up demand for labor and other farm inputs and cutting off the 4 supply of raw materials to local processors. Second, farmers who sold their surface water and maintained their crops by pumping more groundwater than usual were in some cases seen as reducing the overall quantity and quality of supplies available to other water users. Community concerns can be exacerbated if the seller makes a multiyear commitment. For the proposed long-term transfer of water from the Imperial Irrigation District to San Diego, negotiated as part of California’s 4.4 Plan to reduce Colorado River use, leaders in the Imperial Valley have been reluctant to agree to multiyear land fallowing. Sending water toward urban coastal areas adds to the uneasiness of such long-term transfers: Will such areas take precedence in the future, regardless of the terms of the transfer? Some observers express a general skepticism regarding the market for water, given the sharp regional disparities between the wealthy metropolitan areas on the coast and California’s agricultural valleys, which lag far behind in terms of income, employment, and education levels. In this view, arguments about market efficiency are little more than a justification for draining the poorer hinterland of the resources it may need for more favorable growth and development in the years ahead. Although the tone of the debate is at times highly charged, local officials’ resistance to transfers can be seen as a precautionary approach to a potential irreversibility problem: Once the water is gone, they will lack the money and political influence to get it back. Reinforcing this concern is the specter of the Owens Valley affair—the notorious deal in the early 20th century wherein Los Angeles bought up so much of that region’s water that the local agricultural economy collapsed. Community opposition to water transfers challenges the notion that water rights belong to users and suggests instead that water is a community resource. The county ordinances are an attempt to impose a level of local oversight on water transfers not provided for in state law, which is not uniform in its treatment of transfers or in the assessment of effects on third parties. State approval is required only for transfers involving surface water rights acquired since 1914, certain types of groundwater banking, and any water that is “wheeled” (conveyed) through a publicly owned facility (e.g., the California Aqueduct). Only in the latter case must the state ensure against unreasonable negative 5 economic effects on the source county; for the other transfers, there is simply an obligation not to harm other legal users of the surface water (i.e., other rights-holders), fish, and wildlife. There is no provision to protect other groundwater users, as groundwater rights remain largely unregulated by the state. Although there is some question as to whether the county ordinances would stand up to a legal challenge, there is no doubt that they arise where communities are not confident about the existing legal and political process governing transfers. The question nevertheless remains as to whether counties are managing community interests effectively; specifically, an overly conservative policy on transfers can lead to missed opportunities for enhancing collective welfare. The Scope for Resolving Third-Party Issues This study examines the issue of third-party effects of water transfers in California from the economic, institutional, and legal perspectives. It evaluates potential mechanisms for resolving the conflicts between those wishing to trade in water and the wider community. By drawing on a range of data sources, including original survey work, the analysis aims to answer the following questions: How has local resistance to water transfers affected California’s water market to date and what are the likely future effects? What distinguishes the cases where conflicts have been successfully resolved and the stalled deals? Are revisions of state water law a necessary or desirable means for dealing with third-party issues, or should solutions be left to local institutions? The report begins by providing some background on the water market and on the extent of community response. Chapter 2 presents an overview of the statewide water market since the mid-1980s, drawing on a new dataset developed for this study. It shows the evolution of total volumes transferred and the composition of the market—by geography, types of water users, and shifts in end use—in response to the state and federal policy changes outlined above. The next two chapters document the rise in local restrictions on water marketing, with a specific focus on the adoption of county ordinances. Chapter 3 summarizes the statewide trends and discusses some of the legal, economic, and operational issues that the ordinances 6 raise. Drawing on county-level economic indicators and qualitative information gathered from interviews with county officials and water users, Chapter 4 explores why certain counties have adopted export restrictions and others have not. Chapter 5 examines whether and how local opposition affects the water market. This examination includes a review of the permitting process for exports in counties that have adopted restrictions and a statistical analysis of the factors influencing water transfer behavior at the county level. The study then turns to the policy questions that arise from the economic effects of land fallowing and the physical effects on water users arising from transfers involving groundwater. Both economic theory and institutional practice provide justifications for mitigating negative effects of water transfers. In theory, any transfer that truly raises efficiency by moving a resource into a higher-value use will generate sufficient gains to enable the losers to be compensated, such that the well-being of all parties is enhanced or maintained. For transfers that affect the availability of the physical resource to other water users, compensation is justified on grounds of efficiency as well as equity (Coase, 1960). This principle already underlies the state’s policy to “make whole” other surface water users affected by a transfer. More generally, it forms the basis for environmental mitigation of development projects. Measures might include providing alternative sources of water supply to the affected party or devising some other form of financial or in-kind compensation. Compensation mechanisms are not always easy to devise and reach agreement on, however, particularly when the damages are unclear or difficult to quantify. Chapter 6 examines these issues as they apply to the economic effects of land fallowing, and Chapter 7 investigates the physical effects on water users of groundwater transfers and the related practice of groundwater banking. Mitigating harm to groundwater users implies the existence of an effective system for managing groundwater resources more generally. Because these resources are almost always shared by multiple parties, groundwater management usually requires collective oversight. In California, there is a vigorous debate concerning the appropriate level of governance for groundwater. At one extreme are those who argue that 7 groundwater should come under the jurisdiction of the state alongside surface water (Sax, 2002). At the other extreme are those who argue against any restrictions on individuals’ rights to pump. The middle ground in this debate has sought solutions involving local institutions. Two potentially competing approaches have grown in popularity since the early 1990s—the establishment of groundwater management plans by local water entities overlying a shared basin and the coordination of groundwater management at the county level through groundwater protection ordinances. Chapter 7 examines the policy and institutional questions that arise from these approaches, with a particular focus on the role counties are and might be playing. Chapter 8 concludes by summarizing the policy issues that face state and local governments in addressing third-party issues and the role local entities can and should play in the statewide water market. 8 2. California’s Water Market, by the Numbers This chapter documents the evolution of the state’s water market from the mid-1980s to 2001, drawing on a new dataset on individual water transfers developed by the author from a variety of state, federal, and local sources. The data allow an analysis of volumes transferred by duration, region of origin and destination, initial and final use, type of transacting party, and affiliation (if any) with the large state and federal water projects. The discussion highlights two areas of particular interest. First, to what extent has the market responded to the policy measures to encourage trading over the past two decades? Second, does the trading that has occurred correspond to expectations? Who Can Sell Water and What Kinds Can They Sell? The state’s Water Code provides two basic guidelines on who can participate in the water market and what they can sell: Sellers must be rights-holders, and the water they sell must be “wet.” “Wet water” is the term commonly used to contrast with “paper water”—water rights held on paper for which actual water is not available. Under the appropriative rights doctrine governing most of the state’s surface water, the “use it or lose it” clause dictates that rights lapse for any water not used for five consecutive years. The Water Rights Division of the State Water Resources Control Board (SWRCB) estimates that there are roughly three times as many paper claims on surface water as water available in the system. Water-rights-holders must therefore demonstrate that the water they propose to part with is indeed water that they would otherwise use in some way. Without this safeguard, the seller would end up transferring “paper” water that someone else is already legally using. 9 Sources of wet water are of four basic types: surface storage, excess surface water, conserved surface water, and groundwater. The first two sources are not widely available. Only a limited number of rights-holders have surpluses available in surface storage, and the use-it-or-lose-it principle limits conditions under which excess surface supplies are actually considered wet. Contractors with the CVP and SWP projects are the only ones who have been able to sell regular excesses of surface supplies, on grounds that their rights are determined by contract rather than the appropriative doctrine.1 Otherwise, rights-holders may sell the excess surface water generated in very wet years. These are times when overall market demand is more limited. Conservation and groundwater are more generally available options. Conservation can be achieved through investments to improve the efficiency of the conveyance and use systems (e.g., canal lining, installation of drip irrigation, recycling), or through land fallowing. Groundwater can be transferred directly or, more commonly, can be used on-site in lieu of surface water transferred to another party. This latter practice, known as “groundwater substitution” or “groundwater exchange,” is a form of “conjunctive use” of groundwater and surface water. With conjunctive use, the groundwater aquifer serves, in effect, as an underground reserve that can be drawn on to a greater or lesser degree as the quantity of available surface water varies. In some places where prior pumping activity has generated unused space in the aquifer, active storage or banking projects are increasingly popular. As a consequence, storage in underground banks is becoming a new source of water for the market, much like surface storage. Although the initial appropriation of water rights in the second half of the 19th century primarily involved individuals and private companies, most surface water rights today are held by local public agencies: special water and irrigation districts and some municipalities. Legally, some of these parties actually hold long-term “contract entitlements” rather than “rights” to surface water; the large state and federal projects they contract with hold the actual water rights. In some ____________ 1A loosening of restrictions on such trades among Central Valley Project contractors is arguably the main effect of the CVPIA on the water market to date. 10 water districts, individual farmers have specific contractual amounts of water (or “allocations”) assigned to them and are therefore in a position to sell their surface water. Generally, the right to pump groundwater (and hence, potentially, to sell it) is available to all private individuals overlying the aquifer. As we will see in greater detail in subsequent chapters, there are few places where rights are “quantified” (i.e., where users have rights to a specific quantity of water) or where local agencies, rather than individuals, effectively control access to the aquifer. Current groundwater banking operations typically involve local agencies (for instance, the Semitropic Water Storage District and the Arvin-Edison Water Storage District in Kern County each run programs) or consortia of private and public agencies (for example, the Kern Water Bank). Such programs could conceivably be run by a single private entity if it were sufficiently large to cover the banking area. A case in point is the recently abandoned Cadiz groundwater storage and dry-year supply project. In this project, a private agribusiness firm with land holdings in eastern San Bernardino County would have banked and sold water to the Metropolitan Water District of Southern California. The approval process for transfers varies according to the nature of the water right as well as the source of water. The SWRCB must approve transfers (changes in purpose or place of use) involving surface water rights established from 1914 onward, the year the state’s “modern” Water Code became effective. Transfers of surface water among contractors of the federal and state water projects generally do not require SWRCB approval because they do not involve a change in the purpose and place of use assigned to the overall water right. The projects themselves must authorize these sales. The general practice is for farmers to go through the water district, which negotiates the transaction, and for the state or federal project to decide whether the district may make the sale. Transfers of groundwater and of surface water held in pre-1914 appropriative rights do not require SWRCB approval. However, they do come under state or federal jurisdiction if government-owned conveyance facilities are involved, which is likely to be the case in most parts of the state. 11 It was not possible to track systematically the type of water being sold in the data presented below. On average, local public agencies (water districts, irrigation districts) account for over 90 percent of the volumes sold since the mid-1980s, with private parties and mutual water companies making up the balance. The share of direct private activity was highest during the early 1990s, when the state contracted directly with individual farmers for a substantial portion of the water acquired under the drought water bank. The data presented focus on annual flows of water resulting from temporary transfers (under one year), long-term transfers (over one year), and what we have termed “deferred exchanges,” which involve a promise that the buyer repays the water (along with a cash payment in some cases) to the seller at a later date. The data on a fourth category, the permanent transfer of water rights or contract entitlements, are presented separately. Such transfers amount to an outright sale of the rights to use the specified amount of water in perpetuity or for the remaining duration of the contract in question. Because the actual amount accessible to the buyer can vary with the conditions of the water year, it is not strictly appropriate to consider an annual flow of water transferred. For more details on the sources and methods used to construct the water transfers database, the reader is referred to Appendix A. Overall Market Trends The statewide water market got a jumpstart during the last multiyear drought, which began in 1988 (Figure 2.1). Market growth was largely driven by direct interventions of the Department of Water Resources. DWR began making dry-year purchases to offset lower deliveries to State Water Project contractors and wildlife refuges in the first year of the drought. These early operations, which involved only a handful of Sacramento Valley water districts (most notably the Yuba County Water Agency), quickly brought the total volume traded to over 500,000 acrefeet, three times the pre-drought levels. In 1991, when the dry-year market was opened up to any willing buyers and sellers, DWR purchased 821,000 acre-feet of water for resale, bringing the overall market volume to over 1.1 million acre-feet. Water banks and other dry-year purchases were also operated in 1992 and 1994. From 1988 to the end of the 12 Acre-feet traded (thousands) Rainfall index 1,400 1,200 1,000 800 600 400 200 0 Rainfall Acre-feet traded 18 17 16 15 14 13 12 11 Wet 10 9 8 Normal 7 6 5 Dry 4 3 11111111111122111999999999999009999999888889990099932109876556410987 NOTES: For details, see Table A.1. Rainfall is measured by the Sacramento Valley 40-30-30 index, an indicator of water supply conditions for the state’s primary river system (see Appendix D). Figure 2.1—Short- and Long-Term Water Transfers in California Since 1985 drought in 1994, state and federal dry-year purchases for resale and environmental uses accounted for over 40 percent of a market that had jumped from an average of 150,000 acre-feet to over 600,000 acre-feet per year. 2 Although the second half of the 1990s saw a succession of wet years, market activity remained strong, with volumes typically exceeding the drought-year levels, especially by the end of the decade. The only dips in ____________ 2The average market volume as measured by end-user purchases is just over 500,000 acre-feet for the 1988–94 period. This discrepancy arises mainly because DWR’s purchases were higher than its resales to other users in some years. The gap was most notable in 1991, the first year of the water bank, when purchases exceeded resale volume by over 400,000 acre-feet. In several years in the 1990s, there were also smaller discrepancies between purchases and sales of water user pools, notably the SWP “turnback” pool, described in the text, and the Sacramento River Water Contractors’ Association, a pool run by some CVP contractors. 13 a generally upward trend in purchases occurred in the exceptionally wet years of 1995 and 1998, when many areas of the state experienced flooding. With annual trades now exceeding 1.2 million acre-feet— eight to ten times the levels of the mid-1980s—the market appears firmly established as a component of California’s water allocation process. It is also useful to consider the size and scope of the market from three other perspectives. First, the statewide market at current levels represents only 3 percent of all water used by Californians for municipal, industrial, and agricultural purposes (Department of Water Resources, 1998). Second, although there has been an increase in the number of long-term transfer agreements, the market continues to be dominated by short-term transactions, negotiated on an annual basis, which account for about 80 percent of the total volume transferred. Finally, the size and scope of the market are strongly influenced by the intervention of state and federal authorities. This influence stems not only from their important direct role in purchases but also from the relative ease water users have in gaining approval for transfers within the confines of the state and federal projects. Since 1988, direct government purchases have accounted for nearly one-third of the total volume traded. Transfers among contractors within the same projects (SWP, CVP, and the Colorado River Project) account for more than half of all water sold (Figure 2.2). Transfers within the CVP and the SWP have generally increased in response to the more-flexible trading rules adopted by those projects in the early 1990s. The growth has been most pronounced within the SWP, for which internal transfers were rare before the Monterey Agreement. By contrast, the “open market,” a residual category defined broadly to include any transfers between water users not associated with the same project, accounts for only 15 percent of the water transfers recorded over the 14-year period. This share initially increased immediately after the drought, but it has been on the decline again recently, as direct government purchases for environmental programs have been on the rise. In this context, it is noteworthy that a provision in the Central Valley Project Improvement Act to allow project water to be sold to noncontractors, considered a major innovation at the time, has gone virtually unused. Such transfers invoke a $25 per acre-foot 14 Within Colorado River Project, 16% Within SWP, 10% Direct government purchases, 31% Within CVP, 28% Open market, 15% Figure 2.2—Share of Total Water Transfers, by Type of Market, 1988–2001 environmental surcharge, a fee that appears to have proven prohibitive until now.3 Water for the Environment: A Key Factor in Market Growth Market growth in the aftermath of the drought has been largely driven by environmental concerns. The influence of environmental policy is most readily seen by comparing the patterns of water purchases during the multiyear drought (1988–1994) to those in the more recent period (1995–2001), when rainfall has generally been above normal (Figures 2.3a and 2.3b). The most obvious element of the new role for the environment is the rise in direct purchases for instream uses to support aquatic life and for wildlife refuges, through federal and state programs and most recently ____________ 3With the aid of a simulation model, Loomis (1994) predicted that this surcharge would make it unlikely for out-of-project transfers to occur except under severe drought conditions, when buyers would be willing to pay enough to cover it. 15 Acre-feet per year (thousansds) 1,000 900 800 700 600 500 400 300 200 100 0 1988–1994 NOTE: For details, see Table A.2. 1995–2001 Mixed uses Other farmers San Joaquin Valley farmers Environment Cities Figure 2.3a—Annual Volumes Purchased, by Type of End User 100 90 80 70 60 50 40 30 20 10 0 1988–1994 NOTE: For details, see Table A.2. 1995–2001 Mixed uses Other farmers San Joaquin Valley farmers Environment Cities Figure 2.3b—Market Share of End Users Percentage 16 the CALFED EWA. As one of the beneficiaries of DWR’s drought purchases, the environment already accounted for 12 percent of demand during the 1988–94 drought. Since 1995, this share has doubled on average; it reached a third of total demand in 2001, the first full-fledged year of the EWA. On an average yearly basis, environmental purchases have increased by 155,000 acre-feet, out of a total market gain of 430,000 acre-feet. The less obvious component of environmentally related demand is the rise in purchases by San Joaquin Valley farmers. Although this group’s change in market share is less dramatic (moving from 37 percent to 44 percent over the two periods), its increase in volume (228,000 acrefeet per year) accounts for over half of total market growth. Much of this growth can be linked to the changes introduced under the CVPIA in 1992, which mandated that a portion of project water be returned to instream uses. Since then, the CVP’s agricultural service contractors located south of the Delta received full project deliveries in only the two very wet years, 1995 and 1998. One outcome has been the development of an active water market, as certain contractors (most notably Westlands Water District) sought to offset reductions in deliveries via purchases. Within the San Joaquin Valley, CVP service contractors’ market share of purchases has risen from 63 percent to 87 percent across the two time periods. The corollary of this growth has been a decline in the relative importance of municipal and industrial water users. Whereas cities were the main recipients of traded water during the 1988–1994 drought with 42 percent of all purchases, their share since 1995 has been halved. With the exception of 1991, when purchases nearly reached 500,000 acre-feet, volumes have remained relatively flat, at around 200,000 acre-feet per year. The modest role of urban demand contrasts with the frequently voiced assumption that a water market would develop primarily as a response to population growth and the ability of urban dwellers to pay more than agricultural users for water.4 As we will see below, cities are significant players in defining the market for long-term and permanent ____________ 4See Phelps et al. (1978) and Vaux and Howitt (1984) for early economic arguments in favor of a water market in California. 17 transfers of water. This pattern suggests that their role may increase as this part of the market develops. Agriculture’s Leading Role in Market Supply Economists have typically assumed that agriculture would be the leading source of water supply, and this assumption is confirmed by the data. In most years, agricultural water users provide at least 90 percent of supply. It is nevertheless interesting to note that the introduction of “turn-back” pools within the State Water Project has made a significant amount of urban agency water available to agricultural users in wet years. Under the program, introduced as one of the reforms in the 1994 Monterey Agreement noted in Chapter 1, SWP contractors are able to sell back amounts of project water they will not need if there are willing SWP buyers.5 From 1998 to 2000, the turn-back pool generated 200,000 acre-feet or more of water per year, or roughly one-fifth of total market supply. Given the primacy of agricultural supply, it is not surprising to find that the main source regions are the Central Valley (served by the CVP, the SWP, and several large, autonomous, surface water projects) and the agricultural valleys to the south served by the Colorado River Project (Table 2.1). In most years, the Central Valley has furnished about threequarters of the total volume transferred. Within this region, there are pronounced shifts, depending on the nature of the water year. From 1988 to 1994, Sacramento Valley farmers and districts supplied about 45 percent of all water sold. Although this share fell considerably after 1994 (in some years, to under 10 percent of the statewide market), the region again provided over 40 percent of the water in 2001, the first dry year experienced since the previous drought. In the interim, the market shifted to the San Joaquin Valley, which established a vibrant intraregional market to supply water-short districts within the region. In every normal to wet year since 1993, the first year of CVPIA ____________ 5Because the sale price is fixed below the level of per-acre-foot charges incurred by the seller for project water, the pool’s attractiveness to sellers is relative, not absolute: It enables them to pay less than full price for contract amounts in excess of current needs. 