Water Resources Development, Vol. 26, No. 3, 321–349, September 2010

Financing Water Management and Infrastructure: A Review

FRANK A. WARD Department of Agricultural Economics and Agricultural Business, New Mexico State University

ABSTRACT Many of the world’s irrigated regions face the problem of aging infrastructure and declining revenues to maintain and repair irrigation structures. Policy debates over climate change, population growth, food security, and impacts of irrigation on ecological assets compound the problem, raising the urgency to invest in irrigation infrastructure. Meanwhile, a global call for full- cost recovery for water infrastructure investments increases the need to identify the economic value of sustaining irrigation infrastructure. Despite the growing debates, little comprehensive research has been conducted summarizing factors affecting irrigation investments or policy options available for sustaining irrigation infrastructure. This paper reviews research on factors affecting the level and value of irrigation infrastructure investments. It also reviews research on policy instruments for sustaining irrigation infrastructure, considering both market and institutional approaches. Several market approaches have been found to have the potential to influence the economic attractiveness of investments in irrigation infrastructure. These include infrastructure subsidies, clearing titles to water rights, marginal cost pricing, and non-volumetric pricing. Institutional approaches described include regulatory measures, transboundary agreements, and water user associations. Results may contribute to current debates in various regional, national, and international forums on whether and how water should be priced for agricultural use.

Background Three important goals of irrigated agriculture worldwide are a secure food supply to serve a growing world population, increased water conservation, and reduced environmental costs of agricultural production. Despite the role of irrigated agriculture as a supporter of food security and competitor for key ecological assets (ESCAP, 2006), numerous studies point to inadequate investment in the maintenance of irrigation water application and delivery systems, which can lead to water waste and leakages (Farmani et al., 2007). Some studies estimate losses of up to 25% for delivery systems, as much as 20% from on-farm pipelines, and a further 10–15% lost from inefficient water application technologies, all of which can reduce water needed to sustain key ecological assets. Some of these losses return to the water environment with degraded quality, while others transport pollutants such as salts into , streams, and lakes. For the most part, few contingencies have been made for infrastructure renewal. In the face of increasing transfers of water

Correspondence Address: Prof. Frank A. Ward, Department of Agricultural Economics and Agricultural Business, New Mexico State University, Las Cruces, NM 88003. Email: [email protected] 322 F. A. Ward infrastructure from governments to farmers or irrigation districts, this transfer raises questions about future sources of financing (Simon, 2002). The transfer of financial control and investment responsibility into private hands may require farmers to seek new sources of financing to raise capital to infrastructure. Table 1 shows the distribution of irrigated lands for the world’s major irrigating countries. For purposes of this paper, irrigation infrastructure refers to dams, canals, pipelines, aqueducts, pumping plants, drainage and flow regulating structures (Table 2). In many of the OECD countries, irrigated agriculture faces increasing pressure to transfer water to other uses, including urban uses and for instream flows to support key ecological assets (Connor, 2008). Finding ways to finance maintenance of irrigation infrastructure can provide a source of water to protect key ecological assets as well as

Table 1. Irrigated Land By Country, 2003

Land (sq km) Percentage irrigated Country Irrigated All land of all land of farm land Afghanistan 27,200 647,000 4.2% 33.8% Argentina 15,500 2,737,000 0.6% 5.4% Australia 25,450 7,618,000 0.3% 5.4% Bangladesh 47,250 134,000 35.3% 56.1% Brazil 29,200 8,456,000 0.3% 4.4% Burma 18,700 657,000 2.8% 17.0% Chile 19,000 749,000 2.5% 82.4% China 545,960 9,326,000 5.9% 47.2% Egypt 34,220 995,000 3.4% 99.9% France 26,190 528,000 5.0% 13.3% India 558,080 2,973,000 18.8% 32.9% Indonesia 45,000 1,826,000 2.5% 12.4% Iran 76,500 1,636,000 4.7% 40.2% Iraq 35,250 432,000 8.2% 58.6% Italy 27,500 294,000 9.4% 25.7% Japan 25,920 374,000 6.9% 54.7% Kazakhstan 35,560 2,670,000 1.3% 15.7% Korea 14,600 218,000 6.7% 50.1% Mexico 63,200 1,923,000 3.3% 23.2% Pakistan 182,300 778,000 23.4% 82.4% Philippines 15,500 298,000 5.2% 14.5% Romania 30,770 230,000 13.4% 31.3% Russia 46,000 16,995,000 0.3% 3.7% Saudi Arabia 16,200 2,150,000 0.8% 42.7% South Africa 14,980 1,220,000 1.2% 9.5% Spain 37,800 500,000 7.6% 20.4% Sudan 18,630 2,376,000 0.8% 10.7% Thailand 49,860 511,000 9.8% 28.2% Turkey 52,150 771,000 6.8% 20.0% Turkmenistan 18,000 488,000 3.7% 79.4% Ukraine 22,080 603,000 3.7% 6.6% United States 223,850 9,162,000 2.4% 12.5% Uzbekistan 42,810 425,000 10.1% 84.9% Vietnam 30,000 325,000 9.2% 33.7%

Source: The World Factbook, 2008, and calculations by the author Financing Irrigation Water Management and Infrastructure 323

Table 2. Typical kinds of irrigation infrastructure

Cost Type Use Operation, maintenance, repair Capital storage holds water dam/reservoir medium high tank low medium farm pond low medium canals delivers water low medium drainage removes water low high gates delivers water low low pumps raises water high medium pipes delivers water low medium wells accesses water high medium grading levels land low medium water application delivers water to crop sprinkler medium high gated pipe low medium drip system low high furrow medium low flood low low barrages raises surface water low medium levees channels water medium medium computers times flows medium medium meters measures flow low medium meeting growing demands for other uses of water (Rodriguez, et al., 2006). Extreme events like droughts and floods as well as longer term outcomes of climate change and variability provide more challenges (OECD, 2009). Both public and private options are possible. A number of recent research studies have examined the performance of investments in irrigation infrastructure. For example, the Spanish National Irrigation Plan (PNR) aims to help irrigated agriculture by restoring aging infrastructure (Barbero, 2005). A 2003 study from Spain found that accounts for more than half of the total economic value of irrigated agriculture in Spain while only consuming 20% of its total volume of water. Results show how infrastructure improvement can substitute for limited groundwater supplies. A recent Italian study identified economic, social and environmental indicators for the country’s irrigated farming systems (Bartolini et al., 2007). Several recent works from Australia have examined the economics of irrigation. A 2008 study found that improved infrastructure can substitute for reduced water quality or quantity (Connor, 2008). One information source on the scale and financing of irrigation infrastructure investments in the U.S. comes from the 100-plus year experience of its federal Bureau of Reclamation (Reclamation). Most large dams and water diversion structures in the American west were built by, or with the assistance of Reclamation. As of 2009, Reclamation’s infrastructure provides water to 31 million people and provides irrigation water for 10 million acres of farmland that produce 60% of U.S. vegetables and 25% of its fruits and nuts. Reclamation reported in 2008 that its current infrastructure systems are in generally good condition. But the long term trend will show some decrease in reliability of its facilities. 324 F. A. Ward

A recent study from Pakistan examined impacts on reduced poverty from irrigation infrastructure development. Results suggest that access to irrigation infrastructure reduces the incidence of poverty; it also showed that upgrading watercourse lining saves water, resulting in higher cropping intensity, higher crop productivity, greater food security, and improved farm incomes (Hussain et al., 2007). Debate continues to surround decisions on irrigation infrastructure in both the developed and developing world, particularly where irrigation water competes with other uses such as recreation, environment, ecosystem functions, urban, or hydropower uses. Non-specialists may be overwhelmed by the wide range of benefits and costs that are claimed for various irrigation infrastructure improvements. Better understanding may require consultation with lengthy sources and require considerable background in economic analysis. Few published works are available that distill the scope of modern economic concepts and tools that could be used to support modern water management and policy debates for irrigated agriculture. Little comprehensive research has been conducted reviewing the economics, financing, and policy options for irrigation infrastructure investments. In light of these gaps, the objective of this paper is to review research on factors affecting the level and value of irrigation infrastructure investments. A related objective is to review the research on policy instruments for sustaining irrigation infrastructure, considering both market and institutional approaches. This paper analyses the reasons for under-investment in irrigation infrastructure maintenance, including weak incentives, complex property rights, and financial constraints. It also discusses the economically optimum level of investment. It concludes with a discussion of the various market based initiatives and institutional approaches to provide the necessary finance in cost effective ways. In this discussion, consideration is given to future sources of finance for upgrading and new capital infrastructure, and private public partnerships to raise capital and improve long term asset management for renewal of irrigation infrastructure.

