The only entirely non-volumetric system of water charging is based on the size of the farm. In some Indian states, the land tax depends on whether the land is capable of being irrigated; elsewhere, particularly where a two-part water charge is imposed is, one component is often related only to the size of holding, thereby providing the irrigation agency with a guaranteed source of income. In some areas where rice is the prevalent crop, the water charge is dependent on the area actually irrigated. In perhaps the most common form of charging system, the payment is based on the area irrigated and the crop - so that higher rates may be payable for more water-consuming crops. These various systems in fact introduce a degree of volumetric payment. A farmer who irrigates five hectares has received more water than a farmer who irrigated two hectare, and a farmer irrigating sugar cane uses more water than a farmer irrigating cotton. However, there is no direct linkage between volume and payment - so that the same charge is paid in a particularly dry season as in a very wet season. Assessment based on irrigated area and crop requires considerable resources and effort to carry out. The system is also prone to abuse, particularly collusion between the farmer and the assessor to reduce the charge. Box 6 illustrates some of the practical difficulties in the operation of an effective area-based charge.
|
BOX 6 Assessment based on irrigated area - experience from Sindh Province, Pakistan Irrigation water charging is based on the area cultivated and the crops grown. Much effort and time is needed to collect information on these two parameters for an area as large as Sindh. The basic unit of assessment is the revenue village. Assessment is carried out after each cropping season, theoretically based on field walkthroughs. Farm areas and owners are identified, based on the area maps, which are often outdated. Each farm is then divided into cropped acreage plots and the assessment of each plot is carried out by applying the rate for that crop. Under this method, the revenue officials use their skill and experience, and sometimes judgement, to determine whether a selected acre has produced a full yield or some percentage of full. This figure is used to calculate the water charges. However, the method is open to manipulation and leads to underassessment. There are nine main charge rates, including kharif and rabi crops. Furthermore, rates for government and private lift schemes are double and half the gravity rates, respectively. All these factors increase the opportunities for misreporting. For several years, the assessment by the Revenue Department was double-checked by the Irrigation Department. Assessments were generally found to be as much as 50 percent higher than those of the Revenue Department. There is considerable ill feeling among farmers towards the water charge assessment by the Revenue staff, which makes them very unwilling to pay. The main complaint concerns the arbitrary assessment of the area under cultivation. A common grudge is that, as staff are forced to be lenient towards large landowners, they try to achieve targets by overcharging smaller landowners. |
Other non-volumetric methods described in the literature are: output pricing, where the water fee is levied on each unit of output produced by the user; and input pricing, where a farmer pays for irrigation water indirectly through higher prices for inputs purchased from the government or water agency. Both input and output pricing avoid the need to measure the volume of water diverted or consumed. However, neither measure is favoured by economists because of distortion effects on the price of crops (Rhodes and Sampath, 1988). This review found no evidence of the application of these two methods in practice.
Where the water flow is reasonably constant, charging for time of delivery is an implicit form of volumetric charging. This method is easier to monitor and is practised on many small-scale, farmer-managed irrigation systems. Payment is often in kind rather than in cash, and the main objective is to ensure fairness of distribution rather than efficient allocation of a scarce resource. However, the principles are otherwise similar (Small and Carruthers, 1991).
Abstraction licences for groundwater pumping can function as a proxy for volumetric charging or may depend on volumetric measurements. They are more common in developed countries where individual farmholdings are larger. The farmer meets all capital, operating and maintenance costs of pumps or other infrastructure and in addition pays an annual licence fee for abstraction, either a flat-rate annual tariff based on the pump capacity or a two-part tariff. In the United Kingdom, a two-part licence fee is used. A fixed element, making up 25-50 percent of the annual charge, is determined by the maximum volume that is permitted; the remaining component is determined by the volume actually abstracted (OECD, 1999). This system requires metering of water use.
Volumetric methods charge per unit volume of water supplied at the measuring point (Box 7). They require:
information on the volume used by each farmer or a defined group of users below the measuring point;
a central water authority or water users’ organization to set the price, monitor use and collect fees.