18 Table 2.1 Regional Sources and Destinations of Water Sales by water users Sacramento Valley San Joaquin Valley Southern California San Francisco Bay Area Other 1988–1994 1995–2001 1,924,937 1,363,037 970,942 87,195 3,055 1,057,064 3,715,039 1,577,597 82,575 88,694 Total sales 4,349,166 6,520,969 Purchases by water users Sacramento Valley San Joaquin Valley Southern California San Francisco Bay Area Other 135,079 1,450,917 1,187,157 313,197 6,152 515,509 3,253,292 1,234,555 43,505 14,993 Purchases of environmental water % from Sacramento Valley % from San Joaquin Valley 408,672 74 25 1,484,255 17 81 Total purchases 3,501,174 6,546,109 Exports (imports) of nonenvironmental water Sacramento Valley San Joaquin Valley Southern California San Francisco Bay Area 1,488,725 (190,683) (216,215) (230,738) 288,383 (739,455) 343,042 9,070 NOTES: For details, see Tables A.3 and A.4. The bulk of the difference between total purchases and total sales in the first period is the surplus purchased by DWR and distributed through means other than sales to other end users. There were also some small discrepancies in both periods between purchases and sales of user pools. These discrepancies account for the fact that nonenvironmental exports and imports do not sum to zero. See footnote 2 in this chapter. All water measurements are in acre-feet. implementation, San Joaquin Valley farmers and districts have furnished at least half of the total amount transferred. It is common for agricultural districts in this region to restrict out-of-district transfers to cases where land in the recipient district is owned or leased by the same farmer. 19 The two Central Valley regions are the principal suppliers of environmental water (Table 2.1). There has been a major source shift from north to south of the Delta since the early 1990s drought, with the institution of restoration programs along the San Joaquin River system, supplied by area water districts. Kern County water users have also been major suppliers of the EWA. Most environmental water is used in the region of purchase; the main exceptions are EWA purchases of water north of the Delta, which facilitate the delivery of project water to users in the south while protecting Delta fisheries. The Southern California region, defined broadly to include both the desert counties and the coast, generates most of the remaining quarter of total supply. The largest single source has been the Imperial Irrigation District, which has operated a 110,000 acre-feet per year long-term transfer with the Metropolitan Water District of Southern California (MWDSC) since 1988. In the mid-1990s, there was also a large twoyear transfer to the MWDSC by the Palo Verde Irrigation District, another agricultural contractor on the Colorado River Project. With the exception of the early 1990s drought, when several Bay Area cities made substantial purchases, no other region has played a significant role in the market on either the supply or demand side. The Bay Area’s share in demand has dropped from 9 percent of the total in the drought years to only 1 percent in the more recent period. This region’s share in supply is about 2 percent. Most Transfers Are Local or Regional We have already seen that the San Joaquin Valley is both a major supplier and a purchaser of water. The market in Southern California is also primarily regional in nature (Table 2.1). The only exports leaving the region are the transfers by municipal agencies to the SWP turn-back pool, purchased by San Joaquin Valley agricultural districts. In the Sacramento Valley, the only region in the state where water users purchase significantly less than the volume sold, exports are concentrated in dry years. In years with normal to high rainfall, two-thirds of the water is transferred to other water users within the region. Another way of seeing the shares of local and regional markets is to look directly at the source of water obtained by users. Figure 2.4 shows 20 Percentage 100 90 Seller in other region 80 (through bank or pool) Seller in other region 70 (direct) 60 Seller in same region Seller in same county 50 40 30 20 10 0 1990–1994 1995–2001 (525,000 acre-feet/year) (720,000 acre-feet/year) Figure 2.4—Nonenvironmental Water Purchases, by Location of Selling Party this breakdown, for nonenvironmental water only, according to the location of the selling party in relation to the buyer. Nearly one-quarter of total volume is purchased from parties in the same county and another half from parties in the region. Interregional transfers account for the remaining 25–30 percent of the market.6 Only a relatively small fraction of these transfers (one-fifth) are negotiated directly between parties in different regions; the lion’s share moves through banks and pools run by DWR. This pattern highlights again the important role played by the government agencies in California’s water market. Given the need for federal and state agency approval for the use of conveyance facilities, this role is indispensable for any interregional transfers of water physically moving across the Delta. Approval decisions are complicated because pumping additional water through the Delta can affect water quality standards and put protected fish species in danger. When an agency acts as an intermediary, it is able to facilitate the movement of water across the Delta. Agency input has also been crucial ____________ 6Because this graph presents data from the standpoint of end-user purchases, it includes only the water bank volumes that were actually resold in the first period, not the full amounts acquired by DWR. 21 in successful district-to-district transfers—for instance, the 2001 transfer of 160,000 acre-feet from Sacramento Valley CVP contractors to the Westlands Water District. Many observers believe that the absence of state or federal agency sponsorship significantly complicates the approval process for cross-Delta transfers. From Farms to Cities: A Key Element of LongTerm and Permanent Transfers Our records show 15 approved long-term transfers and 14 permanent transfers of surface water rights or entitlements from 1985 to 2002.7 At least ten additional transfers were pending approval in late 2002. As noted, long-term transfers have generally accounted for about one-fifth of all trades since the late 1980s. Volumes traded surpassed the 250,000 acre-feet mark for the first time in 2001. Contract duration runs from a low of two years to a high of 35, with an average of 15 years. The permanent transfers, bunched at the end of the decade, total another 175,000 acre-feet. These mainly concern the transfer of SWP contract entitlements under the Monterey Agreement (representing over 110,000 acre-feet) and the transfer of pre-1914 water rights among parties within Kern County. Almost all these transfers involve shifts of water from agricultural to urban uses. The handful of exceptions includes one long-term deferred exchange and one long-term transfer between districts with a large municipal customer base (Solano County Water Agency and the Mojave Water Agency; Placer County Water Agency and Northridge Water District, respectively) and several long-term transfers from agricultural districts to the environment (most notably, as part of the negotiated agreement for the restoration of the San Joaquin River). Only two longterm transfers between agricultural districts appear, one in the Sacramento Valley and one in San Joaquin, and two permanent transfers of contract entitlement among agricultural users (both in the San Joaquin Valley). ____________ 7For details on the transactions discussed in this section, see Tables A.5 and A.6. 22 This trend holds for transfers pending approval in 2002. Key among these are two large long-term agriculture-urban transfers within the Colorado River Project (Imperial Irrigation District to San Diego, Palo Verde Irrigation District to MWDSC) and several permanent transfers of CVP contract entitlement from agricultural water districts to towns in the San Joaquin Valley. Thus, agricultural users remain the largest single source of demand for water in today’s market, but they conduct their purchases almost entirely through temporary arrangements. Also noteworthy is the highly local character of much of the longterm and permanent transfer market. Transfers involving CVP contractors and water agencies with their own surface supplies have essentially taken place in the neighborhood, between districts and municipalities in the same or adjacent counties. The only cross-regional movements of long-term or permanent water involve SWP contractors and Colorado River contractors.8 In both cases, the transfers came about mainly as part of global renegotiations of project operating rules. Under the Monterey Agreement, Kern County agriculturalists negotiated the sale of contract entitlement to municipal users in Southern California and the San Francisco Bay Area in exchange for greater certainty of supply for their remaining SWP entitlement. Contractors within the Colorado River group have been under intense pressure to set up transfers from the agricultural valleys of the desert counties to coastal municipal agencies as part of California’s obligation to reduce its use of Colorado River water under the 4.4 Plan referred to in Chapter 1. The stronger presence of municipal agencies in the market for longterm and permanent water contracts is logical, given their need to ensure reliability of supplies for growing populations. Looking ahead, we should expect residential demand to increase, as a consequence of demographic trends themselves and recent legislation that makes a tighter connection between water supplies and land-use planning. In 2001, the legislature passed the “show me the water” bills, SB 610 (Costa) and ____________ 8Although technically within the same region (Southern California), most of the Colorado River transfers involve buyers and sellers at quite some distance from one another. The one deal involving close neighbors is the pending transfer from Imperial Irrigation District to Coachella Valley Water District. 23 SB 221 (Kuehl), which require that local governments demonstrate the adequacy of water supplies for growth.9 Long-term transfers are among the measures considered adequate for this purpose. Summing Up Jumpstarted by a prolonged drought in the late 1980s and early 1990s, California’s water market has now become a firmly established feature of the state’s water allocation process. The market remains largely intraregional in nature, with the state’s Department of Water Resources directly mediating most transfers across regions. The market is also highly segmented, with over half of the volumes traded among contractors of the large state and federal water projects and another third involving direct purchases by state and federal agencies for drought relief and environmental mitigation. Outside drought years, when urban agencies have been important buyers, the main sources of demand have been directly and indirectly linked to environmental concerns. Direct purchases for instream uses and wildlife reserves have accounted for over one-third of the increase in purchases since 1995. The other growth sector, accounting for over half of market expansion, has been agriculture in the San Joaquin Valley, as farmers whose contractual water deliveries have been cut because of environmental mitigation programs have turned to the market for replacement water. Municipal agencies are the principal buyers of long-term and permanent contracts. In light of the state’s rapid population growth, it is not surprising that municipal agencies are taking the lead in negotiating long-term and permanent arrangements for water supply. Legislation passed in 2001 requiring that local governments demonstrate adequate water supplies for development should increase urban demand for longterm water transfers. As we shall see, municipalities’ success in forging these deals and assuring new supplies will depend on their ability to smooth the waters of community resistance in the source regions. ____________ 9See Association of California Water Agencies (2002) and Department of Water Resources (2002b). 24 3. The Rise of Local Restrictions on Water Marketing As state and federal authorities have taken steps to facilitate water marketing in California over the past two decades, concerns over potential negative effects of the market on the source regions have prompted many county governments to erect new barriers to trade. This chapter describes the statewide trends in county adoption of groundwater protection ordinances and raises some of the key legal, economic, and operational questions concerning their implementation. The Mobilization of Rural Counties By the end of 2002, 22 of California’s 58 counties had adopted ordinances that restrict the export of groundwater. Although the specific language of the ordinances varies, one common thread is their focus on the regulation of exports, as distinct from groundwater uses on-site. In most ordinances, “exports” are defined as shipments of water beyond the county’s administrative boundaries. Although several counties apply instead an “out-of-basin” definition of exports, and several others an “offparcel” definition, a review of the implementation record suggests that these nonadministrative boundaries reflect an intent to protect the ordinance against potential legal challenges (discussed below) rather than to regulate groundwater use within the county.1 The precursor to this movement was the adoption by three northern counties (Butte, Glenn, and Sierra) of urgency ordinances prohibiting ____________ 1Counties with “out-of-basin” restrictions include Inyo, Kern, Mono, and Siskiyou. Counties with “off-parcel” restrictions include Tehama, San Benito, and Sierra. Imperial County’s ordinance has separate restrictions on exports leaving the county and on groundwater extractions for within-county use. As discussed in Chapter 5, only three counties within this group—Imperial, San Benito, and Sierra—have enforced a permitting process for within-county uses. For details, see Hanak and Dyckman (2003). 25 the “mining” of groundwater in 1977, a year of severe drought. Modoc County followed suit early in the following year, with an ordinance limiting transfers outside the groundwater basin. Over the next 15 years, a handful of Sacramento Valley and mountain counties introduced ordinances with explicit export restrictions. The slow pace may be explained in part by the fact that two counties, Inyo and Nevada, saw their ordinances successfully challenged at the trial court level during the 1980s.2 The floodgates opened once a third county, Tehama, won an appellate court victory in 1994, upholding its authority to regulate groundwater. Since the Tehama decision, which was widely publicized in water law and county government circles,3 14 counties adopted explicit export restrictions for the first time, and three counties regularized urgency ordinances adopted earlier (Figure 3.1).4 Geographically, the group is concentrated in rural California: the mountain counties to the north and east, the Sacramento and San Joaquin Valleys, and Imperial County to the south (Figure 3.2). By and large, these counties fall into the group traditionally considered “source” regions for the state’s water supply; many have relied heavily on groundwater for agriculture. Over this period, five other counties adopted groundwater protection ordinances that focus on management of groundwater resources within the county or in a particular geographic subarea.5 The regulations include various types of restrictions on extraction for on-site use (e.g., well permitting, flow monitoring, pump taxes). In effect, the county ____________ 2Inyo County’s ordinance was successfully challenged by the City of Los Angeles in 1983 (City of Los Angeles, Department of Water and Power v. County of Inyo, case no. 12,908, July 8, 1983). Nevada County’s ordinance, introduced in 1986, was successfully challenged in a suit brought by the Truckee-Donner Public Utility District in 1988. Inyo pursued the matter through negotiations with the City of Los Angeles and reintroduced a revised ordinance in 1998. Nevada County did not seek review and has not reintroduced an ordinance subsequent to the Tehama holding. For this reason, we have not counted Nevada among the 22 counties with export restrictions. 3See Goldsmith (1995a, 1995b) and Bunn (1997). 4In addition, Modoc County introduced a more stringent ordinance restricting exports out of the county rather than the basin in 2000, and Inyo reintroduced an ordinance in 1998 to replace the one invalidated by its court proceedings in the early 1980s. 5For details, see Table B.1, middle column. 26 Number of ordinances per year 5 Export restrictions 4 Tehama County court victory 3 2 1 0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 NOTES: For details, see Table B.1. The dates refer to the adoption of regular ordinances restricting exports. Figure 3.1—Adoption of County Ordinances Restricting Exports assumes an authority resembling that exercised by other specialized groundwater institutions present in the state: basins adjudicated by the courts, special groundwater management districts created through acts of the legislature, and certain special water districts that exercise full control over access to the resource.6 In adjudicated basins, use rights are attributed to individual users in much the same way as surface water rights. In the special districts with groundwater authority, the districts have the authority to regulate individual water use, typically through pump taxes. Counties with these types of groundwater management institutions tend to be located along the coast and in Southern California; most have significant urban populations. The on-site groundwater protection ordinances appear to substitute for or complement the activities of special districts and adjudicated basins. Ordinances substituting for other measures include those of San Diego and Napa Counties; in Napa, ____________ 6For maps showing the location of these institutions, see Figures B.2 and B.3. 27 Export restrictions NOTE: Kern County’s ordinance is limited to the southeast portion of the county within the Lahontan hydrologic region. Figure 3.2—California Counties with Groundwater Export Restrictions, 2002 officials view the ordinance as an alternative to a costly adjudication process. Counties where the ordinances complement other institutions include Monterey, Mendocino, and San Bernardino. In Monterey, the ordinance applies to several zones not already covered by a special groundwater district. Mendocino’s ordinance applies only to the town of 28 Mendocino, where a special district has groundwater authority. The San Bernardino ordinance applies to desert regions of the county not already under a local management system and notably excludes the Mojave Basin, which is adjudicated. At first glance, several of the mountain counties (Lassen, Modoc, Mono, and Sierra) seem to be exceptions to this geographical division because they have adopted both special groundwater management districts and countywide ordinances restricting exports. However, all but one of the six special groundwater districts in these counties were set up with the primary purpose of controlling exports rather than for local use management.7 Siskiyou County also appears as a partial exception, by virtue of the presence of one adjudicated basin. As noted above, several of the counties with export restrictions have ordinances that would, in principle, provide authority to regulate in-county uses as well. However, this authority appears to be exercised actively in only three counties— Imperial, San Benito, and Sierra. An Overview of Export Restrictions All 22 ordinances restrict the direct export of groundwater; those adopted since 1996 also implicitly restrict exports of surface water by regulating the extraction of groundwater used to replace exported surface supplies. Sacramento’s ordinance overtly restricts surface water exports of any kind. A handful of counties (Madera, San Joaquin, and Fresno) have specific restrictions on the use of county groundwater basins as storage sites for groundwater banking projects. San Joaquin County supervisors recently introduced a separate ordinance restricting the location of aboveground storage as well, in response to a proposed project to convert two islands in the Delta into surface storage areas. In most counties, restricted activities require a county permit, which invokes a review under the California Environmental Quality Act (CEQA).8 Permit applicants are expected to conduct from one to several ____________ 7The exception is the Sierra Valley Groundwater Management District, which was initially intended to regulate agricultural pumping in response to drought conditions. 8In four of the five counties with ordinances oriented toward local groundwater management (San Diego, Mendocino, Monterey, and Napa), the process involves 29 studies. Applications go through a multilayered review by county departments and commissions, with the final decision most often in the hands of a political body (the board of supervisors or a body appointed by the board). If approved, permits generally run from one to three years. Most ordinances provide certain categorical and conditional exemptions to the permitting process. Water districts or landowners with service areas or holdings overlying adjacent counties typically do not need permits for water use on those lands. Most counties also provide a blanket exemption to permitting as long as quantities remain within historical use levels. A number of the mountain counties exempt bottlers of spring water, as long as the bottling is done within the county. Finally, various counties exempt specific types of local entities—such as incorporated cities and water districts—from permitting altogether, either because the counties have assurances that the entities already engage in sound groundwater management practices or because there are questions of regulatory authority and an interest in avoiding jurisdictional conflicts. Legal Issues The question of jurisdictional authority is at the heart of the legal issues raised by the county ordinances.9 In California, municipalities may invoke police powers to protect the public welfare in areas not regulated by the state. This is the basis for the groundwater protection ordinances, because groundwater falls outside state jurisdiction. The lawsuit filed in 1992 against Tehama County by two landowners, Baldwin and Myers, challenged this position, arguing that some provisions of the California Water Code already dealt with local groundwater protection, thereby “preempting the field.” Three areas of state intervention were cited: the specially enacted groundwater management districts (noted above); Section 1220 of the Water Code ________________________________________________________ incorporation of a groundwater review or overlay in a regular ministerial process, rather than application for a discretionary permit with CEQA review. 9For a detailed discussion of the legal issues raised by the ordinances, see Hanak and Dyckman (2003). 30 (enacted in 1984), which restricts the direct export of groundwater out of the Sacramento River Basin; and the recently enacted AB 3030 (Water Code Sections 10750–10753.9), which authorizes existing water agencies to create groundwater management districts. The county lost in trial court but prevailed at the appellate level, with a holding that the cited state legislation did not preempt the county’s ability to regulate an arguably open field by establishing a review process in the interest of public health and safety.10 The Tehama case was the first appellate decision to address the issue of county authority to regulate groundwater extraction. As noted, it followed at least two earlier Superior Court rulings that took the opposite position, holding that the state had preempted the power of cities and counties in this domain. Although there have been no further proceedings against county groundwater ordinances, several potential legal concerns were not addressed by the Baldwin v. County of Tehama case. The first of these relates to provisions of the Water Code governing surface water transfers, where it would be difficult to argue that the state has not occupied the field. Sacramento’s ordinance, which openly restricts any surface water exports, would suffer under such a legal challenge. County ordinances that indirectly restrict surface water exports by restricting groundwater substitution may also conflict with the state’s authority in this area. Some ordinances may also be open to legal challenge on the grounds that they exceed the county’s police power. Court decisions have tended to take the view that governments, in exercising this power, should take the minimum steps needed to protect public health, safety, and welfare. In counties where it is possible to demonstrate that there is already a significant problem of overdraft, imposing export restrictions is a classic use of police power, as it protects residents’ ability to exercise their property rights. However, a number of counties are not in a position to justify the export restrictions on the basis of current needs. The widespread exemptions for historic use levels are, in effect, an admission that there is ____________ 10Baldwin v. County of Tehama, 31 Cal. App. 4th 166, 173-74 (1994, 3rd Dist.); review denied, Cal. Sup. Ct., March 17, 1995. 