Economic Framework for Irrigation Investments Economic Efficiency Where economic efficiency is an important objective of investments in irrigation infrastructure, those investments will take place whenever the additional economic benefits they produce exceed the added costs. Nevertheless, the development and use of water and related infrastructure for irrigation is often accompanied by one or more of the classic kinds of market failure. These include externalities, public goods, decreasing marginal costs, common property resources, and uncertainty. The presence of these market failures can produce an inefficient allocation of water and related taxpayer resources used to finance irrigation infrastructure. Externalities can result from either the development of water for irrigation or its allocation. Economies of large scale and decreasing marginal costs are often found in irrigation developments. Both create problems for financing systems large enough to capture economies of scale. It also presents challenges in establishing economically efficient water pricing mechanisms. The appendix presents a mathematical economic framework for evaluating investments in irrigation infrastructure. Table 3 shows the economics of irrigation investments from two accounting stances: private farm income and national net economic benefit. Economic Financing Irrigation Water Management and Infrastructure 325 performance for each viewpoint is shown without infrastructure maintenance, with low maintenance, and with high maintenance. The framework in the table can be used to evaluate the economic performance of investments in maintaining or improving irrigation infrastructure. Examples of applying the economic framework described in Table 3 to an ongoing irrigation water conservation debate in New Mexico’s Lower Rio Grande Basin are presented later in the paper. Recreational, environmental and ecological services that benefit from irrigation projects rarely deplete water at the expense of other users. In these cases an irrigation project has the characteristics of a public good, in which several water users can simultaneously consume the services of the project. Finally, the flowing, mobile and leaky nature of water can result in high costs needed to establish and enforce property rights to develop, allocate and use the water. In these cases, water services, especially environmental or ecological values produced by leaving water undeveloped or unused, may emerge as a common property resource, for which the complete opportunity costs of its use in irrigation rarely face the irrigator.

Costs and Benefits Sustainable economic and financial performance for irrigation infrastructure investment ultimately comes from additional incomes earned by the farmer or from added net national economic benefits resulting from that investment. Economically motivated farmers will invest in irrigation improvements only if their anticipated discounted net present value of income is positive (Merriam & Freeman, 2007). There is a long history of irrigation subsidies by various government agencies in many countries (Scheierling et al., 2006; Qadir et al., 2007; Fan et al., 2008; Venot & Molle, 2008; Medellin-Azuara et al., 2009). In light of the importance of irrigated agriculture for food security, the need to protect

Table 3. Economic evaluation of investments in irrigation infrastructure or maintenance

With infrastructure renewal With property Without prop- Accounting Without infrastruc- right to use water erty right to use Data stance ture renewal saved water saved Crop price farmer P P P Crop yield/acre* farmer Yo Y1 Yo Production cost/acre farmer Co C1 Co Crop water applied/acre farmer Ao A1 Ao Water price farmer Pf Pf Pf Water price nation Pn Pn Pn Annualized cost/acre of farmer 0 Cf Cf infrastructure renewal Annualized cost/acre of nation 0 Cn Cn infrastructure renewal * Net income earned/acre farmer [P*Yo-Co]-[Pf Ao][P*Y1-C1]- [P*Yo-Co]- * * [Pf A1]-[Cf] [Pf Ao]-[Cf] * Net economic benefit nation [P*Yo-Co]-[PnAo][P*Y1-C1]- [P*Yo-Co]- * * produced/acre [PnA1]-[Cn] [PnAo]-[Cn] Note: *One acre ¼ 0.404 Ha 326 F. A. Ward farmers from wide swings in food prices, and the desire to preserve a rural society, there are strong incentives for these subsidies to continue where they already exist (Malik, 2008). An important debate centers around whether these objectives can be met without distorting trade (World Trade Organization, 2007). Nevertheless, unless infrastructure maintenance avoids sufficient losses for the farm to pay for the maintenance, the economic benefits produced by a public subsidy are unlikely to be sustained. Moreover, governments and donor organizations rarely assign high priorities or large budgets to irrigation infrastructure maintenance (Pitman, 2004).

Level of Irrigation Investment The desired level of irrigation infrastructure investment can be based on many criteria, including economic efficiency, equity, sustainability and food security. Where economic efficiency is desired, the economic value of the infrastructure will exceed its cost in present value terms. Using that criterion, infrastructure will be improved or restored to the point as long as it continues to increase the economic value of discounted net present value of services from the investment. For a national economic view, improved economic efficiency is measured by the net present value of additional benefits from all water users minus the cost of investing in the infrastructure. In irrigated systems, the economically efficient level of irrigation infrastructure differs for different levels of water availability. However very little adjustment in capacity is possible in the short run once that capacity has been established.

Cost Allocation and Water Pricing The desired method and extent of public finance for irrigation projects is widely debated. For example, in the U.S., irrigation financing was widely debated beginning with the inception of the Reclamation Act of 1902. For that country, one outcome of those debates was numerous changes made by the Congress, all of which separated beneficiaries of irrigation development from incurring its full cost. The original intent of the Act was to promote settlement of the western U.S. When it was found that settlers inexperienced with irrigation would not earn enough farm income from the water for several years, the repayment period was soon extended to 20 years and later to 40 years. Amendments to the U.S. Reclamation Law enacted in the 1930s effectively separated repayment requirements from the real cost of developing and delivering the water. Water prices charged to irrigators on reclamation projects were originally based on the principle of cost recovery. Therefore, irrigation water was priced so that revenues paid for the cost of supplying the water. However, water price was later changed from recovering cost to ‘ability to pay’ based on the principle of repayment capacity. Price was no longer based on recoverable cost, but on marginal benefits of water used in irrigation (Easter & Liu, 2005). Irrigators were charged prices they could afford to pay unless their repayment capacity was greatly overstated. Beneficiaries paid only a fraction of construction costs. Still, even if benefits were estimated poorly, the ratio of local benefits to local costs paid was likely to be much larger than 1.0. Such a low fraction of repayment required gave rise to high local support for federally financed irrigation projects in the U.S. This support gave rise to the ‘iron triangle’ of bankers, real estate interests and farmers with shared economic interests Financing Irrigation Water Management and Infrastructure 327 in development of irrigation projects. Not surprisingly, vigorous support of irrigation projects by iron triangles has provided strong incentives to limit the use of rigorous economic principles by project analysts in the conduct of irrigation project appraisals.

Financing and Cross Subsidies Irrigation infrastructure subsidies are common in many of the world’s dry regions (Perry, 2001). A cross subsidy occurs when losses incurred by pricing one service below its recoverable cost are offset by charging a price above recoverable cost for a related service supplied by the same organization. A common cross subsidy occurs between power buyers and irrigators who benefit from the same reservoir and dam. Managers of reservoir–dam projects that store and channel water for irrigation and generate power are faced with a decision on how to allocate the joint costs common to both activities. These costs are allocated between irrigators and power consumers, including costs of construction and maintenance of the dam and reservoir. A cross subsidy occurs when a smaller part of the costs are allocated to irrigation than would occur under efficient pricing principles. For example in the U.S., Congress adopted the ability to pay principle described above for irrigation reclamation projects, and authorized the difference between the cost of development and repayment charges received from farmers to be paid from cross subsidies paid by hydroelectric power revenues. Electricity sales turned out to be a major source of finance supporting the reclamation program, offsetting the typically weak economic performance of irrigation when considered in isolation. That part of irrigation development not repaid by farmers was financed by Basin Accounts, which allowed deficits from one project to be made up from surpluses from others in the same river basin. The power beneficiary paid 80% or more of the irrigation water costs. Research from many countries has confirmed the presence of cross subsidies favoring irrigation developments. This includes work from a series of developing countries (Sampath, 1992), Spain (Gomez-Limon & Riesgo, 2004), France (Tardieu & Prefol, 2002), India (Scott & Shah, 2004), the Jordan Valley (Doppler et al., 2002); China (Yang et al., 2003), Turkey (Cakmak et al., 2004), Mexico (Scott & Shah, 2004), and Egypt, Indonesia, Morocco and Ukraine (Hellegers & Perry, 2006).