|
BOX 7 Water allocation and rising block tariffs in Israel Israel faces severe water scarcity but water charging is not used as the principal tool to reduce demand in the agricultural sector. The 1959 Water Law nationalized almost all water sources and established the Water Commission Agency to oversee the development and management of water sources and the allocation of water allotments to different users. Priority has been given historically to the agriculture sector, which has support from a strong political lobby. Since the early 1990s, the Ministry of Finance has sought to increase water charges paid by farmers. However, the farming lobby has refused to accept increased prices, pressing instead for the development of additional water sources, including desalinization. At present, farmers receive a water allocation for which they are charged on an increasing block tariff according to the percentage of the allocation used: |
|
|
First 50 percent of allocation |
US$0.18/m3 |
|
51-80 percent of allocation |
US$0.22/m3 |
|
81-100 percent of allocation |
US$0.29/m3 |
|
The Water Commission Agency has responded to shortage by cutting back allocations to the agriculture sector. However, Becker and Lavee (2002) argue a theoretical case in support of water pricing, rather than allocation, to reduce agricultural demand. |
|
Where there are many small farmers and water is distributed in open channels, the costs of installing water measuring devices and monitoring individual users are often prohibitively expensive - especially in systems designed to provide uniform schedules of irrigation over large areas. In these cases, water can be delivered to an intermediate point, e.g. farmers’ organizations, at the head of secondary or tertiary canals, leaving farmers with the responsibility of distributing and charging individuals for water. Favoured by governments and donor agencies, the compromise involves “devolving the most difficult part of the operation, the actual interface between ‘supply’ and ‘demand’, to others [the users]” (Perry, 2001a). Where government agencies are unable to measure or charge individual farmers volumetrically, the assumption that farmers’ organizations will be better equipped to do so is questionable. There is perhaps scope to mix volumetric charging (to a user group) and area-based charging for the members of that group. Sampath (1992) points out that volumetric pricing may be difficult to apply in systems operating under rotational delivery or proportional distribution (a large proportion of the irrigated area in the developing world). There may, however, be some scope for moving towards systems of arranged delivery and volumetric delivery at an aggregated level e.g. to WUAs rather than to individual farmers. This is commonly found in Morocco and other countries where there was a significant French influence on design
Some commentators emphasize the need to charge per unit of water consumed (evaporated or polluted, however that may be determined) as distinct from the volume diverted. They emphasise the point that scarcity is caused by consumption, not by diversion. In most irrigation systems, it is the consumption of water (through evapotranspiration, pollution or loss) that is the key factor in water scarcity, rather than the quantity of water diverted. In some cases, water that is 'lost' from a system will be stored (e.g. in an aquifer) and used later in the season or elsewhere. An example of water actually lost from a system is where it becomes polluted in the recycling process and cannot be used downstream. Rosegrant (1997) adds that water can also be lost if its recovery becomes too expensive, e.g. by percolation to a deep aquifer. However, from the perspective of capital investment and financial management of the system, the volume of water withdrawn is significant, as costs relate to the volume of water diverted.
The magnitude of reuse will vary between basins, depending on geology, topography and levels of demand. The example in Box 8, illustrating water reuse in the Nile basin downstream of the Aswan High dam, shows a situation of intensive reuse. Of the almost 55 bm3 diverted for irrigation, only 38 bm3 is actually consumed, but the 'losses' are used in other parts of the basin through recycling of water from the drains and particularly from the underlying shallow aquifer.
For the foreseeable future, volumetric charging for water will continue to be based on the volume diverted or abstracted from the source, rather than the volume consumed by the crop or lost to a sink. Not only do problems of assessment make alternatives difficult, but the capital and operating costs of a system are determined by the volumes of water abstracted and conveyed, not the volume consumed. To summarize, water scarcity and water pricing to reduce demand (resource sustainability) should focus on the volumes of water consumed while concerns for cost recovery (financial sustainability) are linked to the volume diverted and managed, irrespective of whether it returns to the basin for use by other downstream users.
In practice, volumetric methods of supply to individual farmers are probably not feasible in large parts of the developing world at present because of the costs and complexity of installing large numbers of measuring devices, and the vulnerability of available devices to accidental and malicious damage.
|
BOX 8 Water reuse in the Nile river system downstream of the High Aswan Dam |
|||
|
Data for water year 1992/93 |
|
Billion m3 |
|
|
Inflow to basin |
|
|
|
|
|
Releases from High Dam, rainfall, deep groundwater |
57.94 |
|
|
Reported diversions from Nile and drains |
|
|
|
|
|
For irrigation |
54.97 |
|
|
|
For other uses |
11.17 |
|
|
|
|
66.14 |
|
|
Reuse |
|
|
|
|
|
From drains and shallow groundwater |
40.01 |
|
|
Consumptive use/evaporation |
|
|
|
|
|
Irrigation |
38.03 |
|
|
|
Other |
4.98 |
|
|
|
|
43.01 |
|
|
Proportion of inflows reused |
69% |
|
|
Figures based on data assembled by Keller (Personal Communication, 2004) from different Egyptian government agencies.