31 no current threat to public health, welfare, and safety. In such cases, applying restrictions could amount to barring present use somewhere in the state to preserve future use in the areas of origin. In addition to potentially exceeding the extent of police powers, an overly protective ordinance may violate Article X, Section 2 of the California Constitution, which requires that “the water resources of the state be put to beneficial use to the fullest extent of which they are capable.”11 Many counties are in the process of incorporating the ordinances into their General Plans as a way of shoring up the basis for the restrictions. In a different vein, there is some debate over whether the ordinances could be subject to a federal Commerce Clause challenge. As noted above, the language in most of the ordinances is jurisdictionally based, restricting exports beyond the administrative boundaries of counties rather than hydrologic basins or some other distinction that reflects the physical links between groundwater extraction and harm to adjacent users. This language raises the potential for a challenge of discrimination based on arbitrary distinctions.12 Although this issue has not been raised in any of the court challenges to date, it has been of some concern for ordinance drafters in several counties, as reflected in the move to an “outof-basin” or “off-parcel” permitting system.13 A final, and arguably more significant, legal concern is that the ordinances generally do not distinguish between native groundwater and imported surface water banked underground. As a result, even ____________ 11In the Tehama case, the 4th Appellate Court rejected this argument, observing in a footnote that the issue “is not so simple as plaintiff’s cursory argument supposes.” Counties may nevertheless have reason for some concern over a potential future legal challenge against “hoarding water.” Note too that the source of county regulatory authority—police power—relates to current threats and not anticipated future threats. 12The case law precedent is Sporhase v. Nebraska ex rel. Douglas (1982), in which the U.S. Supreme Court held that discriminatory groundwater export regulation interferes with interstate commerce (Getches, 1997). 13Counties that moved from a county to a basin restriction include Inyo, Mono, and Siskiyou. Discussions with officials in Inyo and Mono revealed that compliance with the Commerce Clause was a factor in this decision. Kern County drafted its ordinance with basin restrictions for similar reasons, following discussions with Inyo County officials. Constitutionality issues were also a factor in the drafting of the Tehama County ordinance, which restricts use off-parcel rather than out-of-county. 32 ordinances that do not directly address groundwater-banking projects do so implicitly because the county might claim authority to restrict the reexport of water brought in for temporary storage. The legal ambiguity on this question arises from overlapping and potentially conflicting jurisdictions, as the state governs the surface water brought into the county, whereas counties may seek to govern its reexport. The Water Transfer Workgroup convened by the SWRCB in 2000 recommended that county authority over imported banked water be limited to preventing injury to local users, which could arise if the banking project negatively affected either quality or quantity of water available locally.14 Economic Issues The legal basis for county oversight of groundwater reflects the underlying economic justification for regulation of a collective resource. In an unregulated situation, a collective resource such as groundwater risks mismanagement. In general, individual users or water districts will not have the incentive to prevent overuse, with negative consequences for both quantity and quality of the water available in the basin. In the absence of state regulation, there is a strong case to be made for local oversight mechanisms that encourage sound management. Against this background, California’s county groundwater protection ordinances raise two questions from an economic standpoint. The first relates to the scope of the ordinances themselves; the second to whether counties are the appropriate level of local jurisdiction. Concerning scope, the question is whether ordinances whose sole focus is restricting exports can contribute to better management of the groundwater resource. Export restrictions could be a sensible management tool, from an economic standpoint, in two situations. First, when there is uncertainty about the characteristics of the groundwater basin (size of the aquifer, recharge rate, etc.), restricting exports could be a reasonable precautionary measure. Second, in the absence of consensus on mechanisms to manage groundwater use by ____________ 14See the discussion of this issue in the report of the Water Transfer Workgroup (2002). 33 those within the basin, restricting exports at least ensures some upper bound on extraction rates. Even where this precautionary strategy is a sensible first step in groundwater protection, however, it can quickly reach it limits as a management tool. A policy focusing on export restrictions can discourage active management of groundwater basins, a practice that can benefit local water users financially and enhance the regularity of supply in overdrafted basins. Active management can include various programs involving the conjunctive use of groundwater and surface water, such as the intentional recharge of aquifers that have been drawn down and the intentional drawdown of aquifers that are full.15 In most places, active management programs require interactions with water users outside the area—notably through transfers of surface water in lieu of groundwater and through banking of imported surface water. Prerequisites include setting up information systems to better understand the aquifer and establishing some form of local oversight on groundwater use. All of this implies moving beyond export restrictions to a more comprehensive system of groundwater management. In areas where the conjunctive use of surface and groundwater is not feasible because of limited surface supplies, the case for moving beyond export restrictions to active management is admittedly weaker. Even here, however, there can be benefits to understanding the limits and opportunities of the resource base. When groundwater supplies and recharge rates are more than adequate, a well-structured export program could be a boon to the local economy. Are counties the appropriate level of jurisdiction to provide this leadership? The economic literature on the management of collective resources shows that success is greatest when local oversight institutions reflect a high degree of alignment of interests among concerned parties (Ostrom, 1990). The alignment can be geographical, increasing in the extent to which the physical boundaries of the resource management problem coincide with the jurisdiction of the local governing institution. The alignment can also be relational, increasing with the capacity of ____________ 15For a description of conjunctive use programs used in California, see Purkey et al. (1998). 34 concerned parties to participate in or affect the deliberations of the governing institution. In the case of groundwater protection and management, a natural point of geographical organization might be the aquifer—or the larger watershed draining into it. In all but a few cases, these physical boundaries correspond neither to those of the county nor to those of other local institutions, such as water districts. Recognition of this “misalignment” has led to the formation of multiparty institutions to address groundwater and watershed issues in many parts of the state. For groundwater, the most common institutional forms are groundwater management programs created under the AB 3030 legislation noted earlier or under joint powers authority. According to Department of Water Resources records, roughly 20 multiparty programs of this type currently exist. As we will discuss below, some of these programs show promise as a structure for local resource management. Creating new institutions is not without difficulties, however, and there is some merit to considering counties as a potential rallying point for local resource management, even if their administrative borders do not coincide with the limits of the aquifer or watershed. Counties have the merit of having well-established representative institutions and public consultation mechanisms and can provide a convening point for parties. As such, they offer the potential for a high degree of relational alignment. Together with cities, they are also the only local institutions with the authority to invoke police powers, which could be a useful component of a local resource management program. Operational Issues The extent to which the ordinances can play a beneficial role in local water management also depends on the way certain operational issues are handled. Foremost among these is the review process for permitting. In many ordinances, this process is open-ended and involves significant upfront costs. A strong case can be made that those wishing to transfer water should be responsible for conducting the necessary environmental reviews (as is the practice for surface water transfer projects at the state level). In a number of counties, however, the initial requirements appear disproportionately high—effectively calling for a full-fledged study of the 35 aquifer before allowing any transfer project to go through, no matter how small. Whether intentional or not, this requirement can deter those seeking permits. It raises a potential legal issue of undue burden because the first applicant bears the cost of the background studies for all those who follow. In many counties, questions can also be raised about the transparency of the review process and the extent to which the technical characteristics of the project will be considered by the political body empowered to grant permits. Finally, the number of reviewing entities and effective amount of review time create other deterrents. Summing Up In the 25 years since the state adopted a policy in favor of water transfers, a movement to regulate the water market has gained momentum in California’s rural counties. As of late 2002, 22 of the state’s 58 counties had adopted groundwater protection ordinances requiring a permit to export groundwater or to extract groundwater used in substitution for transferred surface water. Although this policy can be justified on economic grounds as a firststep precautionary measure in the face of uncertainty about the resource base, it is harder to justify in the longer run in the absence of a broader water management program. A strictly precautionary policy prevents the water users in the county from actively managing their groundwater resources, a practice that can reap financial and water supply benefits. It also makes counties susceptible to charges of “hoarding water,” which is disallowed by the California Constitution. Because groundwater is a shared resource, active water management requires some form of local oversight. Whether counties or other local institutions are better suited to the task is an open question to which we will return below. 36 4. Why Do Some Counties Adopt Export Restrictions? As we have seen, counties with ordinances restricting water exports are concentrated in California’s inland rural regions—the mountain counties to the north and east and the Sacramento and San Joaquin Valleys in the center. Despite this common denominator, however, there are some clear distinctions within the group. Some counties are agricultural powerhouses; in others, farming is a marginal activity. Some are sparsely populated; others contain large and fast-growing metropolitan areas. Nor is geography a complete determinant: Over a third of the counties in these regions have not adopted ordinances. This diversity suggests that the rural county ordinance movement is not monolithic; a range of factors may influence individual counties’ choices. This chapter explores some of these factors, using two approaches. First, we examine whether there is a statistical basis for predicting which counties adopt ordinances, taking into account characteristics about county economic and water conditions and institutional factors. Second, we use a more qualitative lens, drawing on information obtained from interviews with county officials and water users. This discussion highlights the specific dynamics at play in each of the three regions noted above. Factors That Make a Difference Because the timeline of county adoption is idiosyncratic, we will confine the statistical analysis to a simple “yes” or “no” prediction of ordinance adoption and not attempt to model the factors determining the year of adoption. This leads us to focus on characteristics that vary across counties, measured, when possible, in the mid-1990s, when the adoption movement was fully under way. 37 Data on County Water Economy and Institutions We could expect two structural factors to be important in the decision to adopt water export restrictions: the local economy’s dependency on water and the extent to which exports might compromise the groundwater resource base. To measure the first of these, we have identified four indicators. Two capture the role of agriculture, the primary water-dependent economic activity: the share of farm jobs in total employment and the share of agriculturally related jobs, a category including agricultural services and agro-processing.1 A third indicator, the share of irrigated acreage in total farm acreage, reflects the intensity of water use as an input. Counties with higher proportions of dry-land farming or rangeland will have lower agricultural water needs. The fourth indicator, the share of residential population dependent on groundwater, captures the relative importance of groundwater as a water source. Ideally, we would have included a comparable measure for agriculture, but the share of groundwater in farming is not known in many counties. We might expect all four of these measures to be positively related to the decision to adopt an ordinance restricting water exports. We might also expect counties whose groundwater basins are experiencing serious overdraft to be more inclined to adopt precautionary measures than those whose aquifers are full. Identifying these locations proved difficult. A good measure of overdraft would be the extent to which the water table in an aquifer was declining on a long-term basis. There are no statewide measures of overdraft so defined, in part because the Department of Water Resources does not have access to well data in many areas, in part (and relatedly) because the issue raises thorny legal questions. Designation of overdraft has implications for the rights of groundwater appropriators and can serve as a basis for adjudication proceedings. The last time DWR ventured an official designation of groundwater basins in difficulty was in the 1980 Bulletin 118-80 (Department of Water Resources, 1980). The bulletin provides a list, established ____________ 1See Appendix C for a description of data sources. 38 through a process of data analysis and public review, of 11 basins that are “critically overdrafted” and four basins “with special problems.”2 Although conditions of individual groundwater basins in the state may have changed for better and for worse in the intervening years, this list remains the best indicator of counties with an at-risk groundwater basin. We would expect that counties overlying such basins would be more likely to have adopted some type of restrictions. It is also likely that institutions play a role in the decision to restrict exports. One unquantifiable but integral aspect is the degree of harmony among the county’s institutions. Counties where the water agencies are not in conflict with each other and with the municipal and county governments will be less likely, all else equal, to have the motivation to impose export restrictions as a controlling mechanism. This institutional concordance can result from a range of factors: limited number or low diversity of institutions, overlap of county and water agency governing bodies, or good working relationships borne of individual initiative. Another important institutional factor is an affiliation with the Regional Council of Rural Counties (RCRC), an association of counties with populations generally below 300,000. It began in the early 1970s as an association of the mountain counties and subsequently expanded to cover most of the small counties in the state.3 In 1995, on the heels of the Tehama County court victory, RCRC established a water program to promote the protection of member counties’ water rights. The program encourages members to adopt groundwater ordinances and to reflect these in their general plans. RCRC influence is transmitted in two ways. For some counties, the secretariat has provided technical and legal advice; more generally, the board of directors, made up of a supervisor from each member county, serves as an important forum for information exchange. ____________ 2See Figure B.4 for a map showing the location of these basins. 3As of 2002, RCRC had 30 member counties: Alpine, Amador, Butte, Calaveras, Colusa, Del Norte, El Dorado, Glenn, Imperial, Inyo, Lake, Lassen, Madera, Mariposa, Merced, Modoc, Mono, Napa, Nevada, Placer, Plumas, San Benito, Shasta, Sierra, Siskiyou, Sutter, Tehama, Trinity, Tuolumne, and Yuba. Imperial County officially joined in October 2002 but was informally affiliated with the group for some time before membership. 39 Cross-County Results For several of these indicators, there is a considerable and statistically significant difference in the mean values between counties with export ordinances and those without (Table 4.1). On average, counties with export restrictions have twice the share of farm employment and 40 percent higher residential dependence on groundwater. They are also twice as likely to be members of RCRC. The ordinance adopters also register a third more irrigated farmland and appear nearly twice as likely to overlie a critical or specially designated groundwater basin, although the differences are not statistically significant at the 90 percent level of confidence. The one area where the groups do appear alike is in their share of agriculturally related employment, at roughly 2 percent of all jobs. Several of these factors contribute to the probability of a county adopting an export restriction in the expected ways.4 The most sizable effects are associated with membership in RCRC and the presence of an at-risk groundwater basin. These factors increase the likelihood of ordinance adoption by 30 and 26 percent, respectively. Farm employment also raises the likelihood of adoption. A county with 10 Table 4.1 Average Characteristics of Counties With and Without Export Restrictions Counties With Restrictions Farm employment (%) 8.0*** Agriculture-related employment (%) 2.0 Irrigated farmland (%) 34 Residents using groundwater (%) 75*** Counties overlying critical/special basin (1980) (% of counties in group) 45 RCRC membership (% of counties in group) 77*** Sample size 22 Counties Without Restrictions 3.8 1.9 24 54 All Counties 5.4 1.9 27 62 25 33 36 52 36 58 ***Indicates significant difference of group means at the 99 percent level of confidence. ____________ 4See Appendix C for a more detailed presentation of the statistical results discussed here. 40 percent of its workforce in agriculture is a third more likely to adopt an ordinance than a county with only 2 percent of farm jobs.5 Somewhat surprisingly, agriculturally related employment appears to have the opposite effect. For a given level of on-farm employment, counties with a higher share of value-added activities related to agriculture are less likely to adopt export restrictions. In effect, the geographical link between agriculturally related jobs and farming is less direct than is commonly thought. Many counties with a relatively low share of farming have significant processing activities. The statewide county average ratio of off-farm to on-farm agricultural jobs is 73 percent. Only one county with an export ordinance (Sacramento) exceeds that level. Thirteen counties without ordinances do so— including virtually all of the San Francisco Bay Area and Southern California. Many of these counties have more comprehensive groundwater management systems, which do not discriminate against transfers. Regional Issues To understand the reasons for adopting export restrictions, we interviewed county and water district officials and other resource persons in 36 counties, including most having some type of groundwater ordinance and the key agricultural counties without one.6 The interviews sought background on the following types of questions: Was the ordinance a response to a specific local event or a general precautionary measure? Is the main concern with activities of private individuals or water districts? Which types of prospective buyers raise red flags—farmers in neighboring counties or distant municipal water suppliers? Is the objective mainly to control water exports or groundwater banking projects? Within the county, how contentious was ____________ 5Neither the share of irrigated farmland nor the share of residential dependence on groundwater has statistically significant effects on the adoption of export restrictions. 6The only counties with ordinances for which interviews could not be conducted are Siskiyou and Monterey. Counties without ordinances for which interviews were conducted include Kings, Merced, Placer, Plumas, Solano, Sonoma, Stanislaus, Sutter, Trinity, Tulare, and Yuba. All 58 counties were contacted to verify whether an ordinance was in place or under consideration. 41 the adoption process? Were exemptions granted because of jurisdictional issues or because the exempted parties were considered to have an adequate groundwater management system in place? Although each county has a unique story in some respects, we have opted to highlight the results of these interviews from a regional perspective.7 The three regions singled out have different water supply conditions and a different set of demands for agricultural and residential uses. These factors lend a specific regional character to the nature of the perceived threat the water market brings with it. So, too, do historical events relating to water transfers. Within this context, the weight of the ordinance in a given county will depend on the local dynamics of the adoption process: whether it came in response to a specific event, the degree of internal conflict, and the extent of public involvement preceding adoption. The Mountain Counties: The Legacy of Owens Valley Counties in this region are sparsely populated and have limited agricultural production (Table 4.2).8 Many rely heavily—some exclusively—on groundwater, although rivers and streams provide a source of surface water in others. An indicator of the extent to which groundwater is a local concern is the fact that six of the 15 counties overlie basins identified in Bulletin 118-80 as having “special problems.”9 The history of one of these basins, Owens Valley, has indelibly marked the regional consciousness on water issues. In the early part of the 20th century, the City of Los Angeles bought up vast tracts of land in the valley and began exporting large quantities of groundwater. The transfer had immediate consequences for the local economy, where agriculture became unviable, and has proven over time to be deleterious to the environmental health of the valley. The expansion of pumping capacity in 1970 with the addition of a second aqueduct exacerbated the ____________ 7For a list of counties by region and a map showing county locations, see Table B.1 and Figure B.1, respectively. 8The only county with over $100 million in gross agricultural product in 1999 was Siskiyou, which is part of the Klamath Project. Neighboring Modoc, also in the project, had output valued at just over $60 million in that year. 9Only Calaveras overlies a basin identified as subject to critical overdraft. 42 Table 4.2 Regional Characteristics (County Averages) Population Urban population (%) Agricultural output per capita ($) Residents using groundwater (%) Counties overlying critical/special basin RCRC membership Export ordinances Counties in group Mountain Counties 37,870 0 882 67 7 15 8 15 Sacramento Valley 234,335 (118,161) 41 (35) 3,530 78 San Joaquin Valley 430,075 39 4,366 88 08 82 74 10 8 SOURCES: Population figures for 2002 are from the California Department of Finance. Urban population share is from the 1990 Census, the most recent year for which this estimate is available. Per capita agricultural output is from the 1997 Agricultural Census. See also Appendix C. NOTES: Numbers are county average values for each region. For a list of counties in each region, see Table B.1. For the Sacramento Valley, numbers in parentheses indicate values excluding Sacramento County. environmental problems, both in Mono Lake and in the Owens Lake bed. Following years of difficult and acrimonious legal proceedings, Inyo County and a coalition of environmental groups reached mitigation agreements with the City of Los Angeles in the early 1990s. These have resulted in a considerable reduction in the flow of water out of Mono and Inyo Counties through the Los Angeles Aqueduct.10 The mountain counties also have an acute awareness of their status as source regions for the federal and state water projects and the projects developed by the City of San Francisco and East Bay Municipal Utilities District. Although these projects have not necessarily affected the volumes of water available to local water users, whose needs are limited, ____________ 10The settlement between the City of Los Angeles and the County of Inyo over Owens Valley was reached in October 1991, and the Mono Lake settlement was reached in September 1994 (Hundley, 2001). Together, these agreements have reduced the City of Los Angeles’s exports by an annual average level of 150,000 acre-feet, or one-third of the conveyance capacity of the Los Angeles Aqueduct (personal communication, Jerry Gewe, Los Angeles Department of Water and Power, December 2002). 