Pricing Irrigation Water below Marginal Cost Irrigation water is often priced below its marginal cost to encourage rural settlement, increase food production, and promote national food self-sufficiency. Where subsidized water prices are charged to irrigators, farm income gains from expanded irrigation infrastructure can be considerable, so these investments are typically supported politically by irrigation interests. Where possible, irrigation supporters attempt to negotiate a project evaluation framework in which private farm income is substituted for net national benefit. In that case, considerable overinvestment in irrigation can be expected from the view of net national benefit compared to an economically efficient level. Where irrigation water is priced below its marginal cost of supply, farm income gains can be expected to be much higher than gains in net national benefit (Easter & Liu, 2007). From a national economic view, irrigation infrastructure expansions can perform poorly when there are associated high environmental costs such as minimum streamflows required to keep an endangered from going extinct (Molle, 2008). 328 F. A. Ward

Potential to Renegotiate Water Delivery Contracts Irrigators may show strong support for an irrigation project even when a high contracted price of water charged to irrigators is needed to secure financing for irrigation infrastructure. This support may continue even when the overall economic performance of an irrigation project is weak. If farmers believe they can renegotiate the contract after the system is built and the water is flowing, their support may be greater than would be predicted by calculations of discounted net present value of the project (Ramos & Garrido, 2004). For example, in one celebrated study, Martin and his colleagues found that irrigators in Arizona who would receive water from the Central Arizona Project (CAP) had signed long-term contracts amounting to 70% of the CAP water allocated for agricultural use (Martin, 1988). The researchers were surprised that so many farmers had committed to purchase CAP surface water at the high price of $65 per acre foot. In fact, many of those same farmers had installed their own groundwater wells and pumps, and their cost of pumped groundwater was a much lower $30–50 per acre foot. The authors found that the farmers had learned to play the water development game. As long as the costs of playing the game are minimal and there is a good chance of a benefit in the future, farmers need take no action now to reduce uncertain future costs. Even if future contracted irrigation water prices are greater than farmers can rationally afford to pay, historical experience showed that once the water project and its water are in place, the price of the water would be negotiable. Moreover, since the Arizona farmers had developed cheap substitute groundwater available for use after CAP was finished, they would be in a strong position when renegotiating contracts with Reclamation.

When Farmers Expect to Secure Water Rights Where water rights are based on historical beneficial use, irrigators may believe that investments in water conservation measures, such as ditch lining, investing in , or water banking for cash, may cause them to forfeit their conserved water. That is, active steps taken on the farm to conserve part of one’s historical water use may be perceived as failure to demonstrate current beneficial use (Ward, 2007). High current water use, even if not needed, is a common method to demonstrate beneficial use in case the water might be needed in the future. Property rights in water have an important effect on the incentive to conserve, as shown in a recent study of Korean agriculture (Labadie et al., 2007). Securing control of the natural flows of a river basin through irrigation developments may also effectively block an urban, environmental or other competing user from claiming the right to use the water. So the act of securing control of water, even by economically weak irrigation infrastructure expansions, is perceived to be an attractive measure for landing a water right with considerable future value. Indeed, urban water uses continue to grow in the world’s dry places, so cities are often the prospective buyer of water from willing farmers. Communicating water’s value in agriculture to urban interests presents its own challenges. For example, the Australian irrigation sector has found it harder to communicate to the wider population the economic benefits of irrigation. So irrigation water suppliers are examining better ways to communicate (Christen et al., 2005). Growing cities typically pay prices two-to-five times higher than the economic value in irrigated agriculture. Financing Irrigation Water Management and Infrastructure 329

Economic Value of Irrigation Infrastructure Table 3 and appendix equations (1) and (2) show factors that influence the economic performance of irrigation projects for each of several levels of irrigation infrastructure investment. The economic value of food produced in addition to the cost of supplying the water, the price charged for water, and urban and environmental values of water all play a part. The formulas presented are quite general, and can be applied to any country, river basin, climate, economic or political system, or set of institutions that govern irrigation and irrigation infrastructure maintenance. They can be adapted to agricultural enterprises that specialize in grains, fruits and vegetables, or fodder. They can account for farming operations for which irrigation is a supplemental activity to reduce risk of intermittent rainfall or for irrigated agriculture in desert regions that receive almost no . Several factors described below influence the value of irrigation infrastructure.

Water Price A lower price of water charged to irrigators increases farm income and increases the value of infrastructure investments. Lower water prices also increase the economic incentive for farmers to produce high water using crops. Low water prices discourage farmers from growing water saving crops. Finally, lower water prices encourage greater water use and encourage farmers to substitute water for other resources, such as land, labor, capital, and water-conserving technology. For the case of net national benefits as described in the appendix, the opportunity cost of water occupies the place in appendix equation (A2) held by the price of water in the farm income analysis in appendix equation (A1). A lower opportunity cost of water increases the net national benefits of measures that maintain irrigation infrastructure. A higher opportunity cost does the opposite. For example, in a region where values of water outside agriculture are high, due to sensitive or high-valued environments, or due to growing cities, then investments in irrigation infrastructure maintenance will perform weakly.

Infrastructure Cost Governments rarely assign high priority to using taxpayer resources to maintain irrigation infrastructure already built. A common belief held by governments is that even if it subsidizes the development of irrigation initially, they are less willing to assign adequate budgets to keep infrastructure in top form. The high cost of maintenance is a major reason. Another belief is that since the farmer or other water user is the main beneficiary, they should be able to pay for its upkeep out of the additional income it produces. It is unlikely that the debate will soon be resolved on who should have the responsibility of maintaining infrastructure. Nevertheless much can be said about the economics of maintaining infrastructure based on its cost. A lower price of infrastructure maintenance increases the quantity of it demanded, since its demand is derived from the demand for its final services, summarized as additional irrigation farm income for the case of irrigation infrastructure. Table 3 and appendix equation (A1) both show that the gain in farm income from infrastructure maintenance is higher when the real cost (per unit land) of that maintenance is lower. With no maintenance needed, farm income is unaffected by changes in the price of maintenance. 330 F. A. Ward

However a given level of maintenance required provides a greater boost to farm income as the maintenance price charged to farmer is lower. What this means is that advances in technology, labor supplies, or institutions that make it cheaper to restore or maintain existing infrastructure all are translated into greater farm income produced. There is a greater farm income produced from a given planned level of infrastructure maintenance when repair costs fall. Also the planned level of maintenance increases because of the added farm income induced by those cost economies.

Water Saved by Infrastructure Maintained Investments that maintain irrigation infrastructure include improvements in dams, canals, pipelines, aqueducts, pumping plants, drainage and flow-regulating structures (Table 2). These investments are generally for the purpose of improving the quantity, quality, timing, or reliability of water itself at the place where the farmer needs it. In much the same way as a lower cost of maintaining infrastructure improves farm income, greater amounts of water saved from that investment that is made available for irrigation use on the farm have a similar economic effect. Still, water lost to poor infrastructure maintenance at one point in a river basin can result in greater water supplies at another point or another time period in the same basin (Ward & Pulido-Velazquez, 2008).

Crop Yields Other things being equal, greater crop yields produce larger farm incomes, raise the value of water in irrigated agriculture, and increase the economic productivity of measures that maintain or improve irrigation infrastructure. When a given measure to maintain infrastructure is applied to places where crop yields are higher, the farm benefits are higher. A good example is a study from Spain’s Lemos Valley Irrigation District that found that farmers abandoned irrigation because they lack economic motivation to irrigate. The authors found greater incentives to invest in irrigation infrastructure where its economic return was highest. This occurred where the infrastructure increased crop yields, saved more water, or substituted for effects of unreliable water supply (Cuesta et al., 2005). As a general principle, we can expect high performance both for farm income and for net national benefit from infrastructure improvements when crop yields are already high, but much weaker incentives to invest where crop yields are low. This set of economic forces gives rise to the cycle of deteriorating infrastructure reducing yields, raising current poverty and food insecurity, which in turn causes low investments that sustain infrastructure, which increases future poverty.

Capital Cost An important element that affects the decision to invest is the cost of capital needed to sustain and support the investment—the interest rate. While not stated, the formulas presented in Table 3 and appendix equations (1) and (2) refer to the economics of irrigation for a single period. Most investments in irrigation infrastructure last for several years, so the added costs and returns need to be summed over relevant times in the planning horizon. A lower interest rate makes any investment in infrastructure Financing Irrigation Water Management and Infrastructure 331 development or maintenance more attractive, since most costs are incurred in the first few periods, while benefits are reaped over many future periods.