Rosegrant and Binswanger (1994) suggest that water markets provide a flexible and efficient way to allocate water while providing incentives that are beneficial for water users. Where water savings are tradable, they provide extra income to farmers, while pricing leads to a reduction in income. This analysis misses the difficult issue of diversion versus consumption. A farmer with a right to divert can, by changing his technology from (say) flood irrigation (40 percent on-farm efficiency) to sprinkler (80 percent on-farm efficiency) reduce his diversions substantially while maintaining the same consumptive use. If his previous excess diversions were contributing to aquifer recharge, and he sells his saved water to a user elsewhere in the basin, the aquifer conditions will change. Politics may complicate markets too: agricultural water users may use their political power to intervene in markets in order to prevent the “logical” redistribution of water out of agriculture to higher value uses. The same authors stress the difference between administered prices, set to reduce demand, and tradable water rights. The value of subsidized irrigation services is often capitalized into land prices. In such cases, raising water charges to reflect the value of the irrigation service is unsustainable for farmers who have paid the premium price for land plus subsidized water. Moreover, the move would be unpopular among other farmers who have profited from cheap water. However, a defined tradable water right, with charges related to the cost of the irrigation service, allows farmers to continue farming or to sell the right at an open market price to the highest bidder. The system has the long-term benefit of allowing water to move to the highest value use, while providing a stable charging environment for the irrigation agency and those not participating in water trading.
Water markets can be formal or informal. Informal water markets around privately-owned wells exist in India, Pakistan, Chile and Mexico. Transactions are typically small scale and local, selling surplus water to neighbouring farmers or towns. There is an extremely well-developed informal market for water in Bangladesh, where water from shallow tubewells, of which there may be 700 000 in the country, is sold to groups of 14-17 farmers. Through this informal market, more than 10 million farmers gain access to irrigation water.
Formal markets in Spain and the United States of America involve tradable water rights, permanent or seasonal transfers or transactions between sectors and jurisdictions. Probably the most advanced system of tradable water rights now in operation is in the Murray Darling Basin in Australia, where diversion entitlements are traded at seasonal and permanent levels with a defined security of availability (Box 8).
The debate concerning water markets focuses on their feasibility (high transaction costs, externalities, lack of legal and institutional framework) and on equity issues. There is concern that poor farmers or households will not be able to pay high prices for water, and that they will be disadvantaged by markets. Referring to small farmers who cannot afford their own pumping equipment, Meinzen-Dick (1997) argues that informal markets increase these farmers’ access to water. Ultimately, markets are rarely undistorted even in sophisticated economies. Market-led systems alone cannot overcome potential conflict between the different objectives of productive efficiency and poverty reduction.
In the water supply and sanitation sector, Boland and Whittington (1998) make a critical assessment of RBTs, which “have become the tariff structure of choice”, being used widely in water supply projects in the developing world. The price charged per unit volume rises with increasing consumption. This mechanism cross-subsidizes the cost of a basic supply to poorer consumers by charging more than the supply cost to wealthier consumers, who consume much greater volumes. Experience cannot be directly transferred between the two water sectors, nonetheless, Boland and Whittington question the value of RBTs on a number of counts and urge the greater use of simpler, two-part tariff structures - a fixed, service charge and a variable charge determined by the volume consumed.
RBTs and simpler two-part tariffs require volumetric measurement of supply. For this reason, they are not in widespread use in the irrigation sector. Such tariffs have been used in Jordan and Israel, where the supply to individual farmers is metered. However, in both countries, central government imposes an upper limit, which determines the maximum volume that may be used for irrigation. Within that allocation, some water saving may be achieved through the incentive of increasing tariffs. Shevah (undated) states that the RBT system has led to irrigation water savings of 10-15 percent in Israel. In Jordan, discussion of pricing reflects concerns over cost recovery in the state-managed Jordan Valley Authority rather than an attempt to use price to control demand (Box 9).
Most irrigation agencies rely on simple, fixed-cost, pricing structures, most frequently based on the area irrigated. Even in an advanced and water-scarce economy, such as Spain, the most widespread charging mechanism is a fixed charge per hectare, although there have been trials of water metering and two-part tariff structures on three Spanish schemes. On these schemes, the variable element reflects the energy cost associated with pumping and pressurizing water delivery systems rather than a direct charge for water according to the volume diverted. Once again, in water-scarce basins, water allocation, rather than price, is used to control demand and ensure adequate supply to industrial, municipal and environmental sectors. In France, Tardieu and Prefol (2002) describe the current system of water quotas for farmers, arguing that wider use should be made of step pricing to ensure compliance with the quota, rather than reliance on metering and financial penalties for over-consumption.