43 there is a tradition of vigilance in defending the area of origin status, by which counties can reclaim water they may need for future growth.11 Moreover, contention over the regulation of instream flows can at times be significant. One case in point is the current dispute over Central Valley Project appropriations from the Trinity River. The federal government’s decision to reduce off-take, in response to concerns by Native American tribes and environmentalists, has become the subject of legal proceedings with project contractors in the Sacramento and San Joaquin Valleys. A second is the dispute on the allocation of water between instream and agricultural uses on the Klamath River, which has pitted agricultural water districts in Siskiyou, Modoc, and neighboring Oregon counties against environmental and tribal advocates within the same region. Against this backdrop, one might expect these communities to be wary of transferring water. Several of the basins bordering the state of Nevada have been the subjects of intense controversy for just this reason. A project to market groundwater to Nevada from Long Valley, one of the “special problem” basins underlying Lassen and Sierra Counties and neighboring Washoe County in Nevada, prompted the introduction of the state’s first special groundwater management district act in 1980. Since then, potential export projects to Nevada from the Honey Lake Basin (Lassen) and the Surprise Valley Basin (another “special problem” basin underlying Modoc) led county officials to obtain special district status for these basins as well. In addition to the ability to limit exports, the special groundwater management districts have the authority to negotiate directly with the state of Nevada concerning joint basin matters.12 County-level export restrictions have been adopted as an additional safeguard, with little substantive consequence for the management of these basins. ____________ 11This includes the “County-of-Origin” statute (1927; Cal. Water Code Section 10505), the “Watershed Protection Act” (1933, Cal Water Code Sections 11460– 11464), the “Delta Protection Act” (1959), and the “Protected Areas” statute (1984; Cal Water Code Sections 1215–1222). See Hundley (2001), pp. 531–533. 12This authority has been essential, but the border counties nevertheless find themselves at a disadvantage in cross-border basin management negotiations, because groundwater extraction in Nevada is regulated at the state level. 44 Although the potential for groundwater exports to Nevada has also been an issue in Mono and Inyo Counties (as well as San Bernardino, further south), the more-pressing concerns there have arisen from modern-day projects to export groundwater to Los Angeles. Inyo County’s initial ordinance, passed by a ballot referendum in 1980, was part of the county’s attempt to seek mitigation for the ongoing transfers to the City of Los Angeles begun decades earlier. But mitigation of both Mono Lake and the Owens Valley has spawned a new set of concerns. The freed-up capacity in the Los Angeles Aqueduct has sparked the interest of several private firms, which see the potential to use it for conveying water to the coast. The Inyo County ordinance was revised in 1998 to expressly limit transfers through the aqueduct and any sales to Los Angeles. By contrast, in Calaveras and Tuolumne Counties, ordinance adoption appears to have been largely preventive in nature, sparked more by the Tehama ruling and RCRC encouragement than by any specific threat to the groundwater resources. On the whole, the introduction of ordinances in this region has been spearheaded by county officials and has proceeded with little local dissent. The one exception is Calaveras County, where there were protracted negotiations on the terms of the ordinance because of the concerns of local farmers and the main surface water district. The challenges to Inyo’s initial ordinance were raised by landowning “outsiders”—a private company, Anheuser-Busch, which hoped to export water to use in its plants in Los Angeles, and the City of Los Angeles itself, which sued the county and prevailed in the Superior Court.13 Subsequent versions of the ordinance exempt the City of Los Angeles from the permitting process, as its water use is regulated by a separate agreement with the county. Among the counties without ordinances, at least two have alternative institutional mechanisms in place for management. Plumas County overlies the Sierra Basin, for which a special groundwater management district was set up in 1980 to deal with in-basin supply problems, rather ____________ 13City of Los Angeles, Department of Water and Power v. County of Inyo, case no. 12,908, July 8, 1983. 45 than in response to export threats. One commentator assured us, “Were there a need to, the Board of Supervisors could vote in an ordinance in no time.” El Dorado’s only groundwater supplies come from fractured rock—a geological formation that does not lend itself to recharge in the way alluvial basins do. Well permitting has been strictly controlled in the county for many years. Sacramento Valley: A Balancing Act Between the Surface Water “Haves” and “Have-Nots” The Sacramento Valley is an important agricultural region, producing rice, tomatoes, and various fruits and nuts (Table 4.2). It also contains one of the fastest-growing metropolitan areas, centered on the city of Sacramento. Unlike the mountain counties, the valley is simultaneously a source region for the large surface water projects and a major surface water user. The region is also relatively rich in groundwater, which is a primary supply for residential uses and for some farmers. In 1980, not a single aquifer in the valley was on the critical basin list. Although several counties are now concerned about cones of depression, or pockets of overdraft, the valley would still be absent from that list if it were drawn up today. Debates on water marketing are colored by two characteristics of the region’s water supply: the uneven distribution of surface water rights within the valley and the valley’s overall abundance of supplies relative to other parts of the state. The distribution of surface water rights within the valley reflects the historical patterns of water district formation and construction of storage and conveyance facilities. A number of water districts and mutual water companies have senior rights to Central Valley Project and State Water Project deliveries, by virtue of having laid claim to the water before project inception. Most of these senior contractholders are agricultural water users in the south-central and eastern parts of the valley, in Glenn, Colusa, Butte, and Sutter Counties. To the east, Yuba and Placer Counties are also rich in surface water, thanks to autonomous projects with aboveground reservoirs. Districts along the west side of the valley, served by the Tehama-Colusa Canal, have junior rights to CVP water, typically involving both lower per acre allocations and less-reliable supply from one year to the next. In several counties, 46 independent groundwater pumpers—i.e., those not affiliated with water districts—constitute a significant portion of the farming population. “Area of origin” concerns—notably the perception that Southern California’s thirst for the north state’s water would never be quelled— are a longstanding component of the valley’s water lexicon. A landmark event was the 1982 defeat at the ballot box of the project to build the Peripheral Canal, which would have facilitated the movement of water past the San Francisco Bay Delta to southern contractors of the federal and state water projects (Hundley, 2001). Voters in this region (as elsewhere in Northern California) were overwhelmingly opposed to the project, which still looms as a symbol of the need for vigilance on water rights. Soon after this defeat, at the urging of valley and mountain counties, the legislature added Sections 1215-22 to the Water Code to firm up the area-of-origin protections. Under Section 1220, it is illegal to directly export native groundwater appropriated after 1975 outside the Sacramento River Basin without the authorization of the overlying county. To authorize, the county first has to establish a groundwater management plan to ensure that the export does not compromise supplies for local needs. Water purveyors south of the Delta were the intended targets of the section, because earlier area-of-origin statutes provided safeguards only with respect to the state and federal projects.14 It appears that the measure was largely preventive in nature, rather than a response to specific transfer projects under consideration. The multiyear drought of the late 1980s and early 1990s prompted renewed fears about the need to protect native groundwater. One source of the problem was actual transfer activity that occurred under the stateoperated drought water bank. Various water districts and individual farmers in the region participated in the bank in all three years of its operation (1991, 1992, and 1994). In 1994, bank purchases sparked a well-publicized controversy in Butte County, where water districts with senior rights sold some of their surface water and engaged in additional ____________ 14Personal communication, Clyde McDonald, September 2002. Mr. McDonald was a member of the staff of the bill’s author, Senator Norman Walters, at the time of passage. 47 groundwater pumping for irrigation. When wells in parts of the county ran dry, some independent pumpers linked this development to the transfer activity. Because there was no mechanism in place to monitor the effects of the pumping or to mitigate third-party effects, the problem festered and a heated conflict erupted among local water users.15 Although the Butte conflict may well have had repercussions beyond county lines, the state’s considerations about using the region’s groundwater as a longer-term source of supplemental supply for users south of the Delta were probably at least as significant in generating concerns about groundwater protection. In the wake of the water bank experience, the Department of Water Resources began development of a supplemental water purchase program, intended to be a more systematic approach to future water transfers. Initially, the program envisaged annual purchases of up to 400,000 acre-feet of groundwater from Sacramento Valley basins.16 Upon release of the draft document in 1996, local reaction was immediate and sufficiently spirited to condemn the proposal to mothballs. In our discussions with water users and officials throughout the region, the imprint of this proposal remains strong, although the details have faded into the haze. In explaining the reasons for county mobilization, a typical recollection runs something like this: “Back in the early 1990s, DWR decided to put a straw into the aquifer, so we knew we had to do something.” The other well-publicized groundwater transfer controversy, which provoked the drafting of the Tehama County ordinance, demonstrates that exports to points south of the Delta are not the only concern, however. This case involved the direct pumping of groundwater off a small parcel in proximity of the Tehama Colusa Canal, which the landowner planned to use to irrigate his farmland in Colusa County, farther south along the canal. Section 1220 of the Water Code would not be applicable in this type of situation, as it only limits the direct ____________ 15Notably, DWR did not reserve the right in its purchase contracts to require cessation of pumping in the event that effects were encountered. A detailed case study is provided in Thomas (2001). 16Department of Water Resources (1996b), as discussed in Thomas (2001). 48 exportation of groundwater out of the valley, not transfers within the region.17 Thus, Sacramento Valley counties have put ordinances in place to control two types of behavior, depending on the specific water supply issues they face. In several of the west side counties with limited surface water supplies—Tehama, Shasta, and Yolo—the primary target has been direct groundwater exports by private landholders located alongside conveyance facilities, whatever the ultimate destination. In a larger set of counties, the key issue is the dichotomy between water districts with senior surface water rights and other users who rely on groundwater. Concerns have focused on the potential for surface water districts to engage in indirect groundwater exports, by pumping more groundwater and selling their surface water. This problem has played out in quite different ways across the region. In Butte, the controversy went to the ballot box. A coalition of citizens’ groups angry with the water transfers sponsored an initiative to give the county fairly sweeping control over groundwater management, including the potential to introduce pump taxes. In reaction, the county farm bureau and local water districts proposed an alternative ballot measure, limiting the scope of county oversight to exports. It was this second measure, supported by greater campaign funding, that voters adopted in the fall 1996 elections. Although the air has cleared in the six years since its passage, the ordinance-by-initiative has left its mark on local water politics. At the opposite end of the spectrum, Yuba County has avoided the controversy over groundwater substitution altogether thanks to a particular set of institutional conditions. The Yuba County Water Agency (YCWA), the wholesale purveyor to most of the county’s agricultural water districts, has the entire county as its service area. All five county supervisors also serve as directors on the agency’s board. The agency has adopted a policy of immediate mitigation in the event of any ____________ 17The common interpretation has also been that Section 1220 does not apply to indirect groundwater exports via groundwater substitution, hence another argument for county ordinances. 49 well problems that could be linked to groundwater pumping for transfers. In the three other counties with significant potential for groundwater and surface water interaction—Colusa, Sutter, and Glenn—this high degree of overlap between county and water district jurisdictions does not exist. Ordinances were proposed as preventive measures to protect local users from harm. In Colusa, the county took the lead with RCRC support, and the ordinance appears to have been passed with little dissension by water districts. In Sutter, county officials proposed an ordinance as early as 1995 or 1996 but have held off on adopting it in response to water district opposition. Glenn is the only one of the three counties where, as in Butte, the ordinance adoption process reflected a serious disagreement between water users in the county. Although the supervisors unanimously adopted an export ordinance in 1990, a group of groundwater users considered that it did not have the teeth necessary to protect them from potential exports by the large surface water districts on the east side of the county. The group sought a legislatively sanctioned ordinance and succeeded in getting both houses to pass one during the 1992 legislative session.18 The governor vetoed the bill on the advice of the county’s water districts. It took a dramatic change in course by the largest water district, Glenn-Colusa Irrigation District (GCID), to enable a new set of working relations to develop among the county’s water users. Soon after the veto event, an internal management crisis within the district provoked the recall of the entire board and the hiring of new management. With the blessing of the district’s new board and the county board of supervisors, one of the new GCID directors took the initiative to organize an ad hoc, countywide water users group. The point of departure for the group was to find ways for the county’s different types of water users to coexist in harmony. GCID and some of the other senior surface rights districts recognized that there would be situations where they would be solicited and would want to transfer water south, as during the recent drought. The question was how to ____________ 18SB 867, the “Glenn County Groundwater Management Act,” introduced by Senator Mike Thompson in March 1991 and vetoed by the governor in September 1992. 50 engage in transfers without harming other local users dependent on groundwater. After several years of informal meetings, the county board formalized the process and nominated water users representing all the subbasins in the county to a Water Advisory Committee. In the discussions on management alternatives to export restrictions that followed, observers identified two key turning points. First, the irrigation districts came forward with a proposal for a new ordinance emphasizing “safe yield” of the aquifer. This proposal allayed the fears of some parties that there would not be safeguards for protecting the aquifer in the event of groundwater substitution-based transfers. The question then became how to operationalize the concept of safe yield, which is difficult, if not impossible, to measure directly.19 With input from DWR’s Northern Office, the committee developed the ordinance to follow the concept of “basin management objectives” (BMOs). Under this system, water users in the different subbasins would be responsible for establishing a monitoring system and determining target levels for the aquifer under different water conditions. The target levels (or BMOs) are, in effect, a surrogate measure of safe yield. Exports will be restricted only if they lead to unacceptable groundwater levels as determined by the targets. Significantly, pumping for in-county use can also be limited if problems persist after exports are cut back. The Glenn County supervisors passed the new ordinance in 2000, and the first set of basin management objectives was adopted in 2001. These are subject to revision annually as data-gathering and knowledge improve.20 Over this same period, Sacramento County was involved in a similar process, on a much larger scale, to find consensual solutions to its water problems through the Sacramento Water Forum. One outcome has been the creation of a Regional Groundwater Authority to facilitate ____________ 19“Safe yield” is generally used to mean a level of groundwater use that will not cause long-term decline of groundwater levels. Its estimation is complicated because records of groundwater extraction are limited and recharge rates are highly variable from one year to the next. See Dudley (2000). 20For a discussion of the background leading up to the passage of the new ordinance, see Brown et al. (2001). For details on the basin management objective philosophy and process, see Glenn County (2001) and Fulton and Dudley (n.d.). 51 groundwater management across a set of water entities.21 The export ordinance, adopted in 1980, was a response to the 1976–1977 drought and to more general concerns about protecting water rights prevalent in that period. County water officials indicate that for local water management purposes, the Water Forum process has largely supplanted the ordinance. The ordinance, recently transferred from the county code to the statutes of the county water agency, may still be invoked for water exports, however. Following Glenn’s experience, a number of other northern counties are in the early stages of adopting a BMO-based system: Tehama, Modoc, Siskiyou, Lassen, Butte, and Yolo. As with Glenn, this represents the potential for a significant departure from the exportoriented approach of the past. We will return to a discussion of BMOs as a groundwater management alternative in Chapter 7. San Joaquin Valley: Coping with Overdraft and Surface Water Scarcity Favorable soils and climate and the availability of water for irrigation have enabled the San Joaquin Valley to become California’s leading agricultural region. Historically, irrigation was developed with a heavy reliance on groundwater, as surface supplies were less plentiful than in the north. The availability of imported supplies through the federal and state water projects was a major boon to the valley’s agriculture. Part of the motivation for these infrastructure investments was to remediate the problem of groundwater overdraft in the valley. In 1980, most of the valley’s groundwater basins were classified as “critical,” a result of decades of agricultural pumping in excess of natural recharge rates. All eight counties overlie at least one of these basins (Table 4.2, Figure B.4). Concern with overdraft has spawned active conjunctive use programs in various parts of the region, whereby water districts take advantage of the higher levels of surface supplies in wet years to recharge the aquifers. The longest-running program of this nature is conducted by the members of the Friant Water Users Association, along the east side of the valley. Overdraft has also provided the opportunity for groundwater ____________ 21Thomas (2001) and McClurg (2002) provide detailed case studies. 52 banking. Kern County water districts have been the most active in this area. Although water tables have improved in some places since 1980 as a result of natural and artificial recharge,22 the region has been under new pressures because of environmental restrictions on Delta pumping. The reduced supply of imported project water, particularly to the CVP agricultural contractors on the west side of the valley, has been an important factor in the growth of the state’s water market. Meanwhile, ensuring adequate water supply for municipal and industrial uses is becoming a bigger issue, as the valley is now one of California’s fastestgrowing regions. Over the next two decades, its population is expected to grow by 51 percent, or 1.7 million inhabitants.23 The only region growing faster is the Inland Empire (Riverside and San Bernardino Counties). Higher populations have also meant that most of the region’s counties do not qualify for RCRC; only Madera and Merced are members. Ordinances restricting exports have nevertheless been proposed in six counties and adopted in four. The debates on adoption have been flavored by the problems of overdraft and surface water scarcity. San Joaquin County put in place an ordinance in 1996 over concerns that groundwater transfers by farmers along the Delta Mendota Canal, in the southwestern portion of the county, had contributed to overdraft of that basin during the early 1990s drought. In the years that followed, ordinances motivated by similar concerns were proposed in Stanislaus, Tulare, and Fresno. In both Stanislaus and Tulare, where the proposals do not appear to have come in reaction to specific transfer activity, water districts successfully argued that they were already engaged in adequate groundwater management practices. In Tulare, most districts are members of the Friant Unit, and there is an active multiparty groundwater management plan linking a number of these.24 In ____________ 22This is the case in Kern County, for instance (Kern County Water Agency, 2002). 23Projections are for the period 2000 to 2020. See Hans Johnson (2002). 24Deer Creek and Tule River Authority, whose groundwater management plan was introduced in 1995 53 Stanislaus, several multiparty groundwater management plans and associations were in place at the time of the proposal.25 In Fresno, the impetus for an ordinance was an impending permanent transfer of surface water. After prolonged public discussions, the districts managed to gain concessions but not to forestall its passage. At stake was a transfer of CVP contract entitlement from a small water district on the west side of the valley. The Widren Water District had agreed to sell its entire entitlement (just under 3,000 acre-feet) to a real estate developer for a new residential development near the town of Tracy, in San Joaquin County.26 County officials raised concerns that this agreement might lead to increased groundwater pumping to replace the transferred surface water. Although the amount of water was small, county officials feared that the transfer could set a bad precedent. The Widren deal was blocked, following litigation under CEQA and negotiation, but it convinced the board of supervisors to press for an ordinance to prevent further transfers out of the county. The Fresno ordinance went through over two dozen formal drafts before adoption. In particular, water districts did not want restrictions on their ability to engage actively in the water market, on which many farmers depended as a supplemental source of supply. The version ultimately adopted in 2000 reflects many of these concerns and makes the Fresno ordinance unique. In addition to the standard exemptions for use by districts with lands in bordering counties, it categorically exempts both water exchanges and temporary transfers. It also provides for a tenyear permit, whereas the other ordinances require renewal every one to three years. Although the initial impetus for the Fresno case was the potential for groundwater substitution, a number of observers share the view that the ultimate target is to retain existing surface supplies within the county, whether or not substitution is involved. While the Fresno ordinance was being debated, another permanent transfer of entitlement was proposed ____________ 25On the east side of the county, two local groundwater associations, including most water districts and municipalities; on the west side, the Northern Delta Mendota Groundwater Basin management plan, including five CVP contractors. 26Campbell (2000) provides a detailed discussion of the background to this ordinance. 