Policy Instruments Market Mechanisms Several market mechanisms influence the economic attractiveness of investments in irrigation infrastructure. These include infrastructure subsidies, clearing titles to water rights combined with market water transfers, marginal cost pricing, and non-volumetric pricing of water.

Infrastructure subsidies. Measures that subsidize the capital cost of infrastructure repair have been tried with some success. For example, the U.S. city of San Antonio, Texas, has partnered with the Lower Colorado River Authority to pay for on- and off-farm water conservation measures by irrigators in the Texas Gulf Coast. A considerable part of these subsidies are paying for repairing aging infrastructure and modernizing and improving water conveyance systems (Ward et al., 2008). Similar arrangements have been negotiated with San Diego, California, paying to repair leaky infrastructure in the Imperial Irrigation District (IID). In paying for the infrastructure to reduce water losses, San Diego hopes that the water made available for their urban needs is cheaper than finding it from alternative sources, such as desalting sea water. IID has agreed to transfer up to 67,000 acre feet per year of water conserved by lining of the All American Canal to the city of San Diego in exchange for San Diego financing the canal lining costs (San Diego, 2009). Similar results were found in the Sultanate of Oman from an analysis of water conservation capital subsidies (Abdelrahman & Abdelmagid, 1993). Tables 4–8 show results from a recently completed study from North America’s Rio Grande Basin summarizing impacts of a range of potential public subsidies of irrigation infrastructure. The tables show that the level of a publicly funded irrigation infrastructure subsidy can have a considerable influence on several things. These include impacts on irrigation technology used, water use per unit land, total water use, water conservation, profitability of farm income, and on net national benefits produced by irrigated agriculture. The methods used to conduct the analysis are based on the economic framework shown in the appendix and Table 3. Table 4 shows the changing distribution of irrigated land between flood and drip irrigation for varying possible levels of public subsidies of farmers’ capital costs of the water conserving technology. Higher subsidies make it more economically attractive for irrigators to invest in drip irrigation infrastructure. Table 5 shows impacts of the same set of subsidies from the view of water applied to the suite of crops. Most crops farmed show a strong responsiveness to increased subsidy levels, with the largest changes in technology from flood to drip irrigation occurring for alfalfa and cotton. Surprisingly, Table 6 presents a different story for impacts on water consumption (ET) than for water application. The same drip irrigation subsidy actually increases water consumption, from a total of about 200,000 acre feet per year with no subsidy to a high of about 215,000 acre feet annually at the highest subsidy. This unexpected result of a negative conservation of about 15,000 acre feet produced by the maximum subsidy occurs because drip irrigation produces higher crop yields than flooding. These greater yields 332 F. A. Ward

Table 4. Land in irrigated agriculture for selected drip irrigation subsidies, lower Rio Grande Basin, North America, annual average, 2006-2025

Subsidy (% of Subsidy Land in drip Land in flood Total land under capital cost) ($/ac/yr) irrigation irrigation irrigation 1000 acres/yr

0 0 29.7 59.6 89.3 10 10 29.7 59.6 89.3 20 20 29.9 59.4 89.3 30 30 30.2 59.1 89.3 40 40 38.5 50.8 89.3 50 50 62.0 27.4 89.3 60 60 62.0 27.4 89.3 70 70 62.0 27.4 89.3 80 80 63.0 26.3 89.3 90 90 63.0 26.3 89.3 100 100 63.4 25.9 89.3 require greater crop water ET to sustain those higher yields. So, subsidizing modern irrigation infrastructure, even when intended to promote water conservation and make more available for ecological and other uses, can increase consumption and reduce supplies available for use outside agriculture (Brinegar & Ward, 2009). Table 7 summarizes impacts on regional farm income. Despite the negative impacts of subsidies on water conservation, drip irrigation subsidies have the effect of increasing farm income. Farmers in the region do not take advantage of the subsidy unless or until their income opportunities increase. Table 8 summarizes impacts on net social benefit, as described in Table 3 for various drip irrigation subsidy levels. The difference between benefits to farmers and national benefits is the financial cost to the taxpayer of the subsidy. That is, net social benefits are measured as farm income minus the monetary cost of the subsidy. Overall, Tables 4–8 show the surprising result that not all subsidies of irrigation infrastructure make more water available to support ecological assets by conserving water in agriculture. So while infrastructure subsidies increase farm income and reduce water applications in agriculture, higher crop water ET consumed under drip irrigation’s higher yields means that negative water conservation occurs when tracking total basin water use.

Water rights, water transfers, and water markets. Water markets can be an economically efficient institution for implementing marginal cost pricing and for establishing the incentives to support maintenance of irrigation infrastructure. When a price is paid or an economic value is secured for water that would be saved by repaired infrastructure, then that price or value signals the marginal cost of leaving the infrastructure unrepaired (Ward et al., 2007). Market mechanisms can encourage a greater economic value of scarce water supplies in the face of growing scarcities of all uses by addressing allocation inefficiencies found in traditional irrigation institutions. In irrigated agriculture, markets provide a mechanism to allocate water in which the water user, buyer and seller all face the opportunity cost of current use patterns. Water markets, which harness market forces to determine the price Table 5. Irrigation water applied by crop, technology, and conservation subsidy, Lower Rio Grande Basin, North America, annual average, 1000 Acre Feet Per Year, 2006-2025

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Alfalfa flood 88.1 88.1 87.0 85.6 117.4 0.0 0.0 0.0 0.0 0.0 0.0

Pima cotton flood 40.2 40.2 40.2 40.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Financ Upland cotton flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Spring lettuce flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Fall lettuce flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ing Fall onions flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Mid season flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Irrigatio onions Spring onions flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Grain sorghum flood 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.0 n

Wheat flood 2.7 2.7 2.7 2.7 2.7 2.7 2.7 2.7 0.0 0.0 0.0 Water Green chile flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Red chile flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Pecans flood 155.4 155.4 155.4 155.4 155.4 155.4 155.4 155.4 155.4 155.4 155.4 Manag Alfalfa drip 24.7 24.7 25.2 26.0 8.9 72.0 72.0 72.0 72.0 72.0 72.0 Pima cotton drip 0.0 0.0 0.0 0.0 21.5 21.5 21.5 21.5 21.5 21.5 21.5 eme Upland cotton drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Spring lettuce drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 nt

Fall lettuce drip 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 and Fall onions drip 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9

Mid season drip 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 Infrastruc onions Spring onions drip 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 Grain sorghum drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4

Wheat drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.4 1.4 1.4 ture Green chile drip 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6

Continued 333 334 F. A. Ward

Table 5. Continued

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Red chile drip 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 Pecans drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total applied flood 287.1 287.1 286.1 284.7 276.3 158.9 158.9 158.9 156.2 156.2 155.4 Total applied drip 77.7 77.7 78.3 79.0 83.5 146.6 146.6 146.6 148.1 148.1 148.5 Grand total total 364.9 364.9 364.4 363.7 359.8 305.5 305.5 305.5 304.3 304.3 303.9 Change in water total 0.0 0.0 20.5 21.1 25.1 259.3 259.3 259.3 260.6 260.6 261.0 applications (ref. no subsidy) Table 6. Irrigation water consumption by crop, technology, and conservation subsidy, Lower Rio Grande, NM, USA, annual average, 1000 acre feet per year, 2006-2025

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Alfalfa flood 37.9 37.9 37.4 36.8 50.5 0.0 0.0 0.0 0.0 0.0 0.0

Pima cotton flood 17.2 17.2 17.2 17.2 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Financ Upland cotton flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Spring lettuce flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Fall lettuce flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 ing Fall onions flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Mid season onions flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Irrigatio Spring onions flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Grain sorghum flood 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.0

Wheat flood 1.2 1.2 1.2 1.2 1.2 1.2 1.2 1.2 0.0 0.0 0.0 n

Green chile flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Water Red chile flood 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Pecans flood 66.8 66.8 66.8 66.8 66.8 66.8 66.8 66.8 66.8 66.8 66.8 Alfalfa drip 24.7 24.7 25.2 26.0 8.9 72.0 72.0 72.0 72.0 72.0 72.0 Manag Pima cotton drip 0.0 0.0 0.0 0.0 21.5 21.5 21.5 21.5 21.5 21.5 21.5 Upland cotton drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 eme Spring lettuce drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Fall Lettuce drip 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 8.0 nt

Fall onions drip 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 8.9 and Mid season onions drip 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2 10.2