|
BOX 9 The Murray Darling Basin, Australia The Murray Darling Basin, in south eastern Australia, spans three states and comprises an area virtually the size of South Africa. Water supplies are scarce and extremely erratic; soil salinity is severe. In a process that began more than 20 years ago, water rights, based on historic patterns of use, have been formalized so that each riparian has an entitlement, or entitlements, specified in terms of volume and security. Highly secure rights are met, or exceeded, in almost every year; less secure entitlements may only be met in one out of every four years. Thus, variations in the security of entitlement permit allocations that are consistent with the erratic nature of water availability. Salinity entitlements - the rights of an area to export salt - are also specified, and each area must stay within its entitlement, or face significant financial penalties. Water deliveries are measured at the farm gate, primarily in order to confirm that entitlements have been taken. Once water rights had been established and documented, the possibility existed to allow their trade. The system is complex, given the possibility to buy and sell seasonal or permanent entitlements to water, high security entitlements, and low security entitlements. The complexity of ‘definitions’ of water entitlements reveals the body of knowledge and legislation required to specify water rights. An additional complication, being addressed currently, is the possibility to trade water at significant distances, i.e. outside a local jurisdiction - which inevitably involves third-party impacts on river flows, recharge, etc. This process, which introduces interstate trading, is now being implemented. It involves a number of key components:
The entire process of transferring a water right involves 12 steps. To date, water trading is limited. Permanent water transfers are taking place at an average rate of about 1 percent of total availability per year; temporary transfers are occurring at a rate of around 10 percent of total availability per year. Transfers are almost exclusively within agriculture, rather than between sectors. The impacts of this trading have been:
The general lesson is that water trading can work to move water from lower to higher value uses, provided:
|
Source: Murray Darling Basin Commission (http://www.mdbc.gov.au/index.htm).
|
BOX 10 Jordan - managing water scarcity Countries whose renewable water resources are less than 1 000 m3 per capita/year are considered to be severely limited in socio-economic and environmental terms. Jordan has 209 m3 per capita/year. Driven by this scarcity, the main objective of the Jordan Valley Authority, the body responsible for the overall operation of the Jordan Valley irrigation system, is to balance supply and demand between the irrigation sector and the municipal demand of Amman. Within this situation of extreme water scarcity and with increasing demands for water to be transferred from agriculture to meet the growing needs of Amman, the existing volumetric water pricing system is not used primarily as a tool to manage demand. Rather, fixed, volumetric allocations are made to farmers at the beginning of the season. Volumetric charging is expected to give the farmers some sense of the value and scarcity of water but the key objective of charging is to recover O&M costs. Similarly, the water supply to Amman is ensured not by pricing but by clear allocation, assisted by preseason simulation modelling of demand and supply (Huppert and Urban, 1999). Water fees cover approximately 50 percent of irrigation O&M costs and would need to triple in order to achieve full cost recovery. However, there is strong political pressure to keep the fees low. It is hoped that increased water fees will produce increased water use efficiency. However, although “the level of water charge has been subject to constant debate in recent years...the institutional aspect of financing has been touched upon less frequently”. The fact that the assigned maintenance budget is independent of the fees collected reduces the incentive to make the charge work effectively. Despite wide-scale adoption of drip technology, application efficiencies for irrigation water have not improved significantly and distribution efficiency remains low. Farmers perceive the water supply to be unreliable. Thus, when water is available, they tend to over-irrigate in order to store water in the soil, a situation that leads to greater “losses”. |
Under area-based pricing systems, which are most commonly used, farmers pay a fixed fee per unit of land, assessed either on the basis of their total holding, the irrigated part of it or the actual crops irrigated. The system is relatively easy to administer, but is open to abuse, particularly through collusion between the farmer and the assessor to reduce the scale of the charge. Assessment based on irrigated area would appear to be the fairer method, but it requires considerable resources and effort.
Volumetric methods may not be feasible in large parts of the developing world because of the costs and complexity of installing large numbers of measuring devices on the supply to small farmers, and the vulnerability of available devices to accidental and malicious damage. In some circumstances, as practised for example in China, measured volumes of water can be delivered to an intermediate point, e.g. a township or farmers’ organization, giving farmers responsibility for distributing and charging for water. Systems of bulk volumetric charging and area-based charging to group members can then co-exist. RBT pricing depends on volumetric measurement and so is not common in irrigation, particularly in the developing world.
Where the flow of water is reasonably constant, the duration of delivery may be adopted as a proxy for the volume passed.
Formal and informal water markets exist in various parts of the world. In shallow groundwater areas, pump owners may choose between farming themselves or selling a water supply service to others. However, in the absence of defined water rights such systems are pumping markets rather than water markets, with no assurance of sustainable usage. Significant trading in water rights is underway in Australia, underpinned by clearly specified rights and a complex administration system to ensure proper accounting of third party impacts.