54 involving Mercy Springs Water District, a neighbor to Widren. Proposals have also been aired concerning the sale of entitlement from Broadview Water District, in the same area. All three of these districts overlie lands affected by drainage problems, which severely reduce their agricultural productivity and the likelihood of groundwater pumping. If the county attempts to block future transfers of surface entitlement, districts may argue in court that the county is overstepping its authority. However, the net effect of county opposition to transfers so far has been to keep most of the water within the county.27 In Madera and Kern, the two remaining counties with ordinances, the reasons for adoption were atypical for this region. The Madera ordinance was introduced to provide the county with oversight for groundwater banking projects. The controversial “Madera Ranch” project that mobilized county residents and officials involved the ranch owner, the U.S. Bureau of Reclamation (USBR), and several Central Valley Project contractors.28 The project generated widespread county opposition over potential negative effects to local water quality and supply. Some parties also believed that the primary project participants were intentionally withholding information from local stakeholders. Although the Madera ordinance also regulates both direct and indirect groundwater transfers, these are generally not perceived to be major issues in the county. The two surface water districts that are members of the Friant Water Users Association were granted exemptions, as was the City of Madera. At first glance, the appearance of an export ordinance in Kern County may seem surprising, given the major role water agencies there have played in the development of the state’s water market over the past decade. The Kern County Water Agency and districts such as ArvinEdison Water Storage District, a Friant Unit member, have actively facilitated water exchanges and transfers and have been leaders in the ____________ 27There is still no long-term deal concluded for Widren’s water, which is transferred annually to other agricultural users in Fresno. Part of the Mercy Springs entitlement will pass over to users in Santa Clara and Santa Cruz Counties, and part will stay in Fresno (see Table A.6). The county is also encouraging prospective Monterey County purchasers of Broadview’s water to take only part of the supply. 28For case studies, see Campbell (2000) and Thomas (2001). 55 groundwater banking movement that has already helped ensure dry-year supplies. Closer examination reveals that the ordinance has nothing to do with water management in the San Joaquin Valley part of the county. Rather, the reasons for Kern County’s ordinance hearken back to the water marketing concerns in Inyo and Mono. The southeastern part of Kern overlies the Lahontan Basin, a hydrologic area that also encompasses Inyo and Mono Counties. The ordinance was developed in response to concerns over a proposal by a private water marketing company to buy a former alfalfa ranch overlying the basin and transfer 10,000 acre-feet per year over a 40-year period to the City of Los Angeles through the Los Angeles Aqueduct. In light of the low rainfall and lack of surface water in this desert and foothill area, people worried that Los Angeles would “recreate an Owens Valley in Kern County.” Following a stopgap emergency ordinance, a permanent ordinance was rapidly put in place to require environmental review of any groundwater transfer out of the basin. This was passed with the approval of the Kern County Water Agency, and there was never any serious consideration of extending the ordinance to cover the San Joaquin Valley portion of the county. Only two valley counties, Kings and Merced, have never had formal consideration of export ordinances. One observer suggested that Kings County has not felt the need for one because there is a relatively harmonious atmosphere among water districts, most of which operate with a fairly high level of groundwater monitoring and management. There is, notably, an active multiparty groundwater management plan for users overlying the Tulare Lakebed. Another observer noted that one local agency, the Kings County Water District, has had its own ordinance to prohibit groundwater mining since the 1950s. Like the Yuba County Water Agency, this district has board members who are also county supervisors. Merced County has benefited from a highly effective water planning approach, spearheaded by the county’s two main purveyors—the City of Merced and Merced Irrigation District. In the mid-1980s, dissatisfied with the ten-year planning horizon required by CEQA, the city decided to do a 40-year general plan. The plan raised questions over whether the groundwater system in place would be capable of sustaining the much 56 larger city of the future. The city began discussions with the irrigation district, and in 1991 the two parties embarked on a regional water supply planning exercise, with considerable input from outside consultants for the modeling work. A Technical Advisory Committee with representatives of county departments, other cities, the large water-using industries, USBR, the regional water quality board, and the state and federal environmental agencies provided regular input into the planning process. The Merced Basin regional water supply plan was completed in the mid-1990s and provided the basis for a long-term regional conjunctive use program that could meet the basin’s needs.29 Following its release, the two lead agencies spearheaded the development of an AB 3030 groundwater management plan, involving 15 water purveyors in the basin. Founded in 1997, the Merced Area Groundwater Pool Interests, or MAGPI, has a board of directors that meets quarterly and a Basin Advisory Panel involving a wider range of stakeholders that meets monthly. Initially, the group focused on establishing an effective monitoring system for the basin. It has also begun investigating conjunctive use projects in association with the Department of Water Resource’s Integrated Storage Investigation program. This process, both highly participatory and involving a great deal of technical input, has preempted the need for alternative protective mechanisms in Merced County. Summing Up A statistical analysis of California’s counties suggests that their likelihood of adopting export restrictions varies with economic, water supply, and institutional characteristics. Counties with a higher share of the workforce in agriculture are more likely to impose restrictions. Other key factors are the presence of a groundwater basin designated as “critical” or having “special problems” by the Department of Water Resources in 1980 and membership in the Regional Council of Rural ____________ 29The water plan was updated in the late 1990s in collaboration with the University of California, in connection with the plans to establish UC Merced. 57 Counties, an association that has promoted groundwater ordinances since the mid-1990s. A qualitative review of the reasons for ordinance adoption suggests a strongly regional character to the patterns observed. Many mountain counties have responded to specific threats of long-term groundwater exports from their basins to Nevada and to the Los Angeles region. Ironically, the mitigation settlements for Owens Valley and Mono Lake have kept the threat of new groundwater exports very much alive because of freed-up conveyance capacity in the Los Angeles Aqueduct. The menace of uncontrolled private groundwater pumping for sale to Los Angeles also motivated Kern County’s ordinance, which covers only the southeast portion of the county adjacent to Inyo. Elsewhere in the San Joaquin Valley, ordinances have responded to concerns about transfer and banking operations in overdrafted basins and in an overall context of surface water reductions. In the Sacramento Valley, a major issue has been local disparities in the distribution of water rights—notably the ability of senior surface water-rights-holders to engage in groundwater substitution while others risk shortages. In general, Central Valley counties without ordinances have benefited from greater institutional cohesiveness on water matters. Countywide, umbrella water agencies have played key leadership roles in both Yuba and the San Joaquin Valley portion of Kern County. Elsewhere, substantial efforts have been undertaken to create new institutions. In Merced, an initiative begun by the city and the largest water district in the early 1990s has led to a regional water planning exercise and an active groundwater management plan involving all players overlying the Merced Basin. In Glenn and Sacramento, stakeholder consultation has produced programs to actively manage the groundwater basin. In Glenn’s case, this process culminated in the adoption in 2000 of a new county ordinance based on basin management objectives, which has effectively supplanted the exportpermitting ordinance on the books since 1990. A number of other counties are now following Glenn’s lead. 58 5. Water Market Effects: Do County Restrictions Have Teeth? What effects, if any, are county policies having on the state’s water market? Because most ordinances address both direct groundwater exports and surface water exports that result in additional groundwater extraction, we might expect significant overall market effects. In counties with restrictions, the only types of transfers not subject to local approval are those involving water held in surface storage (available to few water purveyors), excess surface water (available mainly in very wet years), or water conserved through efficiency gains or land fallowing (available only through investment outlays or forgone crop income). If the ordinances reflect a public view that out-of-county sales from any source should be discouraged, even these types of transfers could be affected. Of course, it is also possible that county restrictions have little practical consequence if enforcement efforts are ineffective. This chapter assesses the market effects of export restrictions from two angles. First, it draws on information from county-level interviews to evaluate ordinance implementation. This discussion focuses on how the permitting process has worked for out-of-county transfers. Second, it looks at the effects of the restrictions on county water marketing activity, drawing on the database on water transfers presented in Chapter 2. Export Permitting: A Largely Uncharted Territory In counties with ordinances restricting exports, those wishing to engage in the restricted activity must obtain a county permit, which invokes review under CEQA. The permit-granting authority—in this case, the county—determines the extent of CEQA review. CEQA obligations can be met with a relatively simple environmental assessment, 59 if the county is satisfied that the permitted activity is likely to have little or no negative effect, or if adequate provision is made for mitigation. The county can also require a full-scale environmental impact review (EIR), in addition to preliminary studies. Moving from an environmental assessment to an EIR easily increases costs for the applicant by a factor of ten or more, with EIRs typically costing $300,000 or more. In the years since counties introduced a review process for water exports, there have been few requests for permits and even fewer permits granted. Among the 22 counties with export ordinances, a total of 16 export permit applications have been tendered in four counties: Inyo (12), Mono (2), Tehama (1), and San Joaquin (1).1 Of the 16 applications, only one has been granted, in Tehama, and four remain actively on file, in Inyo. Most of the applications from Inyo and Mono have concerned private sector transfers to Los Angeles. The permit granted in Tehama was for the movement of a small quantity of groundwater to land owned by the same farmer in another county. The application in San Joaquin was for the reexport of imported banked groundwater by the East-Bay Municipal Utilities District in 1997.2 After a protracted review process, the permit was denied in response to concerns about granting a noncounty entity access to the aquifer. A permit for groundwater banking was nearly filed in Madera by a private water firm, Azurix, a partial subsidiary of the Enron Corporation, which intended to pursue the Madera Ranch project that was at the origin of that county’s ordinance. Whether Azurix’s sale of the property in 2001 without filing ____________ 1As noted in Chapter 3, three counties in this group have actively required permitting for some in-county water movements. In San Benito, there were five applications for small residential subdivisions, of which three were approved and two withdrawn. In Sierra County, one permit was granted for a transfer of treated wastewater from an industrial property to adjacent property in agricultural use. In Imperial County, 19 conditional use permits have been granted since 1994 under the well-permitting process instituted in 1972. Recently, Imperial County’s Board of Supervisors rejected a groundwater use petition that had been pending for eight years for use on a farm near the San Diego County border. Among reasons for the rejection were apparent concerns that the applicants might subsequently transfer some of the water to the Borrego Water District in the neighboring county (Mitchell, 2003). 2For a detailed case study, see Thomas (2001). 60 was more a result of its own financial problems or the continued local controversy remains a matter of speculation. A broad local coalition of interests was against both the project and the involvement of an outside firm, and popular sentiment was prominently displayed on billboards along Route 99 admonishing Enron/Azurix to stay out of the valley. In two other counties, those interviewed indicated that permit applications might be on the horizon. In Calaveras, where the ordinance is quite recent, the Calaveras County Water District is investigating a potential conjunctive use project and intends to launch an application if the study confirms the project’s viability. In Sacramento, officials anticipate that a groundwater export being considered as part of a water quality mitigation program may trigger a permitting requirement. The limited permitting experience in counties with export restrictions stands in stark contrast to that of counties whose groundwater protection ordinances were put in place to regulate withincounty uses. In San Diego, Napa, and Mendocino, the review process has been active, with numerous approvals granted.3 When asked to account for the relative dearth of permitting activity in counties with export restrictions, observers offer three explanations. The first holds that there is little local interest in transferring water outside the county anyway; the ordinance is merely an expression of the popular consensus. The second explanation is that the ordinance may be ineffectual in screening exports that continue to occur. This could arise either through lack of public awareness of the permitting requirement or lack of good surveillance to ensure compliance of those inclined to avoid permitting. The third and most common view holds that the permitting process itself discourages transfers. Up-front costs of environmental review, the likelihood of rejection in places hostile to transfers, and the limited length of time for permits once granted are all cited as deterrents. In a number of counties, officials admit that this “discouragement factor” is at least partially intentional. Legally, counties cannot prohibit exports, but they can make it difficult for potential sellers to obtain approval. ____________ 3San Bernardino’s ordinance was approved late in 2002, too recently to have established a record on permitting. We were unable to obtain information on the implementation of Monterey’s ordinance. 61 In principle, the information requirements and up-front study costs depend on the hydrological characteristics of the basin and the extent to which a knowledge base has already been established through prior studies. In practice, another consideration is the degree of public confidence in the process. In places where prior experiences have engendered a distrust of water transfers, the up-front requirements are bound to be higher than where there is no particular local history. A case in point is Butte County, where water officials would like to see a permitting process that effectively screens transfers for third-party effects without blocking the possibility of water marketing. In light of the heated controversy surrounding the adoption of the ordinance, however, it is likely that the first test case would need to go through a full-scale environmental impact review. The cost might be worth it for a long-term transfer program, but it virtually precludes anyone wishing to engage in short-term transfers from applying. Faced with the opportunity to sell water under DWR’s dry-year purchase program in 2001, farmers in the Western Canal Water District elected to generate surface water through land fallowing rather than applying for a permit to engage in groundwater substitution. As it happened, rice prices were sufficiently low that year that some farmers found it worthwhile to forgo crop revenues. More generally, it might have been better for all parties to consider groundwater substitution, but the district was effectively prohibited from pursuing that option. Effects on the Water Market The Butte example highlights the fact that the ordinances will not always limit trade in water if alternative means such as fallowing are available and acceptable to farmers. Because this will not always be the case, however, ordinances limiting the use of groundwater for transfers could have aggregate effects on the market. These effects would be compounded if the ordinances have a more general dissuasive effect on water exports from any source, which appears to be the intent in some counties. Aggregate market effects of two types might be expected. In counties where there are willing buyers locally, the ordinances might be expected to shift supply from exports to in-county uses. In counties without local 62 demand, or where the local demand is more limited than the potential supply, there would be an overall reduction in the volume traded. The alternative prediction—of no effects on the volumes of total transfers and transfers leaving the county—would correspond to a scenario where the ordinances are generally ineffectual. Data Sources To examine whether there is evidence of these effects, we will employ the database on water transfers in California presented in Chapter 2.4 The analysis covers the 12-year period beginning in 1990, the point at which data on counties of origin and destination become more precise. The aim will be to determine the effect of export ordinances on two county-level measures of water trading: annual sales and annual “exports,” defined as transfers going to users outside of the county. We will focus the analysis on two geographical groups. The first is the set of 34 “water trading” counties—counties that appear at least once in the transactions database over the period under review (Figure 5.1). This includes all 18 Central Valley counties, the Southern California region, the inland portion of the San Francisco Bay Area, and San Francisco itself. It excludes the mountain counties and counties along the north and central coast. These nontrading counties are excluded for statistical reasons; the key econometric models cannot be estimated when they are included. Moreover, there may be structural reasons for the lack of trading activity in many of these counties: Counties along the coast and to the far north rely on local river and groundwater sources and are not hydraulically connected to the state’s main water arteries. It is possible that some local trades occur in these regions that we have not been able to trace with our sources. The second geographical group is the set of 18 Central Valley counties. It is interesting to look separately at this region for two reasons. First, Central Valley counties have been the major source of water for the market since the early 1990s. Second, this region has the greatest potential for groundwater-related transfer activity, both through ____________ 4For a detailed description of data sources and methods used in the analysis, see Appendix D. 63 Transfer activity Figure 5.1—California’s Water-Trading Counties, 1990–2001 groundwater substitution transfers and through groundwater banking.5 The potential role of groundwater in this region stands in stark contrast to the situation in Imperial County, the other major water supplier. Although Imperial has imposed groundwater export restrictions since ____________ 5See Purkey et al. (1998) and related work from the Natural Heritage Institute on the potential for groundwater banking in the Central Valley. 64 1996, these restrictions are unlikely to have much practical influence on the overall volume of water sales. Imperial County has vast quantities of surface water rights from the Colorado River and few areas with usable groundwater.6 To see whether the presence of an ordinance affects county trading behavior, it is necessary to control for other factors that might also be important. Multiple regression techniques allow us to isolate the effect of the export restrictions while holding the influence of other factors constant. We have selected a set of seven variables—described in Appendix D—to account for agricultural and residential demand and water supply conditions. In addition, we will test for the effects of state and federal policy changes on the trading environment by including a time trend in the regressions. If the new operating rules are effectively increasing transfer activity, above and beyond what would occur because of changes in the other variables noted above, this variable should be positively related to sales. A time trend also captures the effect of “learning-by-doing” by water users as they gain familiarity with the market. Results County regulations have noticeably restricted sales in the statewide market. In any given year, the typical county with an export restriction sold 14,308 acre-feet less than a county without one. The estimated effect on exports is larger—at 16,948 acre-feet—although the difference is not statistically significant. These results suggest that the county restrictions have not only reduced sales but have also resulted in some shifting of water from external to within-county buyers, by an average of 2,640 acre-feet per year. The quantitative effect of export restrictions is even more pronounced when the analysis is restricted to the 18-county Central Valley sample. Holding other factors constant, export ordinances reduce overall sales by 20,789 acre-feet and out-of-county ____________ 6In large parts of Imperial County, the groundwater is too saline for agricultural uses. Salinity is also a limitation for groundwater use in San Benito County, the other non–Central Valley county within the “trading counties” group that has an export ordinance. 65 sales by 26,245 acre-feet. The corresponding increase in local sales induced by ordinances is nearly 5,500 acre-feet per county per year. Since 1996, the point at which a number of counties began to adopt ordinances, the restrictions have reduced exports by 932,000 acre-feet, or 19 percent of all out-of-county sales (Figure 5.2). Of this total, 145,000 acre-feet that would otherwise have been exported have been sold locally. The lion’s share (787,000 acre-feet) has simply been kept off the market. In all, this represents a 14 percent reduction compared to the level of predicted sales in the absence of county restrictions. For the Central Valley counties, these effects are even larger: a 39 percent reduction in exports and a 25 percent reduction in overall sales because of restrictive ordinances. The resulting shift from exports to the local market appears to have increased within-county sales by nearly 50 percent. Meanwhile, state and federal measures to improve the trading environment, as measured by the time trend, have had a substantial positive effect on water sales. The typical county was likely to sell 42,000 acre-feet per year more in 2001 than in 1990, under identical conditions of water supply and agricultural demand. During the first six years analyzed here, the positive effect of state and federal support far outweighed the negative effect of county restrictions (Figure 5.3). As the 7,000 6,000 5,000 Actual sales Sales if no export restrictions Acre-feet (thousands) 4,000 3,000 2,000 1,000 0 All sales Exports Within-county sales Figure 5.2—Market Effects of Export Restrictions Since 1996 66 Acre-feet (thousands) 900 800 700 600 500 400 300 200 100 0 1990–1995 Increases in sales from state and federal policy Reductions in sales from county restrictions 1996–2001 Figure 5.3—Market Effects of State and County Policy Environments number of counties with restrictions has grown, this has ceased to be the case. From 1996 to 2001, county restrictions cancelled out the positive effect of state and federal policies to encourage trade. Counties introducing export restrictions may have reduced their trading activity for reasons other than the ones captured in our model. In this case, the strong correlation we observe between ordinances and the reduction in sales and exports would not imply causality. In light of the reasons given for ordinance adoption, however, we regard this as unlikely. The bigger question is whether the ordinances will continue to wield the same effect in the years ahead. The pending long-term transfers from the agricultural to the municipal contractors of the Colorado River Project, negotiated as part of California’s program to reduce its overall use of project water, will substantially increase overall volumes traded statewide. The combined transfers from Imperial Irrigation District to San Diego (200,000 acre-feet), Imperial to Metropolitan Water District of Southern California and Coachella Valley Water District (100,000 acre-feet), and Palo Verde Irrigation District to MWDSC (111,000 acrefeet)—achieved through a combination of land fallowing and efficiency gains—represent over 30 percent of the amount traded in 2001. This 67 shift of activity by water users in Imperial County, which has an export ordinance, is bound to alter the statistical relationship between trades and county restrictions in the statewide market.7 For counties in the Central Valley, the other main source of market supply, nothing on the immediate horizon suggests a weakening of the effect of export restrictions. Some Sacramento Valley observers have suggested, however, that the restrictions could loosen under another major and prolonged drought. With external pressure to make water available, counties may have little choice. A more positive impetus for change lies in the possibility that counties will move beyond export restrictions to a wider groundwater management system, thereby ensuring protection of local users without discouraging market activity. The Glenn County experience in 2001 indicates how this might work. That year, several of the county’s CVP settlement contractors participated in a program to send water to Westlands Water District, using a combination of fallowing and groundwater substitution to free up surface water supplies. The potential groundwater effects of the transfer were monitored through the county’s new basin management objectives system as part of the more general monitoring program being established by the county’s water users. In conjunction with the export, Glenn-Colusa Irrigation District also made surface water available to some water-short entities within the area, at a lower price. Summing Up Have county-level groundwater export restrictions had an effect on the water market? The record on permitting suggests that they have. The very low number of permit applications supports the view that this process is more useful as a deterrent than as a screening mechanism. ____________ 7As an indication, we incorporated all pending long-term transfers listed in Table A.5 to the sales and export data for 2001 and reran the regressions for the 34-county sample. With just one year of the new Colorado River Project trades (assumed at full volume), the estimated effect of county restrictions diminishes in size and statistical significance. 