Spring onions drip 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 12.1 Infrastruc Grain sorghum drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.4 Wheat drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 1.4 1.4 1.4 Green chile drip 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6 6.6

Red chile drip 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 7.2 ture Pecans drip 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

Continued 335 336 F. A. Ward

Table 6. Continued

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Total consumption flood 123.4 123.4 123.0 122.4 118.8 68.3 68.3 68.3 67.2 67.2 66.8 Total consumption drip 77.7 77.7 78.3 79.0 83.5 146.6 146.6 146.6 148.1 148.1 148.5 Grand total total 201.2 201.2 201.3 201.4 202.3 214.9 214.9 214.9 215.2 215.2 215.3 Change in water con- total 0.0 0.0 0.1 0.3 1.2 13.8 13.8 13.8 14.1 14.1 14.2 sumption (ref. no subsidy) Table 7. Farm Income by Crop, Technology, and Conservation Subsidy, Lower Rio Grande, NM, USA, Annual Average, 1000 US Dollars Per Year, 2006- 2025

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Alfalfa flood 11,103 11,103 10,975 10,796 14,797 0 0 0 0 0 0

Pima cotton flood 283 283 283 283 0000000Financ Upland cotton flood 00000000000 Spring lettuce flood 00000000000

Fall lettuce flood 00000000000ing Fall onions flood 00000000000 Mid season onions flood 00000000000Irrigatio Spring onions flood 00000000000 Grain sorghum flood 266 266 266 266 266 266 266 266 266 266 0

Wheat flood 2104 2104 2104 2104 2104 2104 2104 2104 0 0 0 n

Green chile flood 00000000000Water Red chile flood 00000000000 Pecans flood 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 Alfalfa drip 4,143 4,477 4,918 5,420 1,986 16,962 17,939 18,915 19,891 20,867 21,843 Manag Pima cotton drip 00001056371,169 1,701 2,233 2,765 3,297 Upland cotton drip 00000000000 eme Spring lettuce drip 00000000000

Fall lettuce drip 1,121 1,284 1,446 1,609 1,772 1,934 2,097 2,260 2,422 2,585 2,748 nt

Fall onions drip 6,676 6,806 6,936 7,067 7,197 7,327 7,457 7,587 7,717 7,847 7,978 and Mid season onions drip 574 704 834 964 1,094 1,224 1,355 1,485 1,615 1,745 1,875

Spring onions drip 1,257 1,387 1,517 1,648 1,778 1,908 2,038 2,168 2,298 2,428 2,559 Infrastruc Grain sorghum drip 0000000000259 Wheat drip 00000000295 256 217 Green chile drip 1,408 1,506 1,603 1,701 1,799 1,896 1,994 2,091 2,189 2,287 2,384

Red chile drip 2,237 2,335 2,433 2,530 2,628 2,725 2,823 2,921 3,018 3,116 3,213 ture Pecans drip 00000000000

Continued 337 338 F. A. Ward

Table 7. Continued

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Total income flood 35,349 35,349 35,221 35,041 38,760 23,963 23,963 23,963 24,066 24,066 24,133 Total income drip 17,416 18,499 19,688 20,938 18,358 34,615 36,871 39,128 41,289 43,585 45,821 Grand total farm total 52,765 53,848 54,909 55,980 57,118 58,578 60,834 63,090 65,355 67,651 69,954 income Change in farm total 0 1,083 2,144 3,215 4,353 5,813 8,069 10,325 12,590 14,886 17,189 income (ref. no subsidy) Table 8. Net Social Benefit from Farming by Crop, Technology, and Conservation Subsidy, Lower Rio Grande, NM, USA, Annual Average, 1000 US Dollars Per Year, 2006-2025

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Alfalfa flood 11,103 11,103 10,975 10,796 14,797 0 0 0000

Pima cotton flood 283 283 283 283 0 0 0 0000Financ Upland cotton flood 0000 0 0 0 0000 Spring lettuce flood 0000 0 0 0 0000

Fall lettuce flood 0000 0 0 0 0000ing Fall onions flood 0000 0 0 0 0000 Mid season onions flood 0000 0 0 0 0000Irrigatio Spring onions flood 0000 0 0 0 0000 Grain sorghum flood 266 266 266 266 266 266 266 266 266 266 0

Wheat flood 2104 2104 2104 2104 2104 2104 2104 2104 0 0 0 n

Green chile flood 0000 0 0 0 0000Water Red chile flood 0000 0 0 0 0000 Pecans flood 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 24,133 Alfalfa drip 4,143 4,143 4,234 4,362 1,501 12,082 12,082 12,082 12,082 12,082 12,082 Manag Pima cotton drip 000022,022 22,022 22,022 22,022 22,022 22,022 22,022 Upland cotton drip 0000 0 0 0 0000 eme Spring lettuce drip 0000 0 0 0 0000

Fall lettuce drip 1,121 1,121 1,121 1,121 1,121 1,121 1,121 1,121 1,121 1,121 1,121 nt

Fall onions drip 6,676 6,676 6,676 6,676 6,676 6,676 6,676 6,676 6,676 6,676 6,676 and Mid season onions drip 574 574 574 574 574 574 574 574 574 574 574

Spring onions drip 1,257 1,257 1,257 1,257 1,257 1,257 1,257 1,257 1,257 1,257 1,257 Infrastruc Grain sorghum drip 0000 0 0 0 0 0 02205 Wheat drip 0000 0 0 0 02407 2407 2407 Green chile drip 1,408 1,408 1,408 1,408 1,408 1,408 1,408 1,408 1,408 1,408 1,408

Red chile drip 2,237 2,237 2,237 2,237 2,237 2,237 2,237 2,237 2,237 2,237 2,237 ture Pecans drip 0000 0 0 0 0000

Continued 339 340 F. A. Ward

Table 8. Continued

Water conservation subsidy (percentage of capital cost) Irrigation Crop technology 0 10 20 30 40 50 60 70 80 90 100 Total flood 35,349 35,349 35,221 35,041 38,760 23,963 23,963 23,963 24,066 24,066 24,133 Total drip 17,416 17,416 17,507 17,636 12,752 23,333 23,333 23,333 22,926 22,926 22,721 Grand total total 52,765 52,765 52,729 52,677 51,512 47,296 47,296 47,296 46,992 46,992 46,854 Change in net total 0 0 236 288 21,253 25,469 25,469 25,469 25,772 25,772 25,911 social benefits (ref. no subsidy) Financing Irrigation Water Management and Infrastructure 341 for irrigation water, are also more flexible than centrally controlled, allocation mechanisms. These markets require the foundation of well-defined, tradable water rights and supporting reservoirs, canals and other infrastructure to distribute water. Externalities like return flows, third-party effects, and instream uses (Hellegers & Perry, 2006) raise the challenge of improving water use efficiency through the market mechanism. Where culture permits, informal water markets have often developed on their own when water is scarce or when local, regional or national governments are unable to quickly respond to rapidly changing water shortages. However, an institutional and technical structure is needed to carry out a market-based policy. Add to these the problems of externalities, interdependencies across time, and uncertain water supplies, markets alone are unlikely to secure improved economic efficiency. The development or improved efficiency of water banks, water rentals, water leasing, water trading, or other market forms of water transfers provide an excellent set of economic incentives whose effect could contribute to improving aging irrigation infrastructure (Ward et al., 2007). The presence of these various kinds of water markets provides a mechanism to harness the power of market incentives to self-finance irrigation infrastructure. To the extent that water markets can be developed that reflect the income value of infrastructure improvements, considerable burdens and frustrations are reduced from searching for external financing by governments or other external organizations. One of the biggest problems related to the development of water markets is the high cost of initiating the change produced by the transition from existing regulatory or traditional water allocation mechanism to a market environment. A good example is the high cost of discovering historical water use patterns or another basis for a property right in the face of poor or absent records documenting historical use. This information typically is required prior to adjudicating water rights and may be a necessary prerequisite for establishing a market system, which itself may be needed to economically justify irrigation infrastructure improvements. Where irrigation infrastructure needs repair, and water losses would result from failure to repair, it is only clear who has the incentive to bear responsibility for taking care of it when the property rights in water are well-defined. Without well-defined rights to water, infrastructure repair suffers all the well-known problems of a common property resource, with little incentive for anyone to contribute their share of the financing (Herrera et al., 2006).

Marginal cost pricing. Many cultures treat water as a free resource. While free water has desirable equity properties to water users lucky enough to secure a reliable supply, it can damage incentives to repair aging infrastructure. Economic analysis faces major challenges in the search for institutions that encourage more efficient use of high-cost water. One way to promote economically efficient water use by sending the right signals to repair infrastructure is to establish institutions that confront all water users with the real cost of their actions. A good example of this principle has been for irrigation water pricing in France. That pricing aims to recover costs and reduce the cost of public financing for the operation and maintenance costs for irrigation projects. A large part of the capital cost of irrigation infrastructure, ranging from 15% to 60% is charged to farmers, with heavy reliance on volumetric pricing (Rieu, 2005). Economically efficient water supply requires clear price signals that provide incentives for economically efficient use of water by individual consumers, resulting in total benefits of the water supplied exceeding costs by the greatest amount possible. One method for 342 F. A. Ward promoting that economic efficiency is marginal cost pricing in which each water user pays a price that reflects the incremental cost of their use on the system. Marginal cost pricing equates the price of a unit of water consumed in agriculture with the marginal cost of making it available for use. The marginal cost of an additional unit includes delivery costs, costs of building or restoring the infrastructure to make the unit available, and all other opportunity costs, including those of other beneficial uses displaced, including key ecological assets lost, and impacts of hydropower, municipal and recreational water uses displaced. Where the price charged to irrigators reflects all these marginal costs, resulting allocation among current irrigation and other current and future uses displaced can be economically efficient (Johansson et al., 2002). A significant challenge posed by marginal cost pricing of irrigation water is the development of ways to account for all the marginal costs and benefits when establishing the price. For irrigated agriculture, costs include the collection of fees as well as providing system operation and maintenance; these costs may vary over time because of highly fluctuating supplies and demands by other users including needs by key ecological assets (Connor, 2008). When irrigation water is underpriced, there is little incentive to repair aging infrastructure as long as supplies can be obtained from alternative sources. Still, even if the price is zero, when irrigators suffer shortages from lack of infrastructure repair, some incentives remain to invest in needed repairs, as long as the saved water produced by the repairs reduces the cost of those shortages. The irrigator’s location in the system influences the marginal cost of repairing infrastructure as well as the marginal cost of ignoring repairs. Farmers located at higher elevations, for example, can impose significantly added costs on downstream users from pumping compared to those on level ground. Also, farms located at the lower end of the system may have lower costs of infrastructure repair. One way to find the needed information to implement marginal cost pricing is through the development and application of integrated basin scale models that account for the economic value of water in its several uses, locations, and time periods (Scheierling et al., 2006; Ward et al., 2007).

Non-volumetric pricing. Non-volumetric methods charge for irrigation water based on something else than use. Examples include prices based on a per unit output, per unit input, per unit area, or per unit land. Bos & Walters (1990), in their international survey, found more than half of farmers surveyed on several million hectares of land faced water charges based on the total land area irrigated. This method is convenient to water managers because it is easy to administer, requiring no special measurement of irrigation water supplied (Johannson et al., 2002).

Institutional Approaches Irrigation institutions. Irrigation institutions are structures and mechanisms of social order and cooperation governing the behavior of a group of irrigators. They refer to the interacting policy, legal and administrative structures supporting the development, ownership and use of water (Global Water Partnership, 2000). Laws and formal and informal rules that define the allocation of water to irrigation affect the performance of the system and in some cultures define the performance. The status of water law and property rights is connected to politics and influences the kinds of culturally acceptable water Financing Irrigation Water Management and Infrastructure 343 regulations. Irrigation institutions have an important effect on economic incentives that influence efficient and/or equitable water allocations. In most countries and cultures, rights to use water use have evolved through customary or formal laws. For irrigation, water rights specify how water is divided among individual farmers and among irrigated regions (Torell, 2010). Three systems dominate water rights arrangements in most parts of the world (Johansson & Dinar, 2002). These include: riparian rights that assign the right to use water to people owning adjacent land; public allocation based on priorities determined by local, regional or national government; and prior appropriation, determined by historical or customary use. Water rights among countries can be based on treaties or other less formal arrangements (e.g., the celebrated picnic table talks between Jordan and Israel) for allocating transboundary waters.

Regulatory solutions. Governments can play a constructive role in influencing water allocations and affecting economic efficiency by establishing regulations, standards or requirements for upkeep in irrigation infrastructure. The common property nature of benefits produced by maintained infrastructure and associated potential for free riders serve to block economically efficient infrastructure maintenance. In that case regulations can be an important way to maintain infrastructure. Regulations have been attempted for many dimensions of water, including water use levels and timing, place of use or water transfers, and environmental pollution control. Supported by an underlying legal framework, regulations require, permit or restrict particular activities, or prescribe certain outcomes in connection with water use. For an existing regulation to be economically efficient and to achieve community support, the economic benefits of the regulation need to outweigh its costs, and the costs and benefits need to be shared fairly. Regulation redesign would be targeted and concentrated where the highest economic efficiency (additional net benefits) could be produced. Especially in places where water rights defined informally or not at all, water saved from infrastructure maintenance, regulations could be an economically efficient measure for sustaining infrastructure.

Water user associations. Water user associations are composed of stakeholders who jointly manage water, sometimes at the scale of a watershed (Ashraf et al., 2007). There is growing interest in using these associations to provide information for water resource managers. There is considerable variability in associations’ goals, effectiveness, stakeholder composition, involvement in decision making, types of participation allowed, leadership, financing, economic efficiency and temporal scale. Several analyses have identified a rapid growth of watershed associations in the 1990s. These have blossomed in response to trends of increasingly ineffective forums and processes of water management decision making that have marginalized the role of local stakeholders in problem solving. In most cases, watershed user associations provide a pragmatic vehicle for resource managers and stakeholders to address common concerns in a more economically efficient manner than is otherwise possible.

Transboundary agreements. Transboundary waters shared by two or more nations are common in the world’s river basins. One way of reducing the cost of infrastructure maintenance as well as increasing the size and scope of its benefits is for two or more nations jointly to develop, finance, manage and use common rivers, where elements of 344 F. A. Ward responsibility are assigned to each nation based on comparative advantage. This permits all nations to gain from trade. Transboundary rivers pose major economic and political challenges in policy design. Designing an economically efficient, fair and sustainable measure to allocate scarce and random supplies that meets the needs of all parties is a major challenge, particularly when two or more political units share a water source. The water sharing agreement is one way to allocate these supplies. In the western U.S., for waters extending beyond the borders of one state, interstate compacts have been used with success. An interstate compact is a negotiated agreement among the states that, once ratified by Congress, becomes both a federal law and a contract between the signing states. Beginning in 1922 with the signing of the Colorado River Compact, twenty two such compacts currently divide the waters of western American rivers.

The Information Challenge To put it mildly, information on irrigation infrastructure is typically poor and what little data are available are rarely consistently collected. Without good data, economic principles have little to offer to support major policy choices that would conserve water in irrigation and make more available for the environment. There are considerable needs for better information in order that economic principles are put to good use in evaluating irrigation infrastructure. For example, Briscoe (1999) in an honest, candid and compelling article, states that a then recent World Bank study on irrigation financing concluded “there are no reliable statistics on global irrigation investment.” He also states that in this data- poor environment, estimates of aggregate levels of investment on irrigation are little more than educated guesses, typically ranging from $10–$15 billion per year. For a start, information is needed on cost-sharing arrangements between irrigators and public suppliers of irrigation, impacts on water savings at the farm level, project level, and the basin scale from infrastructure improvement. Good data combined with judicious use of economic principles have considerable potential to productively inform decisions on why, when, where, and how to develop and sustain irrigation and its infrastructure.