68 High up-front costs and the likelihood of negative public opinion guiding the decision process are factors discouraging parties from filing. As the recent experience in Butte County demonstrates, a lack of groundwater permits will not necessarily block transfers if alternatives such as fallowing are available and acceptable to farmers. In the aggregate, however, there is likely to be a market effect, both in reducing total sales and in shifting some water to in-county users (who will typically be willing to pay less than outsiders). We find evidence of both effects in a statistical analysis of county trading behavior from 1990 to 2001. In any given year, the presence of an export restriction reduced a county’s trades by 14,300 acre-feet and shifted 2,640 acre-feet to incounty buyers. In aggregate terms, this reduced out-of-county sales by 932,000 acre-feet, or 19 percent, and total sales by 787,000 acre-feet, or 14 percent, since 1996. Overall, the negative effect of county restrictions cancelled out the positive effect of a generally improved trading environment resulting from state and federal regulatory changes. Looking ahead, the key question is whether communities can move beyond an export restriction mode to one combining protection of local users with the flexibility to allow water trading where feasible. In areas where the groundwater effects of trade are the concern, this means establishing systems for monitoring, mitigation, and actively managing the aquifer. Land fallowing, a key alternative source of water for the market, also raises questions of harm to local communities. As the economics of the water market make this an increasingly attractive option for some farmers, there is a need to consider how and when economic mitigation programs will be a necessary component of water transfers. The following chapter addresses this issue. 69 6. Mitigating the Economic Effects of Land Fallowing Land fallowing has been the basis for several major water transfer programs in California since the early 1990s and is an integral feature of the active agricultural water market within the San Joaquin Valley. Noncontroversial in some situations, the concept of idling farmland to sell water has generated tremendous conflict in others. At issue are both equity questions and the potential aggregate consequences of fallowing. In this chapter, we address the economic, legal, and institutional issues that fallowing for the water market raises, with a focus on the following questions. Under what circumstances is it important to address the distributional consequences of the transaction? What do we know about the conditions under which fallowing would cause harm to the local economy, and what policy alternatives are available to minimize negative effects? We begin with a review of California’s recent experiences with land fallowing for the water market. California’s Recent Experiences with Land Fallowing DWR’s Dry-Year Programs Fallowing was a major component of the 1991 drought water bank, accounting for 415,000 acre-feet of the 821,000 acre-feet purchased. Most of the contracts were negotiated directly with individual farmers, who were paid not to irrigate and were compensated on the basis of the imputed water savings. In Yolo County, where a substantial part of the fallowing occurred, the board of supervisors took the view that the state, as purchaser, should indemnify the county for the losses it incurred as a result of the decreased economic activity. Notably, the county observed an increase in demands for unemployment-related social services as a consequence of lower farm labor employment. Challenging both the 71 legality of the claim and the facts on which it was premised, DWR declined to pay the county the modest amount requested.1 It also cut the fallowing program short. In the 1992 and 1994 banks, water was purchased entirely from surface storage and groundwater exchange. Over time, however, the Yolo episode has led DWR to approach the fallowing question somewhat differently. When it launched a new trial run with fallowing during its 2001 dry-year program, DWR made arrangements to pay a 5 percent fee ($3.75 per acre-foot) to Butte County to handle the associated mitigation costs of the transfer by Western Canal Water District. As we will discuss below, this raised both practical and legal issues at the county level that are still being worked through. It nevertheless appears that the state’s current default position is to compensate for third-party effects of fallowing for its own water purchases. Mitigating the effects of fallowing is also a central focus of the programmatic environmental impact review now under way for the Environmental Water Account, a joint state-federal program. Long-Term Fallowing Along the Colorado River The year following the drought water bank experience, MWDSC launched a two-year fallowing program with the Palo Verde Irrigation District (PVID). Under the trial program, the district idled land to free up 93,000 acre-feet of water per year for MWDSC. Encouraged by the outcome, the two parties then developed a long-term transfer arrangement, under which PVID farmers will fallow up to 29 percent of their 91,000 irrigated acres of land, to send up to 111,000 acre-feet annually to the coast. The 35-year transfer was in the final stages of approval by the boards of both agencies in late 2002 and should get under way in 2003. Both the PVID board and area farmers regard the transfer as an opportunity for the area’s economy on the grounds that it helps stabilize farm incomes.2 A $6 million fund has been set aside to ____________ 1The bill submitted by the Yolo County Board of Supervisors was for $129,305 to cover general assistance and Aid to Families with Dependent Children entitlements for 450 persons deemed to have lost their employment because of the water transfers (Gray, 1994b). 2See “Rural Palo Verde Valley Agrees to Colorado River Pact” (2002) and Lyn Johnson (2002). 72 compensate the community for potential income losses from lower agricultural activity. At the same time that Palo Verde farmers and the PVID board were lauding the forthcoming fallowing program in the local press, representatives of the neighboring Imperial Irrigation District (IID) were going on record with their doubts over whether they would vote for a similar program for sending Colorado River water to San Diego. Unlike the Palo Verde deal, the Imperial deal with San Diego has been one of the biggest water controversies in recent California history. Palo Verde’s transfer was premised on land fallowing, but this method was essentially imposed on Imperial because environmental constraints precluded the preferred option of conserving water through more efficient irrigation practices. Like some other local agencies, IID had a policy disallowing fallowing as a source for water transfers. Accordingly, the initial transfer deal, agreed to in 1998 by the IID and San Diego County Water Authority boards, explicitly ruled out fallowing. It was not until early 2002, during the environmental review phase of the transfer, that objections were raised to the efficiency-based method. By reducing the district’s agricultural runoff, the irrigation improvements would hasten the increase in salinity levels in the Salton Sea, a major aquatic bird sanctuary along the Pacific flyway. By this time, the transfer had become a linchpin of California’s 4.4 Plan to reduce its use of Colorado River water over a 15-year period. Under the threat that the U.S. Secretary of the Interior would immediately reduce California’s annual supplies from 5.2 to 4.4 million acre-feet if parties within the state did not reach an agreement by the end of 2002, IID was pressured to consider land fallowing as an alternative means of conserving water for the transfer.3 ____________ 3Technically, the agreement that needed to be reached among California’s Colorado River contractors by the end of 2002 is the “Quantification Settlement Agreement,” under which the senior agricultural contractors—Palo Verde, Imperial, and Coachella Valley Water District—agree to “quantify” their water rights to a specified amount. Up to now, in the order of seniority, these districts have had the right to any amount of water they can put to beneficial use, within the overall allocation available to the state. The lack of a firm upper limit on use has meant that any transfer deals concluded between any one of the parties and a more junior rights-holder (notably Metropolitan Water District of Southern California and San Diego) would not necessarily result in a reduction of water 73 Negotiations over this transfer have been extremely complex, in part because the fate of the Salton Sea is uncertain, even without the loss of runoff from Imperial’s farms. Created in 1905 when massive river flooding broke through a canal erected by area farmers, the sea has relied on agricultural runoff from the district ever since as its primary source of replenishment. Even without the transfer, the sea is predicted to become too saline to support the fish and other marine life on which the birds feed within one to two decades unless extraordinary actions are taken.4 So far, the science of the problem has proven elusive, with uncertain proposals involving price tags of about $1 billion or more. As a consequence, one of IID’s major concerns in the transfer talks has been to bind its liability limits with respect to the sea’s future health. The other major sticking point has been the fallowing question. Once fallowing became the suggested means of achieving the water savings for the transfer, the deal switched from one that was “win-win” for the district to one involving winners and losers. An efficiency-based transfer program would involve keeping all the land in production while creating local jobs to carry out conservation investments. A fallowing program implies some job losses. The debate over just how many and over what time horizon has been central to the negotiation process. Under the terms of a proposed deal negotiated in October 2002, the new per-acre-foot price to be paid by San Diego includes enough additional money to cover up to $20 million in mitigation funds over 15 years. The maximum annual acreage to be fallowed is 30,000 (of a total of 450,000 irrigable acres in the district). Imperial would be free to switch from fallowing to efficiency-based methods of water savings beginning in year 16. San Diego would agree to cover any excess costs of third-party effects beyond those provided for in the $20 million.5 ________________________________________________________ use by the agricultural contractors. This problem arose with the initial transfer of 110,000 acre-feet from IID to MWDSC begun in 1988 and noted in Chapter 2. The transfer was based on efficiency gains in IID’s network, financed by MWDSC. Although the gains were realized (and allocated to MWDSC), IID actually increased its water use in the subsequent period. 4See Imperial Irrigation District and U.S. Bureau of Reclamation (2002). 5For details of the proposal as per the October 15, 2002, agreement, which has remained the basis of the proposals on fallowing, see “Summary of Water Agreement,” 74 In sharp contrast to the Palo Verde case, no one in the Imperial Valley has gone on record to praise the deal, except to say that it was the best the negotiators could do in a difficult situation. In the months leading up to the October negotiations, both federal and state authorities had made it clear that the district risked having its water rights challenged in the absence of a successful transfer agreement.6 In December 2002, IID’s board voted 3-2 against the terms of the deal brokered in October. Although the effects of fallowing were cited as one of Imperial’s Board’s concerns, the “deal-breaker” in the end appeared to be the lack of adequate guarantees against possible lawsuits over environmental effects to the Salton Sea.7 The fate of the transfer remains uncertain at the time of this writing (May 2003), but the latest proposal brokered by the state involves an additional $200 million in state funds directed to mitigating effects to the Salton Sea. Fallowing in the San Joaquin Valley The planned Palo Verde transfer and the possible sale by Imperial are the state’s first large long-term water transfers based on land fallowing, but they are not the only places where this is taking place. Fallowing has been a regular feature of the temporary agricultural water market within the San Joaquin Valley since the early 1990s. The district-to-district transfers of this type mainly involve moving water from land owned or leased by the same farmer to more productive, water-short land elsewhere in the valley, notably within Westlands Water District and several neighboring CVP contracting districts.8 Fallowing, in these contexts, is accepted as part of the farmers’ overall land management plans. Water districts allowing this activity do vary, however, in the extent to which they impose conditions on the transfer. In some districts, once the land is leased there are no restrictions on transfers to other lands farmed by the ________________________________________________________ October 17, 2002. The maximum amount to be fallowed is not explicitly mentioned in the agreement but has been cited elsewhere by IID directors and others (Vogel, 2002). 6See statements by U.S. Interior Department Assistant Secretary Bennett Raley in Kasindorf (2002). For a discussion of state policy, see Kasler (2002). 7See Conaughton (2002). 8Drainage problems, which are reducing the productivity of some areas within Westlands, have also encouraged land fallowing for water sales within the district. 75 lessee. For farmers within the Kern County Water Agency service area, a casual lease is not sufficient for gaining the right to transfer the water. Land retirement was also the basis for the permanent transfer of up to 130,000 acre-feet of State Water Project entitlement from Kern County farmers to municipal users under the Monterey Agreement. Importantly, however, this proposal came about as a way of reallocating water from land that had already come out of production because of marginal economic conditions; it did not precipitate crop idling. Two recent sales of SWP entitlement by Kings County farmers, one to other farmers and one for municipal uses, were also predicated on removing water rights from lands that are becoming unprofitable to farm.9 Recent proposals by some water users in Kern County would involve a multiyear, rotational fallowing scheme to free up water for municipal users outside the county.10 Whether and under what conditions such programs are acceptable to the wider community is one of the subjects currently on the table in a countywide review of water transfer policy launched by the Kern County Water Agency. In the eastern part of the San Francisco Bay Area and in San Joaquin County, several long-term, local agriculture-to-urban transfers also derive some of the water savings from land fallowing or land retirement.11 Such ____________ 9Sales by Tulare Lake Basin Water Storage District to Dudley Ridge Water District (for agriculture) and Antelope Valley–East Kern Water Agency (AVEKWA) (for municipal uses). For details, see Table A.6. Technically, there is no explicit fallowing requirement under the terms of the transfer; rather, lands for which SWP entitlement is sold are permanently disallowed from receiving future SWP entitlements. The transfer to Dudley Ridge involved lands owned by the same farmer; the water will be used in Dudley Ridge where the land is more productive. The transfer to AVEKWA involved lands being taken out of production by a large ranch, which is scaling back operations. 10See Semitropic Water Storage District (2002). 11This includes the ongoing transfer from Byron-Bethany Irrigation District to Alameda County Flood Control and Water Conservation District, Zone 7 (for which water is made available through temporary fallowing of approximately 600 acres and water savings from 300 acres of land already retired) and the proposed transfer of contract entitlement from the Westside Irrigation District and Banta Carbona Irrigation District to the City of Tracy (made possible by preexisting land retirement). See Tables A.5 and A.6 and the environmental documentation for these transfers (Alameda Flood Control and Water Conservation District Zone-7, 1994, and CH2MHill, 2002a, 2002b). 76 transfers raise somewhat different issues from the cases where fallowing involves sending water to a distant destination. The selling districts defend the action on the grounds that residential development is in all events encroaching on the area’s farmland; the transfers provide a way of lessening costs for the remaining farmers while moving water locally to new uses. Objections, when raised, relate to concerns over the consequences of development within the area, not to an outflow of economic opportunity.12 Future Trends? Recent developments in the Sacramento Valley suggest that fallowing for transfers to points south of the Delta will be an increasingly important component of the water market. As noted, DWR purchased water conserved through fallowing from farmers in Butte County for its 2001 dry-year program. Fallowing was also the primary source of the water for the transfer from CVP settlement contractors in the Sacramento Valley to Westlands in that year.13 As the fate of the Imperial–San Diego transfer remained uncertain in late 2002, officials of Metropolitan Water District of Southern California—the agency that stands to lose the most water if the state fails to meet the deadline for the 4.4 Agreement—unveiled a plan to purchase 205,000 acre-feet from this same group of senior water-rights-holders in 2003. For this transfer, again primarily based on land fallowing, Metropolitan has proposed to include a $5 per acre into a mitigation fund (5 percent of the $100 per acre-foot price negotiated for the water). Once the details of this oneyear deal are worked out, the water districts intend to discuss possible longer-term arrangements combining fallowing and groundwater ____________ 12For instance, the Sierra Club has taken issue with the pending sales to Tracy on grounds that they encourage sprawl (Cooper, 2002). 13Participants in the program included Glenn-Colusa Irrigation District, PrincetonCodora-Glenn Irrigation District, Provident Irrigation District, Reclamation Districts 108 and 1004, several mutual water companies, and private farms with individual CVP contracts. Of a total of 160,000 acre-feet, 91,000 acre-feet were made available through fallowing, corresponding to approximately 27,500 acres. 77 substitution. The fallowing is concentrated on acreage planted to rice, for which prices are at historical lows.14 Thus, we see diverse local reactions to the prospect of fallowing and to mitigation. So far, the only substantial mitigation programs envisaged are for the two large long-term transfers of Colorado River water and for temporary purchases by two big buyers—the state itself and the largest urban water agency. There is, as yet, no real track record on either the consequences of a long-term fallowing program or the “how-to” of mitigation. Before turning to some of the practical questions of designing programs to limit negative effects, we examine briefly the economic and legal issues of fallowing for water transfers. Economic and Legal Issues of Land Fallowing for Water Sales Economic Incentives and Third-Party Effects From the seller’s standpoint, fallowing to free up water for the market is likely to be most attractive, the higher the sale price of water and the lower the value of the water in agricultural uses. This is why short-term land fallowing is often seen as flexible tool for coping with drought conditions. At such times, water will fetch a better market price, thus compensating farmers for forgoing their own use. It is also why longer-term fallowing programs for sale to municipal users may be appealing to some farmers, because the municipal market can generally afford to pay a higher price than can other agricultural users. The incentive structure for individual farmers will depend critically on the rules established by the local water district. In some situations, the bulk of the proceeds will go directly to the farmer who is idling land. In a surprising number of cases, however, the proceeds remain at the district level and go toward keeping down costs to other users. This was ____________ 14See Economic Research Service (2002). Changes in the U.S. farm support programs since the second half of the 1990s have also made it more attractive to fallow when farm prices are low. Under current programs, farmers of commodities eligible for price support (notably rice and cotton in California) are paid on the basis of historical acreage rather than current acreage. As a result, they are not penalized for idling the land to make the water available for the market. 78 the procedure for the Monterey Agreement transfers in Kern County, as well as the transfer in the Bay Area noted above. Without a change in operating rules, this redistribution would also occur in the many districts that do not attribute specific contract amounts to individual landowners but rather deliver water on the basis of annual requests.15 The attraction of “farming water” instead of farming land is clearly greater when individuals, and not districts, stand to benefit from the sale. From the standpoint of the surrounding community, the effects of a fallowing operation will depend on the interaction of two types of effects. On the one hand, there are the effects of the changes in land use on onfarm and farm-related employment, tax revenues, and, indirectly, the wider economic activity of the area. On the other hand, there are the proceeds of the water sale, which, if spent locally, contribute to job and revenue creation. Thus the aggregate effect of a water sale achieved through a reduction in cropland is not necessarily negative, even in the short run. If, as is typically assumed, farmers elect to fallow the lowvalue crops—which produce less profit per unit of water used and which require relatively lower labor inputs—the associated employment and revenue reductions will be limited. If, at the same time, they reinvest proceeds of the water sales into farming operations, for instance, by releveling the land or making upgrades in equipment, this reinvestment can have a positive effect on employment and revenues. Typically, however, there are some losers from land fallowing, even in a generally positive scenario such as the one described. The new investments in land leveling or other farm improvements will give a boost to those sectors, whereas the fall in crop output will reduce the demand for specialized services such as harvesting and processing and, in all likelihood, some farm labor. This is why the notion of mitigation enters the equation. Mitigation is envisaged both as a short-term compensation for income losses and as a means of assisting those who may be permanently affected by a long-term water transfer to adjust to the new economic circumstances. ____________ 15For instance, Yolo County Flood Control and Water Conservation District and Madera Irrigation District. 79 Nevertheless, mitigation for land fallowing poses some distinct economic, legal, and policy questions. The negative effect on those who lose out is an economic spillover effect, or what economists sometimes refer to as a “pecuniary externality”—the effect of one person’s business decision on someone else’s financial outcome. This stands in contrast to a “physical externality,” which occurs when a water transfer negatively affects the quantity or quality of the physical resource available to other users. From the standpoint of economic theory, a transfer that results in a negative physical externality requires compensation. Compensation helps to ensure a socially efficient outcome by preventing overuse of the resource. By contrast, there are no efficiency grounds for compensating those affected by a pecuniary externality. Rather, the issue is one of equitable distribution of the benefits (Howe et al., 1990, Howitt, 1994). Lack of Legal Provisions for Mitigation The legal issues raised by the two types of effects are also quite distinct. There is a legal tradition for protecting third parties from the negative physical externalities associated with business decisions. In California, the no-injury provisions of state water law, established in case law as early as 1862, specifically aim to prevent negative physical effects on other water users, including wildlife.16 However, there is no clear legal tradition for protecting individuals from the effects on their livelihood of a change in other people’s business decisions—and this is what compensation for the economic effects of land fallowing implies. At the federal level, exceptions arise when a region or sector is negatively affected by a policy change considered beneficial for society as a whole. Notably, national programs have been available since the early 1960s to assist workers in industries affected by trade liberalization, and special regional programs were introduced in the 1970s to assist forestry workers affected by the expansion of national park areas in the western states.17 More generally, federal, state, and local governments provide ____________ 16The no-injury rule was established in the court case Butte T. M. Co. v. Morgan, 19 Cal. 609 (Gray, 1994a). The environmental protections for fish, wildlife, and instream beneficial uses were codified in Cal. Water Code Section 1738 in 1980. 