Conclusions Many of the world’s irrigated areas face the challenge of aging infrastructure and a limited revenue base from which to fund maintenance and repairs. The drive towards full cost recovery for storage and delivery services arising from emerging water reform policies means that both water suppliers and irrigators need to consider a strategic evaluation of infrastructure and the cost of renewal. This paper has characterized factors affecting the level and economic value of investment in irrigation and in irrigation infrastructure maintenance. It examined both financial and economic views of the economically efficient level of investment in irrigated agriculture. It also examined some of the historic causes of higher levels of investment in irrigation compared to investments limited to those that pass a rigorous economic analysis. The economic performance of those investments depends on how the additional water produced or saved is managed on the farm. Five factors are identified that increase the level of investment in irrigated agriculture. These include pricing water based on ability to pay rather than marginal cost, financing Financing Irrigation Water Management and Infrastructure 345 mechanisms in which cross subsidies from hydropower charges finance irrigation development, pricing irrigation water below the marginal cost of supply, the potential for irrigators to renegotiate contracts after projects are built, and incentives to overinvest in irrigation when irrigators believe they will secure a water right. Five factors are identified that lead to a higher value of infrastructure maintenance. These include a lower price of water charged to the irrigator, a lower real cost of repairing infrastructure, a greater water savings from infrastructure maintenance, higher crop yields produced by saved water, and a lower cost of capital. Several market approaches are identified that influence the economic attractiveness of investments in irrigation infrastructure. These include subsidies of infrastructure, clear titles to water rights that promote market transfers of water, and marginal cost pricing of irrigation water. Several policy approaches are also discussed. These include various regulatory solutions requiring infrastructure maintenance, development of water user associations and river basin compacts, treaties and transboundary agreements. A considerable challenge results from poor data available to support the rigorous use of economic principles to inform the formulation and implementation of plans to renew infrastructure. Important data needs will focus on assembly of detailed cost and return enterprise budgets for irrigated agriculture that account for financial and economic impacts of greater water supply, increased water supply reliability, and various forms of substitution between water, land, labor and infrastructure. Other important data needs focus on the cost and productivity of measures to maintain dams, canals, pipelines, aqueducts, pumping plants, wells, drainage and flow regulating structures. Improvements in models of water supply and demand for alternative uses could be even more valuable. Economic analysis has considerable potential to improve economic efficiency, equity and sustainability of irrigation infrastructure investments. Still, difficult challenges remain. Without pretending to be exhaustive, the following short list presents a few research challenges: . Designing laws, property rights systems, institutions and incentives for water that rewards farmers for taking care of on farm and public irrigation infrastructure. . Finding measures that signal the marginal cost of water applied and water consumed in irrigation. . Using economic principles to discover which, when and by how much irrigation infrastructure should be repaired, restored, improved or abandoned. . How to price irrigation water low enough to promote farm income and food security while sending price signals that support ecosystem functions and conserve water. . Effect of subsidized irrigation infrastructure and infrastructure repair on water use inside and outside of agriculture, and on food prices and food quantities.

Acknowledgements

The author is grateful for financial support by Organisation for Economic Co-operation and Development and New Mexico Agricultural Experiment Station.

References Abdelrahman, H. A. & Abdelmagid, I. M. (1993) Water conservation in Oman, Water International, 18, pp. 95–102. 346 F. A. Ward

Ashraf, M., Kahlown, M. A. & Ashfaq, A. (2007) Impact of small dams on agriculture and groundwater development: A case study from Pakistan, Agricultural Water Management, 92, pp. 90–98. Barbero, A´ . (2005) The Spanish national irrigation plan. Paper presented at the OECD Workshop on Agriculture and Water: Sustainability, Markets, and Policies, Adelaide, South Australia, 14–18 November. Bartolini, F., Bazzani, G. M., Gallerani, V., Raggi, M. & Viaggi, D. (2007) The impact of water and agriculture policy scenarios on irrigated farming systems in Italy: An analysis based on farm level multi-attribute linear programming models, Agricultural Systems, 93, pp. 90–114. Batterink, M. (2005) Allocation of costs and benefits in the water framework directive: A Dutch exploration. Paper presented at the OECD Workshop on Agriculture and Water: Sustainability, Markets, and Policies, Adelaide, South Australia, 14–18 November. Bear, S., Hall, A. & Pechey, L. (2006) Irrigation infrastructure charging: Non-rival access, cost sharing, and exclusion. Paper presented at the 50th Annual Conference of the Australian Agricultural and Resource Economics Society, Sydney, Australia, 8–10 February. Brinegar, H. & Ward, F. A. (2009) Basin impacts of irrigation water conservation policy, Ecological Economics, 69, pp. 414–426. Briscoe, J. (1999) The financing of hydropower, irrigation, and water supply infrastructure in developing countries, International Journal of Development, 15, pp. 459–491. Cakmak, B., Beyribey, M. & Kodal, S. (2004) Irrigation water pricing in water user associations, Turkey, International Journal of Water Resources Development, 20, pp. 113–124. Chakravorty, U. & Umetsu, C. (2003) Basinwide water management: A spatial model, Journal of Environmental Economics and Management, 45, pp. 1–23. Christen, E. W., Meyer, W. S., Jayawardane, N. S., Shepheard, M., Robinson, D., Marshall, A. & Fairweather, H. (2005) Triple bottom line reporting to promote sustainability of irrigation in Australia. Paper presented at the OECD Workshop on Agriculture and Water: Sustainability, Markets, and Policies, Adelaide, South Australia, 14–18 November. Connor, J. (2008) The economics of time delayed salinity impact management in the River Murray, Water Resources Research, 44, pp. Cuesta, T. S., Cancela, J. J., Dafonte, J., Valcarcel, M. & Neira, X. X. (2005) Social aspects influencing water management in the Lemos Valley Irrigation District, Spain, Irrigation and Drainage, 54, pp. 125–133. Doppler, W., Salman, A. Z., Al-Karablieh, E. K. & Wolff, H. P. (2002) The impact of water price strategies on the allocation of irrigation water: The case of the Jordan Valley, Agricultural Water Management, 55, pp. 171–182. Easter, K. W. & Liu, Y. (2007) Who pays for irrigation: cost recovery and water pricing? Water Policy,9, pp. 285–303. Easter, W. & Liu, Y. (2005) Cost Recovery and Water Pricing for Irrigation and Drainage Projects, Agricultural and Rural Development Discussion Paper 26 (Washington, DC: World Bank). Economic and Social Commission for Asia and the Pacific (ESCAP) (2006) Enhancing Regional Cooperation in Infrastructure Development Including That Related to Disaster Management (New York, NY: United Nations). Fan, S. G., Gulati, A. & Thorat, S. (2008) Investment, subsidies, and pro-poor growth in rural India, Agricultural Economics, 39, pp. 163–170. Farmani, R., Abadia, R. & Savic, D. (2007) Optimum design and management of pressurized branched irrigation networks, Journal of Irrigation and Drainage Engineering-Asce, 133, pp. 528–537. Global Water Partnership Secretariat Technical Advisory Committee (TAC) (2000) Integrated Water Resources Management, TAC Background Paper 4 (Stockholm, Sweden: Global Water Partnership). Gomez-Limon, J. A. & Riesgo, L. (2004) Water pricing: Analysis of differential impacts on heterogeneous farmers, Water Resources Research, 40. Hellegers, P. & Perry, C. J. (2006) Can irrigation water use be guided by market forces? theory and practice, International Journal of Water Resources Development, 22(X), pp. 79–86. Herrera, P. A., Van Huylenbroeck, G. & Espinel, R. L. (2006) Asymmetric information on the provision of irrigation through a public infrastructure: The case of the Peninsula of Santa Elena, Ecuador, Water Resources Management, 20, pp. 431–447. Hussain, I., Jehangir, W., Mudasser, M., Nazir, A. & Ashfaq, M. (2007) Impact Assessment of Irrigation Infrastructure Development on Poverty Alleviation: A Case Study from Pakistan, JBICI Research Paper No. 31 (Japan Bank for International Cooperation). Labadie, J. W., Fontane, D. G., Lee, J. H. & Ko, I. H. (2007) Decision support system for adaptive river basin management: Application to the Geum River basin, Korea, Water International, 32, pp. 397–415. Financing Irrigation Water Management and Infrastructure 347