17See Appendix F in Illingworth et al. (2002). 80 transitional assistance to workers facing negative economic conditions through unemployment insurance and other social programs. In California, the only legal provision for protecting communities from the economic effects of water transfers is found in the wheeling statute of the Water Code (Sections 1810–12), introduced in 1986. Under this statute, the transport of water through public conveyance facilities must be done “without unreasonably affecting the overall economy of the environment of the county from which the water is being transferred.” To date, these protections have not been invoked. In effect, it is difficult to demonstrate that fallowing programs pass the (admittedly vague) unreasonable effect test. Available studies suggest that fallowing programs would need to be considerably more extensive than those in recent years to have significant negative county-level effects (Table 6.1). The studies, which examine the effects of fallowing anywhere from 6 to 25 percent of a county’s irrigated farmland, find that it is likely to have no more than a 1 percent effect on overall county economic activity, even when the payments to farmers for the water transfers are excluded. This level was exceeded in only two counties—Colusa and Glenn—in a study simulating the effects of a 25 percent cut in surface water supplies (with no revenues for water sold). A 22 percent reduction in land farmed in the Palo Verde area under the test program had insignificant effects on income within that limited region, a result that is also anticipated even if farmers engage in maximum fallowing (29 percent of total acreage) under the upcoming long-term program. However, the same studies also demonstrate that there can be significant localized negative effects on individual farm workers and businesses and on local public agencies such as school districts. Thus, there may be ethical grounds for devising mitigation programs, even when a transfer does not trigger the legal requirement to do so. The case for mitigation is stronger when the transfer has negative distributional implications—a concentration of losses to low-income farm workers and processing plant laborers and accrual of most benefits to the relatively wealthy members of the community (or, in the case of absentee landlords, nonmembers). Given the structure of California’s agricultural economy, where wages are low, unemployment rates high, and large 81 Study Area and Subject 1991 drought water bank (11 counties) Table 6.1 Study Estimates of the Overall Economic Effect of Land Fallowing Authors Dixon et al. (1993) Acres and % of County Farmland Fallowed County Job Losses Regional or Sectoral Income Losses 2–3% of agricultural income Overall County Income Losses <1% 1991 drought water Howitt (1994) bank (Yolo and Solano) Solano: 23,500 acres (13%) Yolo: 45,700 acres (13%) 4.7% (model) 1.5% (survey) (farm job loss, both counties) Solano: 3.2% gross agricultural income Yolo: 5% gross agricultural income <1%a 82 1992–1993 Palo Verde Test Fallowing Program M. Cubed (1994) 20,215 acres (22% of region’s farmland) 1.3% regional job Negligible regional Negligible loss income effects 2003–2038 Palo Verde Long-Term Program M. Cubed (2002) 26,500 acres (29% of region’s farmland (program maximum) <2% regional job <1% regional loss income Negligible Irrigation water cuts Lee et al. (1999) in Sacramento Valley (8 counties)b 25% surface supply cut, no replacement with groundwater (4.5% to 6.2% fall in acreage) 305 jobs (8 counties) Colusa: 5% Glenn: 2.5% 5 other counties: 0.5% Sacramento: negligible Proposed Westlands Illingworth et al. 100,000 acres (6%) <1% (Fresno and 5 M&I to M&I Bay Area So. Cal. SWP 1998 Westside WD Colusa County WD 25,000 25 Agr to Agr Sac Valley Sac Valley CVP 1999 Byron–Bethany ID Alameda County FCWCD 5,000 15 Agr to M&I Bay Area Bay Area No 1999 Oakdale ID Stockton East WD 15,000 10 Agr to M&I SJ Valley SJ Valley No 1999 South San Joaquin ID Stockton East WD 15,000 10 Agr to M&I SJ Valley SJ Valley No 2000 Mercy Springs WD Westlands WD, Santa Clara Valley WD, and Pajaro Valley WMAa 6,260 25 Agr to All SJ Valley SJ Valley and CVP Bay Area 2000 Placer County WA Northridge WD 29,000 25 All to M&I Sac Valley Sac Valley No 2001 Kern County WA Western Hills WD 8,000 35 Agr to M&I SJ Valley SJ Valley No 2001 Merced ID U.S. Fish and Wildlife 12,500 12 Agr to Env SJ Valley SJ Valley No 2001 Oakdale ID U.S. Fish and Wildlife 15,000 12 Agr to Env SJ Valley SJ Valley No 2001 SJ River Group Authority U.S. Fish and Wildlife 110,000 12 Agr to Env SJ Valley SJ Valley No 2002 San Bernardino Valley MWD MWDSC 20,000 2002b Imperial ID Coachella Valley WD and MWDSC 100,000 2002b Imperial ID San Diego County WA 200,000 2002b Butte WD Madera ID and Root Creek WD 15,000 2002b Merced ID U.S. Fish and Wildlife 47,000 2002b Palo Verde ID MWDSC 111,000 2002b South San Joaquin ID Cities of Tracy, Escalon, Manteca, Lathrop 75,000 10 M&I to M&I So. Cal. So. Cal. 75 Agr to All So. Cal. So. Cal. 75 Agr to M&I So. Cal. So. Cal. 25 Agr to M&I Sac Valley SJ Valley 10 Agr to Env SJ Valley SJ Valley 35 Agr to M&I So. Cal. So. Cal. 25 Agr to M&I SJ Valley SJ Valley SWP CO River CO River No No CO River No NOTES: Abbreviations: WA = water agency or water authority; WMA = water management agency; FCWDC = flood control and water conservation district; MWD= municipal water district; M&I = municipal and industrial; Agr = agriculture; Env = environment. “U.S. Fish and Wildlife” as buyer indicates purchase by USBR’s Water Acquisition Program for environmental uses; SJ River Group Authority includes Merced ID, Modesto ID, South San Joaquin ID, Oakdale ID, SJ River Exchange Contractors, and Friant Water Users Association. aWestlands WD and Santa Clara Valley WD will receive deliveries until Pajaro Valley WMA completes construction of facilities for the permanent transfer of Mercy Springs WD entitlement. bPending approval in 2002. Table A.6 Permanent Transfers of Surface Water Since 1985 137 Year Seller Buyer Annual Acre- Region of Region of Within Feet (Max) Purpose Origin Destination Project 1998 1998 1998 1998 2000 2000 2000 2000 2001 2001 2001 2001 2002 2002 2002a 2002a 2002a 2002a Corning WD Kern County WA Proberta WD Thomes Creek WD Kern County WA Kern County WA Kern County WA Kern County WA Kern County WA Kern County WA Kern County WA Olcese WD and private ranch Tulare Lake Basin WSD Tulare Lake Basin WSD Banta Carbona ID Lower Tule River ID Mercy Springs WD The Westside ID U.S. Fish and Wildlife Mojave WA U.S. Fish and Wildlife U.S. Fish and Wildlife Alameda County FCWCD Alameda County FCWCD Castaic Lake WA Palmdale WD Alameda County FCWCD Napa County FCWCD Solano County WA Kern County WA Antelope Valley–East Kern WA Dudley Ridge WD City of Tracy City of Orange Cove Westlands WD City of Tracy 2,300 25,000 2,000 2,000 15,000 7,000 41,000 4,000 10,000 4,025 5,756 50,000b 3,000 3,973 5,000 2,000 1,071 5,000 Agr to Env Agr to M&I Agr to Env Agr to Env Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to M&I Agr to Agr Agr to M&I Agr to M&I Agr to Agr Agr to M&I Sac Valley SJ Valley Sac Valley Sac Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley Sac Valley So. Cal. Sac Valley Sac Valley Bay Area Bay Area So. Cal. So. Cal. Bay Area Bay Area Bay Area SJ Valley So. Cal. SJ Valley SJ Valley SJ Valley SJ Valley SJ Valley CVP SWP CVP CVP SWP SWP SWP SWP SWP SWP SWP No SWP SWP CVP CVP CVP CVP NOTES: Abbreviations: WA = water agency or water authority; WMA = water management agency; FCWDC = flood control and water conservation district; MWD= municipal water district; M&I = municipal and industrial; Agr = agriculture; Env = environment. “U.S. Fish and Wildlife” as buyer indicates purchase by USBR’s Water Acquisition Program for environmental uses; SJ River Group Authority includes Merced ID, Modesto ID, South San Joaquin ID, Oakdale ID, SJ River Exchange Contractors, and Friant Water Users Association aPending approval in 2002. bAverage annual volumes. Appendix B Groundwater Institutions and Basins Table B.1 Counties with Groundwater Protection Ordinances, by Region Export Restrictions Mountain Region Calaveras (2002) Inyo (1980) Lassen (1999) Modoc (1978) Mono (1988; 1998) Nevada (1986–1988 only) Sierra (1977; 1997) Siskiyou (1998) Tuolumne (2001) Sacramento Valley Butte (1977; 1996) Colusa (1998) Glenn (1977; 1990) Sacramento (1980) Shasta (1997) Tehama (1992) Yolo (1996) San Joaquin Valley Fresno (2000) Kern (1998) Madera (1999) San Joaquin (1996) North Coast On-Site Groundwater Use Restrictions No Ordinance Alpine Amador El Dorado Mariposa Plumas Placer Sutter Yuba Mendocino (1995) Kings Merced Stanislaus Tulare Del Norte Humboldt 139 Table B.1 (continued) Export Restrictions San Francisco Bay and Central Coast Regions Lake (1999) San Benito (1995) On-Site Groundwater Use Restrictions No Ordinance Monterey (1993) Napa (1999) Alameda Contra Costa Marin Santa Barbara Santa Clara Santa Cruz San Francisco San Luis Obispo San Mateo Solano Sonoma Southern California Imperial (1996) San Diego (1991) San Bernardino (2002) Los Angeles Orange Riverside Ventura NOTE: Year in parentheses is year of adoption. If two dates are listed, the first refers to the adoption of an urgency ordinance and the second to the adoption of a regular ordinance. Many ordinances have been revised at least once subsequently. Imperial County adopted an ordinance requiring conditional use permits for some within-county groundwater uses in 1972 and added explicit export restrictions in 1996. 140 Del Norte Siskiyou Modoc Trinity Humboldt Shasta Lassen Tehama Plumas Yuba Mendocino Glenn Butte Sierra Sutter Colusa Lake Placer Nevada Sonoma Napa Yolo El Dorado Amador Alpine Sacramento Marin Solano Calaveras San Francisco Contra San Costa Joaquin Tuolumne Mono San Mateo Alameda Santa Stanislaus Mariposa Clara Merced Santa Cruz Madera San Benito Monterey Fresno Kings Tulare Inyo San Luis Obispo Kern Santa Barbara Ventura Los Angeles San Bernardino Orange Riverside San Diego Imperial Figure B.1—California’s Counties 141 1. Raymond Basin (1944) 2. Cucamonga Basin (1958, management 14 structure being updated) 3. West Coast Basin (1961) 4. Central Basin (1965) 5. Santa Margarita River Watershed (1966) 6. San Bernardino Basin Area (1969) 7. Brite Basin (1970) 8. Cummings Basin (1972) 9. Tehachapi Basin (1973) 10. Main San Gabriel Basin (1973) and Puente Narrows (1972) 11. Warren Valley Basin (1977) 12. Chino Basin (1978) 13. Upper Los Angeles River Area (1979) 14. Scott River Stream System (1980) 15. Puente Basin (1985) 16. Mojave Basin Area Adjudication (1996) 17. Santa Paula Basin (1996) 18. Six Basins (1998) 7, 8, 9 16 1 10,15 17 2, 6 13 3 4 12,18 5 11 SOURCES: Adjudication dates—Department of Water Resources (2001). Basin contours—Department of Water Resources (2002a). NOTES: The basins illustrated are the full basin boundaries as determined by DWR. Actual adjudicated basin boundaries are defined by the court and are often smaller. The dates indicate the year of final adjudication. Figure B.2—Adjudicated Groundwater Basins 142 Special Groundwater Management Districts 11 1. Monterey Peninsula Water Management District (1977) 2. Long Valley Groundwater Management District (1980) 3. Sierra Valley Groundwater Management District (1980) 10 4. Fox Canyon Groundwater Management Agency (1982) 7 5. Pajaro Valley Water Management Agency (1984) 6. Mendocino City Community Services District (1987) 7. Honey Lake Groundwater Management Agency (1989) 8. Tri-Valley Groundwater Management District (1989) 9. Ojai Groundwater Management Agency (1991) 10. Willow Creek Groundwater Management Agency 6 2 (1993; inactive) 3 11. Surprise Valley Groundwater Management Agency (1995; inactive) Special Water Districts with Groundwater Control A. Santa Clara Valley Water District B. Orange County Water District 8 C. Coachella Valley Water District A 5 1 9 4 B C SOURCES: Agency names and dates are from the Department of Water Resources (1996a); basin contours are from the Department of Water Resources (2002a); agency contours are from the U.S. Bureau of Reclamation (2002). NOTES: Special Water Districts and Pajaro Valley Water Management Agency are represented by administrative district boundaries (U.S. Bureau of Reclamation, 2002). Other district boundaries are represented by underlying groundwater basin boundaries (Department of Water Resources, 2002a). The following district boundaries were estimated using corresponding basins: Monterey Peninsula Water Management District (Seaside Area subbasin of Salinas Valley and Carmel Valley), Fox Canyon Groundwater Management Agency (Oxnard subbasin of Santa Clara River Valley, Pleasant Valley, Arroyo Santa Rosa Valley, and Las Posas Valley), Mendocino City Community Services District (Fort Bragg Terrace Area), and Tri-Valley Groundwater Management District (Eastern Mono County portion of Owens Valley). Figure B.3—Special Groundwater Management Districts 143 A B C Basins Subject to Critical Conditions of Overdraft 1. Santa Cruz Pajaro Basin 2. Cuyama Valley Basin 3. Ventura Central Basin 4. Eastern San Joaquin County Basin 5. Chowchilla Basin 6. Madera Basin 7. Kings Basin 8. Kaweah Basin 9. Tulare Lake Basin 10. Tule Basin 11. Kern County Basin Basins with Special Problems A. Surprise Valley Basin 4 B. Long Valley Basin C. Sierra Valley Basin D. Owens Valley Basin 5 6 17 8 9 10 D 11 2 3 NOTES: The basins illustrated are those identified in Bulletin 118-80 (Department of Water Resources,1980), mapped using boundaries appearing in the 2002 Draft Groundwater Map (Department of Water Resources, 2002a). Owing to name or boundary changes, the following 1980 basin boundaries were estimated using the 2002 definitions: Santa Cruz-Pajaro Basin (Santa Cruz Purisima Formation; Pajaro), Ventura Central Basin (Piru, Fillmore, Santa Paula, Mound, and Oxnard subbasins of Santa Clara River Valley; Pleasant Valley; Arroyo Santa Rosa Valley; and Las Posas Valley). Figure B.4—Critically Overdrafted and Special Problem Groundwater Basins Listed in Bulletin 118-80 144 Appendix C Predicting County Adoption of Export Restrictions This appendix describes the data sources, estimation methods, and results of the statistical analysis of the likelihood of county adoption of export restrictions reported in Chapter 4. Data Sources Farm and Agriculture-Related Jobs The source for farm jobs is the Bureau of Economic Analysis regional accounts data. The source for agriculture-related jobs is the Bureau of the Census “County Business Patterns.” Data are from 1995. Share of Irrigated Agriculture in Total Farmland This series is derived from results of the 1997 Agricultural Census (U.S. Department of Agriculture and the California Department of Food and Agriculture). 1997 data were preferred over those from 1992, a year with unusually low acreage figures as a consequence of the prolonged drought. Share of Residential Population Dependent on Groundwater This series was approximated using information from county environmental health officers and a 1992 survey by the Water Education Foundation (1994). Other Variables Membership in the Regional Council of Rural Counties and presence of a critical or specially designated groundwater basin are presented in Chapter 4. 145 Estimation Methods and Results Probit regressions were used to estimate the effect on a county’s likelihood of adopting an export ordinance of the variables presented above, for which sample characteristics are reported in Table 4.1. The statistical exercise involves considering the effect of each variable on ordinance adoption, while holding the other variables constant. Table C.1 reports the results of two regressions, one including the full set of variables and one excluding residential groundwater share. The variables used in each model are jointly significant at the 99 percent level of confidence. The second model produces slightly tighter coefficient estimates because groundwater share, itself insignificant, is correlated with the measure of critical basins. This exclusion does not affect overall model fit. Both models slightly underpredict adoption of export restrictions: compared to the 22 actual cases (38 percent), the models predict adoption by 20 counties (34 percent). Most of the variables are marginally significant (at the 90 percent level of confidence), with coefficients of the expected sign. The effects of each variable on the likelihood of ordinance adoption are presented in Chapter 4. Because some of the counties in the nonadoption group have other types of more comprehensive groundwater management systems in place—including adjudicated basins, special districts, or groundwater protection ordinances that control local groundwater use—we also performed two tests to see whether the results changed significantly when controlling for this factor: a multinomial logit regression distinguishing among three possible outcomes (export restrictions, comprehensive groundwater management systems, and no groundwater rules), and a binomial probit with a control variable for counties with these types of groundwater systems. The additional “groundwater management” group for the multinomial logit included El Dorado, Los Angeles, Mendocino, Monterey, Napa, Orange, Plumas, Riverside, San Bernardino, Santa Clara, Santa Cruz, San Diego, and Ventura. The groundwater management control variable in the binomial probit included this group plus counties with dual systems: Imperial, Lassen, Modoc, Mono, San Benito, Sierra, and Siskiyou. In neither case was there a substantial 146 Table C.1 Effects of County Characteristics on the Probability of Adopting an Export Restriction Farm employment (%) Agriculture-related employment (%) Irrigated farmland (%) Residents using groundwater (%) Counties overlying critical/special basin RCRC membership Chi-squared test of joint significance of variables in model Log-likelihood Observed probability Predicted probability Model Including Residential Groundwater 0.045* (0.026) –0.13* (0.077) 0.004 (0.003) 0.001 (0.003) 0.23 (0.17) 0.27* (0.16) Model Excluding Residential Groundwater 0.048* (0.025) –0.12* (0.075) 0.004 (0.003) — 0.26* (0.16) 0.30* (0.15) 20.88*** –28.06 38% 34% 20.72*** –28.13 38% 34% NOTES: Coefficients are reported as marginal effects. For the binary variables (critical groundwater basins and RCRC membership), the coefficient is the effect of a discrete change from 0 to 1. Standard errors are reported in parentheses. ***Indicates coefficient significantly different from zero at the 99 percent level of confidence. *Indicates coefficient significantly different from zero at the 90 percent level of confidence. change in coefficient estimates in relation to those reported in Table C.1, although the loss of degrees of freedom in the multinomial logit reduces levels of significance. 147 Appendix D Measuring the Effect of Export Restrictions on County Water Sales This appendix provides detailed information on the data sources, estimation methods, and results of the statistical analysis of the effect of export restrictions on county water sales and exports presented in Chapter 5. Data Sources Annual County Water Sales and Annual County Exports These series are developed from the water transfer database presented in Chapter 2 and Appendix A. Annual county sales are defined as the sum of all short- and long-term transfers by county water users in a given year. For sales by cross-jurisdictional water districts, the approximate share of the district in each county has been attributed to that county. Annual county exports are the sum of transfers not destined for other water users within the county. Environmental water sales were considered as exports from the county. Although this water is most often used for habitat or instream purposes within the region, it rarely is under control of users in the county of origin. For water districts with multiple jurisdictions, we considered the transaction to be “in-county” if the purchaser was in any of the district’s counties. As such, the exports category unambiguously includes only those transfers going from a user within the county to a user somewhere else in the region or state. Because the coverage of transfers within the CVP’s Friant group was not consistently available in all years, we have excluded internal Friant transactions from the sales data. This concerns five San Joaquin Valley counties: Merced, Madera, Fresno, Tulare, and Kern. Since members of the Friant group are effectively exempted in the counties with ordinances, this should not pose a problem for interpretation of the 149 results. Trades between Friant members and other water users are included. County Export Restrictions We consider that counties have an export ordinance in operation beginning in the year of adoption, as indicated in Table C.1. There are two exceptions. Because Kern County’s ordinance applies only to the relatively unimportant desert and foothill region in the southeast, but not to the San Joaquin valley portion of the county where population, agriculture, and surface water entitlements are all concentrated, we have considered Kern to have no ordinance for the purposes of this exercise. The second exception is made for Glenn County, whose export ordinance was effectively removed in 2000, when the new basin management ordinance was adopted.1 The 2001 season is the first during which Glenn water users worked under the new system. State and Federal Policy Environment The general effects of an improved trading environment arising from state and federal policies to facilitate transfers are captured by a time trend. Agricultural Water Demand The model uses three measures of agricultural water demand: average county-level prices for annual crops (defined as all field and horticultural crops), the acreage under annual crops, and the share of perennial crops in total nonrange acreage. All three series are constructed using county agricultural statistics from the California Agricultural Statistics Service databases.2 ____________ 1Although the 1990 ordinance remains on the books, the numerous persons interviewed in Glenn, including two county supervisors, considered that the new ordinance has supplanted it for operational purposes. 2The annual crop price is calculated using the county’s prior year output data, valued at the statewide average price for the current year. This captures the notion that the farmer has an idea of the average market price for the coming season and can calculate what he would earn by farming the same crop mix as in the preceding year. The series is deflated using the western states urban consumer price index, with 1992 as the base year. Ideally, we would measure the value of crops on a per-acre-foot basis to capture the water 150 In principle, we would expect water sales to be inversely related to the average level of crop prices, which reflect the value of using water in agriculture. On average, real price levels have been relatively flat over the period, hovering around $160 per ton in 1992 prices. The range across counties is quite large, however, with averages over $300 per ton along the south and central coast and $100 per ton or less in parts of the Central Valley. As with prices, there has been little movement over time in the average level of annual crop acreages, although the cross-county differences are huge, with at least several hundred thousand acres in Imperial County and most San Joaquin Valley counties (and close to 1 million acres in Fresno), and fewer than 50,000 acres along the coast.3 In part, this range reflects differences in the overall scale of agricultural operations across counties; in part, it reflects a much higher share of perennial crops (fruit trees, nut trees, and vineyards) in some counties. The values span a high of over 90 percent of all nonrange farmland in Napa and over 50 percent in San Diego, Ventura, and Madera, to only 1 percent of cropland in Imperial. Over time, there has been a mild upward trend in tree crops as a share of the total, moving from 22 to 26 percent on average. Because farmers can make adjustments in annual crop acreages fairly easily as a function of water availability, we would expect water sales to be positively related a county’s crop acreage. Conversely, because a higher share of tree crops in total acreage introduces less flexibility in water use, we would expect tree crop share to be negatively related to water sales. ________________________________________________________ intensity of the crop mix. This would require making assumptions about the irrigation technology used in each county for each crop, however. If anything, the use of a per-ton measure probably dampens the effect of this variable, since low-value crops also tend to be those with a relatively high level of water use. Annual crop acreage includes all farm acreage except perennials and rangeland (i.e., including irrigated pasture). The share of tree crops is calculated as the share of perennials in total nonrangeland farm acreage. Both acreage measures are valued at the prior year levels to account for the fact that decisions on water sales are generally made before final planting decisions. 3San Francisco, the only county in the sample with no commercial agriculture (albeit some fine gardens), has no acreage recorded and no positive crop prices. 