Malik, R. P. S. (2008) Towards A Common Methodology for Measuring Irrigation Subsidies (Geneva, Switzerland: The Global Subsidies Initiative of the International Institute for Sustainable Development). Martin, W. E. (1988) Back to the future: A willingness to play reexamined, Western Journal of Agricultural Economics, 13, pp. 112–120. Medellin-Azuara, J., Howitt, R. E., Waller-Barrera, C., Mendoza-Espinosa, L. G., Lund, J. R. & Taylor, J. E. (2009) A calibrated agricultural water demand model for three regions in northern Baja California, Agrociencia, 43, pp. 83–96. Meijer, K., Boelee, E., Auqustijn, D. & van der Molen, I. (2006) Impacts of concrete lining of irrigation canals on availability of water for domestic use in southern Sri Lanka, Agricultural Water Management, 83, pp. 243–251. Merriam, J. L. & Freeman, B. J. (2007) Operational cost benefits study of flexible on-farm irrigation supply systems, Journal of Irrigation and Drainage Engineering, 133, pp. 12–16. Molle, F., Venot, J. P. & Hassan, Y. (2008) Irrigation in the Jordan Valley: Are water pricing policies overly optimistic? Agricultural Water Management, 95, pp. 427–438. OECD (2009) Managing Water for All: An OECD Perspective on Pricing and Financing—Key Messages for Policy Makers (Paris: OECD). Parris, K. & Jung, P. K. (Eds) (2003) Agricultural Water Quality and Water Use Indicators in the Context of the OECD Work on Agri-environmental Indicators (Paris, France and Gyeongju, Republic of Korea: OECD Secretariat). Perry, C. (2001) Water at any price? Issues and options in charging for irrigation water, Irrigation and Drainage, 50, pp. 1–7. Pitman, G. T. K. (2004) Project Performance Assessment Report: Armenia Irrigation Rehabilitation Project (Washington, DC: World Bank). Qadir, M., Sharma, B. R., Bruggeman, A., Choukr-Allah, R. & Karajeh, F. (2007) Non-conventional water resources and opportunities for water augmentation to achieve food security in water scarce countries, Agricultural Water Management, 87, pp. 2–22. Ramos, A. G. & Garrido, A. (2004) Formal risk-transfer mechanisms for allocating uncertain water resources: The case of option contracts, Water Resources Research, 40, pp. Rieu, T. (2005) Water pricing for agriculture between cost recovery and water conservation: Where do we stand in France ? Paper presented at the OECD Workshop on Agriculture and Water: Sustainability, Markets, and Policies, Adelaide, South Australia, 14–18 November. Rodriguez, L. B., Cello, P. A. & Vionnet, C. A. (2006) Modeling stream- interactions in a shallow aquifer, Choele Choel Island, Patagonia, Argentina, Hydrogeology Journal, 14, pp. 591–602. San Diego County Water Authority (2009) Colorado Water Transfer Agreement. Available at http://www. sdcwa.org/manage/pdf/WaterTransfer.pdf Scheierling, S. M., Young, R. A. & Cardon, G. E. (2006) Public subsidies for water-conserving irrigation investments: Hydrologic, agronomic, and economic assessment, Water Resources Research, W03428. doi:10.1029/2004 WR003809:W03428– W03428. Scott, C. A. & Shah, T. (2004) Groundwater overdraft reduction through agricultural energy policy: Insights from India and Mexico, International Journal of Water Resources Development, 20, pp. 149–164. Shah, T., Hassan, M. U., Khattak, M. Z., Banerjee, P. S., Singh, O. P. & Rehman, S. U. (2009) Is irrigation water free? A reality check in the Indo-Gangetic basin, World Development, 37, pp. 422–434. Simon, B. M. (2002) Devolution of bureau of reclamation constructed water facilities. Journal of the American Water Resources Association, 38, pp. 1187–1194. Tardieu, H. & Prefol, B. (2002) Full cost or ‘sustainability cost’ pricing in irrigated agriculture: Charging for water can be effective, but is it sufficient?, Irrigation and Drainage, 51, pp. 97–107. Tilmant, A., Pinte, D. & Goor, Q. (2008) Assessing marginal water values in multipurpose multireservoir systems via stochastic programming, Water Resources Research, 44. Torell, G. (2010) Economic analysis for improved water management, food security, and rural livelihoods in Afghanistan’s Balkh basin, MS Thesis, New Mexico State University. Tsur, Y., Dinar, A., Doukkali, R. M. & Roe, T. (2004) Irrigation water pricing: policy implications based on international comparison, Environment and Development Economics, 9, pp. 735–755. Venot, J. P. & Molle, F. (2008) Groundwater depletion in the Jordan highlands: Can pricing policies regulate irrigation water use?, Water Resources Management, 22, pp. 1925–1941. Ward, F. A. (2007) Decision support for water policy: A review of economic concepts and tools, Water Policy,9, pp. 1–31. 348 F. A. Ward

Ward, F. A., Kealy, M. J., Guy, G. & Manning, L. M. (2008) Efficient conservation measures in irrigated agriculture to sustain urban and environmental water demands, Review of Agricultural Economics, 30, pp. 425–434. Ward, F. A. & Pulido-Velazquez, M. (2008) Water conservation in agriculture can increase water use, Proceedings of the National Academy of Sciences, 105, pp. 18215–18220. Ward, F. A., Michelsen, A. M. & DeMouche, L. (2007) Barriers to water conservation in the Rio Grande basin, Journal of the American Water Resources Association, 43, pp. 237–253. Wichelns, D. (2010) Agricultural Water Pricing: United States (Paris: OECD). World Trade Organization (WTO) (2007) Understanding the WTO. Available at http://www.wto.org/english/ thewto_e/whatis_e/tif_e/utw_chap1_e.pdf Yang, H., Zhang, X. H. & Zehnder, A. J. B. (2003) Water scarcity, pricing mechanism and institutional reform in northern China irrigated agriculture, Agricultural Water Management, 61, pp. 143–161. Zhen, L., Zoebisch, M. A., Chen, G. B. & Feng, Z. M. (2006) Sustainability of farmers’ soil fertility management practices: A case study in the North China Plain, Journal of Environmental Management, 79, pp. 409–419.

Appendix: Detailed Description of Costs and Benefits Private Costs and Benefits The standard of maximum farm income attempts to evaluate impacts of irrigation and its infrastructure maintenance on private income earned by irrigators. It assumes that irrigators ultimately are charged for the cost of infrastructure maintenance. That farm income is measured as:

NYi ¼½PYi 2 Ci 2 PwWi 2 MiLi ðA1Þ where NY ¼ net farm income P ¼ crop price Y ¼ crop yield per unit land (e.g., ha) C ¼ production cost per unit land excluding cost of water and infrastructure Pw ¼ price charged for irrigation water W ¼ crop water applied per unit land L ¼ amount of land in production M ¼ cost of infrastructure improvement per unit land i ¼ index for level of infrastructure investment (0 ¼ none; 1 ¼ low; 2 ¼ high) Equation (A1) presents a framework for assessing investments in infrastructure from the private farm accounting stance. Several terms are directly influenced by the level of infrastructure: These include crop yield, production cost per unit land, crop water applied per unit land, cost of infrastructure itself, and the amount of land in production. For example if a reservoir that is partly silted up is dredged or if its storage capacity is increased in some other way, crop yields will likely increase in the face of a more reliable water supply, the crop mix could change in favor of higher valued crops, production costs may fall, and the amount of land in production could be expected to increase. If the value of these improvements exceeded the cost of the infrastructure improvement, then equation (A1) shows that the investment pays for itself. Financing Irrigation Water Management and Infrastructure 349

National Costs and Benefits Like private farm income, net national benefit also accounts for the economic performance of an irrigation project, but its performance is assessed from a wider point of view. Recent research has shown that several factors important to the nation included in net national economic benefits are excluded from private farm income: the opportunity cost of water displaced from other uses is typically higher than water’s price charged to irrigators; urban values of water associated with a multiple use irrigation project (U) can also be high (Meijer et al., 2006) growing with increased urban populations and with higher urban incomes; environmental values of water (E) can be significant, especially so in environmentally sensitive areas (Chakravorty & Umetsu, 2003). Bearing in mind the distinction between farm income and net national benefits, the latter are measured as:

NBi ¼½PYi 2 Ci 2 PoiWi 2 MiLi þ Ui þ Ei ðA2Þ where NB ¼ Net national benefit P ¼ crop price Y ¼ crop yield C ¼ production cost per unit land excluding cost of water and infrastructure Po ¼ opportunity cost of irrigation water W ¼ crop water applied per unit land L ¼ amount of land in production M ¼ cost of infrastructure improvement per unit land U ¼ urban value of water affected by an irrigation project E ¼ environmental value of water affected by an irrigation project i ¼ index for level of infrastructure investment (0 ¼ none; 1 ¼ low; 2 ¼ high) Equation (A2) recognizes that national related benefits (and costs) can be produced by improvements in irrigation infrastructure beyond those directly accruing to the irrigator. For example impacts on hydropower, ecosystem services, recreation, flood control, and urban water supply can result from improved storage and delivery systems designed for irrigation. The framework described above can be used to evaluate irrigation infrastructure improvements for a wide range of economic, technical, agronomic, climatic and institutional conditions. Equation (A2) also shows that any investment in infrastructure that increased national benefits by more than the cost of the investment is justified economically. For the investments to be sustainable, the economic value of these improvements is ultimately what pays for them.