151 Residential Water Demand County population levels are used to account for residential water demands. The source is the annual population series from the California Department of Finance, based on updates from the 2000 Census (Department of Finance, 2001). Other things equal, we should expect counties with higher populations to be less likely to sell water. Water Supply Conditions Annual deliveries of project water from the CVP, the SWP, and the Colorado River Project are captured in two measures: senior rights and junior rights.4 The senior rights category includes those deliveries with a high degree of reliability, by virtue of the seniority of the contractors. This includes the CVP settlement contractors in the Sacramento Valley and the exchange contractors in the San Joaquin Valley, the SWP’s Feather River contractors in Butte and Sutter Counties, and the Colorado River contractors in Southern California. On average, 8.3 million acre-feet are delivered annually to these contractors. Half of this volume is destined to Imperial and Riverside Counties and a quarter to the four main settlement-contracting counties in the Sacramento Valley (Colusa, Butte, Sutter, and Glenn). The only dips in supply occurred during the early 1990s drought, when CVP and SWP contractors’ deliveries were reduced by 25 to 50 percent in some years. The junior-rights category includes the ordinary project contractors of the CVP and the SWP. On average, these projects have delivered just over 6 million acre-feet annually over the 12-year period, to a much larger number of water users. Only two counties in the sample, San Francisco and Yuba, do not have project contractors. Project deliveries have generally been much more variable from one year to the next, particularly for contractors south of the Delta. ____________ 4For the Colorado River Project and the SWP, actual delivery data were used. For the CVP, we applied the annual allocation rules by type of project contractor (settlement contractors, north of Delta and south of Delta service contractors, and Friant class 1 and 2 contractors). Because California received surplus deliveries of Colorado River water for the entire time period under consideration, the Metropolitan Water District of Southern California is included as a senior rights-holder. Under California’s official allocation of 4.4 million acre-feet, this agency would have its supplies cut back, as the junior rightsholder. 152 In general, we would expect counties with higher water deliveries to be more active in the water market. By the same token, individual counties should be more likely to sell in years when their deliveries are higher. Unfortunately, detailed data on other water supplies—from autonomous projects and from groundwater—are not available. We do have a general indicator of the quality of the water year, however, in the form of the state’s most important rainfall measure—the Sacramento Valley 40-30-30 index.5 Since market demands and water prices are likely to be higher in dry years, we would expect this indicator to be negatively related to sales. The period under review contains an equal number of dry and wet years (Figure 2.1). Sample Characteristics Table D.1 provides summary statistics for the complete set of variables used for both geographical samples used in the estimations: the 34 water-trading counties6 and the 18 Central Valley counties.7 We have excluded 24 nontrading counties from the first group for statistical reasons. Two key econometric models cannot be estimated when these counties are included. The inclusion of counties that never trade adds no information to the estimation of the effects of an export ordinance on trading behavior in a fixed-effects mode. The presence of many counties with all zero trades also complicates the estimation of a random-effects Tobit model. For the state as a whole, the convergence properties of this ____________ 5The Sacramento Valley 40-30-30 Index is the main index used by DWR to measure water conditions in the Sacramento Valley, source region for both the CVP and the SWP as well as several large local projects. It is computed as a weighted average of the current water year’s April–July unimpaired runoff forecast (40 percent), the current water year’s October–March unimpaired runoff forecast (30 percent), and the previous water year’s index (30 percent). For details, see DWR’s website, http://watsup2.water.ca.gov/ hydrologic.cfm. 6The sample excludes the 24 counties for which there are no transactions records in any of the 12 years: Alpine, Amador, Calaveras, Del Norte, El Dorado, Humboldt, Inyo, Lake, Lassen, Marin, Mariposa, Mendocino, Modoc, Mono, Monterey, Nevada, Plumas, San Mateo, Santa Cruz, Sierra, Siskiyou, Sonoma, Trinity, and Tuolumne. 7Butte, Colusa, Fresno, Glenn, Kern, Kings, Madera, Merced, Placer, Sacramento, San Joaquin, Shasta, Stanislaus, Sutter, Tehama, Tulare, Yolo, and Yuba. 153 Table D.1 Summary Statistics for Annual County Water Sales and Water Exports, 1990–2001 All sales (acre-feet) Observed probability of sales Out-of-county exports (acre-feet) Observed probability of exports 34 Trading 18 Central Counties Valley Counties 22,734 31,461 (40,658) (45,842) 0.69 0.89 19,465 (39,075) 0.61 25,419 (44,140) 0.82 Agricultural and residential demand Annual crop prices ($/ton) 162 (126) Annual crop area (acres) 205,893 (226,721) Tree crop area in total (%) 23.2 (19.7) Population 856,177 (1,602,147) 117 (51) 294,918 (230,010) 22.6 (13.9) 286,226 (292,132) Water supply conditions Project deliveries (acre-feet) Senior rights Junior rights Rainfall index 243,826 (537,918) 179,474 (272,842) 8.18 (3.18) 178,751 (210,645) 247,412 (346,500) 8.18 (3.18) County export restrictions (1 = restriction) 0.19 (0.39) 0.30 (0.46) Number of observations 408 216 NOTE: The table reports mean values with standard deviations in parentheses. model are not stable. However, regression results on the full state sample are consistent with the findings we report below on the effects of the export ordinances and other key variables for the 34-county sample. 154 Regression Results A random-effects Tobit model is the main specification used to account for bunching at zero in the distribution of the dependent variable.8 Because it is not straightforward to test for fixed effects using this specification (Arellano and Honoré, 2001), we did so with a linear model. Tables D.2 through D.5 present the results for the randomeffects Tobit model and the corresponding random- and fixed-effects linear models for all sales and exports for the water-trading counties and Central Valley counties, respectively. Despite the censoring in the dependent variable, the results of the random effects linear model correspond closely to the Tobit model results, suggesting that reliance on the linear model for inferring properties about the Tobit model is reasonable. For all four models, Hausman specification tests of the linear model fail to reject the null hypothesis of no fixed effects at conventional levels of significance, suggesting the absence of crosssectional omitted variables that would bias the random-effects Tobit results. Consequently, the discussion in Chapter 5 focuses on the results of the random-effects Tobit model. Overall, the identified variables have the anticipated effects on both sales and exports. Among the control variables, the group capturing the effects of water supply is highly significant. Crop prices and annual crop acreage, two of the agricultural demand variables, are significant for the determination of sales within the full 34-county sample. These factors are not significant for the Central Valley sample, where there is less crosscounty variability. Results pertaining to the key variables of interest—county export restrictions and the time trend used to capture the effect of an improved trading environment—are presented in Chapter 5. The cumulative market effect of ordinances (Figure 5.2) was calculated by multiplying the number of counties affected by the per-county coefficients listed in ____________ 8For the 34-county sample of water trading counties, 31 percent of all sales and 39 percent of all exports are zero; for the 18-county Central Valley sample, the corresponding values are 11 percent of all sales and 18 percent of all exports. 155 the left-hand column of Tables D.2 through D.5. For the 34-county sample, the number of counties with ordinances ranges from two in the early 1990s to 12 in 2000. For the 18-county sample, the range is from two in 1990 to 10 in 2000. The cumulative market effect of state and federal policies (Figure 5.3) was calculated by multiplying the total number of counties in the sample (34) by the coefficient on the time trend reported in the left-hand column of Table D.2. 156 Table D.2 Determinants of Annual County Water Sales in 34 Water-Trading Counties, 1990–2001 Random-Effects Fixed-Effects Random-Effects Linear Linear Tobit Regression Regression Agricultural and residential demand Annual crop prices ($/ton) –75.8** –40.8 –55.6 (33.0) (25.9) (52.5) Annual crop area (1,000 acres) 54.6** 37.0* 143 (23.1) (19.7) (88.4) Tree crop area in total (%) –96.1 –10.2 213 (220) (176) (680) Population (1,000) –1.5 –1.1 –0.8 (2.6) (2.3) (32.4) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index 23.8*** (7.4) 49.2*** (13.1) –3,529*** (749) 21.5*** (7.2) 42*** (10.6) –2,160*** (549) –140** (56.6) 51*** (12.5) –2,099*** (575) State and local institutional factors County export restrictions State and federal policy (time trend) –14,308** (7,246) 3,828*** (681) –12,671** (5,563) 2,442*** (515) –10,770* (6,387) 2,616*** (720) Log-likelihood; overall R2 Hausman specification test –3,419 0.34 0.00 0.33 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 157 Table D.3 Determinants of Annual County Water Exports in 34 Water-Trading Counties, 1990–2001 Random-Effects Fixed-Effects Random-Effects Linear Linear Tobit Regression Regression Agricultural and residential demand Annual crop prices ($/ton) –88.9** –45.0* –59.2 (38.4) (27.0) (52.4) Annual crop area (1,000 acres) 37.6 15.7 97.1 (27.2) (20.6) (88.3) Tree crop area in total (%) –113 –17.9 215 (254) (185) (679) Population (1,000) –2.0 –1.2 8.0 (3.0) (2.5) (32.4) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index 28.3*** (8.0) 43.2*** (14.3) –3,235*** (831) 24.4*** (7.6) 34.5*** (11.0) –1,909*** (551) –160*** (56.6) 44.7*** (12.5) –1,796*** (575) State and local institutional factors County export restrictions State and federal policy (time trend) –16,948** (7,722) 3,729*** (761) –16,276*** (5,632) 2,242*** (517) –13,875** (6,378) 2,321*** (719) Log-likelihood; overall R2 Hausman specification test –3,062 0.26 0.05 0.15 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 158 Table D.4 Determinants of Annual County Water Sales in 18 Central Valley Counties, 1990–2001 Agricultural and residential demand Annual crop prices ($/ton) Annual crop area (1,000 acres) Tree crop area in total (%) Population (1,000) Random-Effects Fixed-Effects Random- Linear Linear Effects Tobit Regression Regression –71.8 (87.7) 11.0 (28.8) 289 (364) –18.5 (21.6) –57.9 (91.1) 1.0 (31.4) 118 (421) –7.4 (24.2) –87.0 (129.3) 224* (130) –278 (1,659) 163 (166) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index –7.2 (25.6) 64.0*** (15.4) –5,410*** (1,049) –16.2 (28.7) 60.0*** (15.0) –4,482*** (969) –116 (75.8) 65.4*** (17.3) –4,639*** (1,037) State and local institutional factors County export restrictions State and federal policy (time trend) –20,789** (8,713) 4,645*** (977) –19,038** (8,397) 4,094*** (934) –13,034 (9,309) 3,412** (1,341) Log-likelihood; overall R2 Hausman specification test –2,321 0.27 0.11 0.58 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 159 Table D.5 Determinants of Annual County Water Exports in 18 Central Valley Counties, 1990–2001 Random-Effects Fixed-Effects Random-Effects Linear Linear Tobit Regression Regression Agricultural and residential demand Annual crop prices ($/ton) –139 –77.7 –105.6 (102) (92.0) (130) Annual crop area (1,000 acres) –30.7 –26.6 166 (32.1) (31.7) (130) Tree crop area in total (%) 409 172 –559 (408) (427) (1,665) Population (1,000) –10.3 –3.5 245 (23.7) (24) (167) Water supply conditions Project deliveries (1,000 acre-feet) Senior rights Junior rights Rainfall index –10.9 (29.0) 58.6*** (16.4) –4,661*** (1,109) –26.4 (29) 51.5*** (15.1) –3,963*** (975) –126* (76) 54.7*** (17.4) –4,072*** (1,041) State and local institutional factors County export restrictions State and federal policy (time trend) –26,245*** (9,350) 4,220*** (1,039) –23.481*** (8,450) 3,659*** (940) –16,902* (9,343) 2,699** (1,346) Log-likelihood; overall R2 Hausman specification test –2,173 0.20 0.04 0.52 NOTES: All models are estimated with a constant. Standard errors are in parentheses. The Hausman specification test reports the probability that the difference in coefficients of random-effects and fixed-effects regressions is not systematic. ***Indicates coefficient is significantly different from zero at the 99 percent level of confidence in a two-way test. **Indicates coefficient is significantly different from zero at the 95 percent level of confidence in a two-way test. *Indicates coefficient is significantly different from zero at the 90 percent level of confidence in a two-way test. 160 References Alameda Flood Control and Water Conservation District Zone-7, Negative Declaration for the Five Year Water Transfer Agreement with Byron-Bethany Irrigation District, Pleasanton, California, July 1994. Arellano, Manuel, and Bo Honoré, “Panel Data Models: Some Recent Developments,” in J. J. Heckman and E. Leamer (eds.), Handbook of Econometrics, Volume 5, Elsevier Science B.V., Amsterdam, 2001. Association of California Water Agencies, Water Supply and Development: A User’s Guide to California Statutes Including SB 221 (Kuehl) and SB 610 (Costa), prepared by McCormick, Kidman and Behrens, Sacramento, California, 2002. Brown, Judy, Toccoy Dudley, and Allan Fulton, Locally Governed Groundwater Management in Glenn County, California, Water Advisory Committee, Glenn County, California, October 2001. Bunn, Thomas S., III, “Counties Have a Stake in Managing Groundwater,” California County, July/August 1997, pp. 10–13. Campbell, Christopher L., “County Groundwater Regulation: A Case Study of Fresno and Madera Counties,” ACWA Continuing Legal Education Workshop for Water Attorneys, Costa Mesa, California, October 2000. CH2MHill, Final Initial Study/Environmental Assessment for the BantaCarbona Irrigation District/City of Tracy Water Assignment Project, prepared for the Banta-Carbona Irrigation District, City of Tracy, and U.S. Bureau of Reclamation, September 2002a. CH2MHill, Final Initial Study/Environmental Assessment for the West Side Irrigation District/City of Tracy Water Assignment Project, prepared for The West Side Irrigation District, City of Tracy, and U.S. Bureau of Reclamation, September 2002b. 161 Coase, R. H., “The Problem of Social Cost,” Journal of Law and Economics, Vol. 3, October 1960, pp. 1–44. Conaughton, Gig, “Water Transfer Still Breathing,” North County Times, December 13, 2002. Cooper, Audrey, “Tracy May Buy Farmers’ Water: Irrigation Districts OK Proposal to Sell to City,” San Joaquin Record, September 12, 2002. Department of Finance, Interim County Population Projections, Sacramento, California, June 2001. Department of Water Resources, Ground Water Basins in California (Bulletin 118-80), Sacramento, California, January 1980. Department of Water Resources, Water Facts: Groundwater Management Districts or Agencies in California, Sacramento, California, 1996a. Department of Water Resources, State Water Project Supplemental Purchase Program: Draft Environmental Impact Report, Sacramento, California, 1996b. Department of Water Resources, California Water Plan Update (Bulletin 160-98), Sacramento, California, November 1998. Department of Water Resources, Water Facts: Adjudicated Basins in California, No. 3, Sacramento, California, January 2001. Department of Water Resources, Bulletin 118 Statewide Groundwater Basin Map, Version 2, Sacramento, California, May 3, 2002a. Department of Water Resources, Draft Guidebook for Implementation of Senate Bill 610 and Senate Bill 221 of 2001, Sacramento, California, September 2002b. Department of Water Resources, “DWR Announces 2003 Dry Year Water Purchase Program,” news release, Sacramento, California, November 15, 2002c. Dixon, L. S., N. Y. Moore, and S. W. Schechter, California’s 1991 Drought Water Bank: Economic Impacts in the Selling Regions, RAND, MR-301-CDWR/RC, Santa Monica, California, 1993. 162 Dudley, Toccoy, “Basin Management Objective (BMO) Method of Groundwater Basin Management,” September 2000, reprinted in Glenn County, Basin Management Objective (BMO) for Groundwater Surface Elevations in Glenn County, California, Willows, California, August 2001. Economic Research Service, U.S. Department of Agriculture, Rice Outlook, Washington, D.C., December 11, 2002. Fulton, Allan, and Toccoy Dudley, Frequently Asked Questions about Glenn County Groundwater Ordinance No. 1115: Its Background and Intent, Organizational Structure, and Status, Water Advisory Committee, Glenn County, Willows, California, n.d. Getches, David, Water Law in a Nutshell, 3rd edition, West Publishing Co., Eagan, Minnesota, 1997. Glenn County, Basin Management Objective (BMO) for Groundwater Surface Elevations in Glenn County, California, Glenn County Board of Supervisors, Willows, California, August 21, 2001. Goldsmith, Janet K., “Counties, Cities May Regulate Groundwater, Appellate Court Rules,” California Water Law and Policy, February 1995a, pp. 42–43. Goldsmith, Janet K., “It’s Official—Counties Can Regulate Groundwater,” California Water Law and Policy, May 1995b, p. 163. Governor’s Commission to Review California’s Water Rights Law, Final Report, Sacramento, California, 1978. Gray, Brian, “The Modern Era in California Water Law,” Hastings Law Journal, Vol. 45, No. 249, January 1994a. Gray, Brian, “The Role of Laws and Institutions in California’s 1991 Water Bank,” in H. O. Carter, H. J. Vaux, Jr., and A. F. Scheuring (eds.), Sharing Scarcity: Gainers and Losers in Water Marketing, Agricultural Issues Center, University of California, Davis, June 1994b. Gray, Brian, “A Model Water Transfer Act for California,” Journal of Environmental Law and Policy, Vol. 4, Fall 1996, pp. 3–22. 163 Hanak, Ellen, and Caitlin Dyckman, “Counties Wresting Control: Local Responses to California’s Statewide Water Market,” University of Denver Water Law Review, Vol. 6, No. 2, Summer 2003. Howe, C. W., J. K. Lazo, and K. Weber, “The Economic Impacts of Agriculture-to-Urban Water Transfers on the Area of Origin: A Case Study of the Arkansas River Valley in Colorado,” American Journal of Agricultural Economics, Vol. 72, No. 5, December 1990, pp. 1200– 1204. Howe, C. W., “Protecting Public Values in a Water Market Setting: Improving Water Markets to Increase Economic Efficiency and Equity,” University of Denver Water Law Review, Vol. 3, No. 2, Spring 2000. Howitt, R., “Effects of Water Marketing on the Farm Economy,” in H. O. Carter, H. J. Vaux, Jr., and A. F. Scheuring (eds.), Sharing Scarcity: Gainers and Losers in Water Marketing, Agricultural Issues Center, University of California, Davis, June 1994. Hundley, N. H., The Great Thirst. Californians and Water: A History, revised edition, University of California Press, Berkeley, 2001. Illingworth, W., et al., “Analysis of Economic Impacts of Proposed Land Retirement in Westlands Water District,” draft prepared for Westlands Water District, September 2002. Imperial Irrigation District and U.S. Bureau of Reclamation, Water Conservation and Transfer Project Draft Habitat Conservation Plan and Draft Environmental Impact Report/Environmental Impact Statement, prepared by CH2MHill, El Centro, California, January 2002. Johnson, Hans P., “A State of Diversity: Demographic Trends in California’s Regions,” California Counts, Vol. 3, No. 5, Public Policy Institute of California, May 2002. Johnson, Lyn, “Water Deal One Step Closer, Maybe. . . . ,” Palo Verde Valley Times, November 1, 2002. Kasindorf, Martin, “War Over Water Splits California Cities, Farms,” USA Today, October 2, 2002. 164 Kasler, Dale, “Powerful Forces Are Bearing Down on Imperial Valley Farmers, Who No Longer Can Ignore. . . . The Call for Water,” Sacramento Bee, September 28, 2002. Kern County Water Agency, Water Supply Report 1998, March 2002. Knapp, Keith, Marca Weinberg, Richard Howitt, and Judith Posnikoff, “Water Transfers, Agriculture, and Groundwater Management: A Dynamic Economic Analysis,” Journal of Environmental Management, forthcoming. Lee, Hyunok, Daniel Sumner, and Richard Howitt, Economic Impacts of Irrigation Water Cuts in the Sacramento Valley, Agricultural Issues Center, University of California, Davis, 1999. Loomis, John, “Water Transfer and Major Environmental Provisions of the Central Valley Project Improvement Act: A Preliminary Economic Evaluation,” Water Resources Research, Vol. 30, No. 6, June 1994, pp. 1865–1871. Lund, Jay R., Morris Israel, and Richard Kanazawa, Recent California Water Transfers: Emerging Options in Water Management, Center for Environmental and Water Resources Engineering, Department of Civil and Environmental Engineering, University of California, Davis, Report 92-1, November 1992. M. Cubed, Regional Economic Impacts of the Palo Verde Test Land Fallowing Program, prepared for the Metropolitan Water District of Southern California, December 1994. M. Cubed, Socioeconomic Assessment of the Proposed Palo Verde ID Land Management, Crop Rotation and Water Supply Program, prepared for the Palo Verde Irrigation District, October 2002. McClurg, Sue, The Water Forum Agreement: A Model for Collaborative Problem Solving, Water Education Foundation, Sacramento, California, 2002. Mitchell, Laura, “Fallowing Debate at County Meeting,” Imperial Valley Press, September 4, 2002. 165 Mitchell, Laura, “Groundwater Permit Denied by Supervisors,” Imperial Valley Press, March 18, 2003. Ostrom, Elinor, Governing the Commons: The Evolution of Institutions for Collective Action, Cambridge University Press, New York, 1990. Phelps, Charles E., Nancy Y. Moore, and Morlie Hammer Graubard, Efficient Water Use in California: Water Rights, Water Districts, and Water Transfers, RAND, R-2386-CSA/RF, Santa Monica, California, 1978. Provencher, Bill, A Quantitative Analysis of Private Property Rights in Groundwater, Ph.D. dissertation, University of California, Berkeley, 1991. Provencher, Bill, “Issues in the Conjunctive Use of Surface Water and Groundwater,” in Daniel Bromley (ed.), Handbook of Environmental Economics, Blackwell Publishers, Cambridge, Massachusetts, 1995. Provencher, Bill, and Oscar Burt, “The Externalities Associated with the Common Property Exploitation of Groundwater,” Journal of Environmental Economics and Management, Vol. 24, 1993, pp. 139– 158. Purkey, D. R., G. A. Thomas, D. K. Fullerton, M. Moench, and L. Axelrad, Feasibility Study of a Maximal Program of Groundwater Banking, Natural Heritage Institute, San Francisco, California, December 1998. “Rural Palo Verde Valley Agrees to Colorado River Pact,” Associated Press, October 22, 2002. Sax, Joseph, “Understanding Transfers: Community Rights and the Privatization of Water,” Hastings West–Northwest Journal of Environmental Law, Policy, Thought, Vol. 1, No. 1, Spring 1994. Sax, Joseph, Review of the Laws Establishing the SWRCB’s Permitting Authority over Appropriations of Groundwater Classified as Subterranean Streams and the SWRCB’s Implementation of Those Laws, State Water Resources Control Board report number 0-076-300-0, Sacramento, California, January 2002. 166 Semitropic Water Storage District, Semitropic Water Storage District’s Policy for Fallowing Land to Make Surface Water Available for Other Purposes, Wasco, California, May 2002. “Summary of Water Agreement,” Desert Sun, Palm Springs, California, October 17, 2002. Sunding, David, “The Price of Water . . . Market-Based Strategies Are Needed to Cope with Scarcity,” California Agriculture, Vol. 54, No. 2, March–April 2000, pp. 56–63. Thomas, Gregory, Designing Successful Groundwater Banking Programs in the Central Valley: Lessons from Experience, The Natural Heritage Institute, Berkeley, California, August 2001. U.S. Bureau of Reclamation, Water Districts Maps: Federal, State, and Private, edition 1.6, Sacramento, California, October 2002. Vaux, H. J., and Richard E. Howitt, “Managing Water Scarcity: An Evaluation of Interregional Transfers,” Water Resources Research, Vol. 20, No. 7, July 1984, pp. 785–792. Vogel, Nancy, “Milestone Water Accord Ok’d: Southland Deal Furthers Shift of Rural Supplies to Cities,” Los Angeles Times, October 17, 2002. Water Education Foundation, The Water Awareness Guide: Where Your Water Comes From, Sacramento, California, 1994. Water Transfer Workgroup, Water Transfer Issues in California, final report to the California State Water Resources Control Board, June 2002. Yniguez, Rudy, “Frank Discussion Hoped For with State Officials on Third-Party Impacts,” Imperial Valley Press, July 25, 2002. 167 About the Author ELLEN HANAK Ellen Hanak is a research fellow at the Public Policy Institute of California. From 1992 to 2001, she was a research economist at France’s Center for Cooperation in International Agricultural Development. Before that, she held positions with the President’s Council of Economic Advisers and the World Bank. Her recent work has focused on the competitiveness of agricultural supply chains, food safety, and water policy. She holds a Ph.D. in economics from the University of Maryland. 169 Related PPIC Publications California’s Infrastructure Policy for the 21st Century: Issues and Opportunities David E. Dowall Making Room for the Future: Rebuilding California’s Infrastructure David E. Dowall and Jan Whittington Building California’s Future: Current Conditions in Infrastructure Planning, Budgeting, and Financing Michael Neuman and Jan Whittington The California Electricity Crisis: Causes and Policy Options Christopher Weare PPIC publications may be ordered by phone or from our website (800) 232-5343 [mainland U.S.] (415) 291-4400 [Canada, Hawaii, overseas] www.ppic.org 171" ["post_date_gmt"]=> string(19) "2017-05-20 09:36:18" ["comment_status"]=> string(4) "open" ["ping_status"]=> string(6) "closed" ["post_password"]=> string(0) "" ["post_name"]=> string(8) "r_703ehr" ["to_ping"]=> string(0) "" ["pinged"]=> string(0) "" ["post_modified"]=> string(19) "2017-05-20 02:36:18" ["post_modified_gmt"]=> string(19) "2017-05-20 09:36:18" ["post_content_filtered"]=> string(0) "" ["guid"]=> string(50) "http://148.62.4.17/wp-content/uploads/R_703EHR.pdf" ["menu_order"]=> int(0) ["post_mime_type"]=> string(15) "application/pdf" ["comment_count"]=> string(1) "0" ["filter"]=> string(3) "raw" ["status"]=> string(7) "inherit" ["attachment_authors"]=> bool(false) }