ArticlePDF Available

Cost Recovery and Water Pricing for Irrigation and Drainage Projects

The World Bank
Agriculture and Rural Development Discussion Paper 26
Cost Recovery and Water
Pricing for Irrigation and
Drainage Projects
K. William Easter
Yang Liu
© 2005 The International Bank for Reconstruction and Development / The World Bank
1818 H Street, NW
Washington, DC 20433
Telephone 202-473-1000
All rights reserved.
The findings, interpretations, and conclusions expressed herein are those of the author(s) and do not necessarily
reflect the views of the Board of Executive Directors of the World Bank or the governments they represent.
The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors,
denominations, and other information shown on any map in this work do not imply any judgment on the part of the
World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries.
Rights and Permissions
The material in this work is copyrighted. Copying and/or transmitting portions or all of this work without
permission may be a violation of applicable law. The World Bank encourages dissemination of its work and will
normally grant permission promptly.
For permission to photocopy or reprint any part of this work, please send a request with complete information to
the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA, telephone 978-750-8400,
fax 978-750-4470,
All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the
Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail
Acknowledgments......................................................................................................................... v
1. Why Cost-Recovery Rates Are Low.................................................................................... 1
2. Cost-Recovery Principles..................................................................................................... 5
3. Pricing Scheme Design ......................................................................................................... 8
Area-Based Pricing..................................................................................................................... 8
Volumetric Pricing.................................................................................................................... 10
Water Markets........................................................................................................................... 12
4. Guidelines for Implementing Cost Recovery and Water Pricing................................... 14
Key Factors for Improving Cost Recovery............................................................................... 14
Key Factors for Reducing Water Use....................................................................................... 20
5. Conclusions and Recommendations.................................................................................. 25
Authority and Responsibility to Farmers.................................................................................. 25
Autonomous Water Supply Entity............................................................................................ 29
Equitable and Simple Fee Structure.......................................................................................... 29
Conservation and Reallocation of Water.................................................................................. 30
APPENDIX 1: Consultations on Cost Recovery and Irrigation Water Pricing ................... 31
APPENDIX 2: Case Studies, Summaries, and Lessons Learned ........................................... 32
Brazil......................................................................................................................................... 33
Bulgaria..................................................................................................................................... 34
China......................................................................................................................................... 35
Egypt......................................................................................................................................... 38
India .......................................................................................................................................... 39
Indonesia................................................................................................................................... 42
Iran............................................................................................................................................ 43
Macedonia................................................................................................................................. 44
Mexico ...................................................................................................................................... 44
Morocco.................................................................................................................................... 45
Nepal......................................................................................................................................... 46
Niger ......................................................................................................................................... 47
Pakistan..................................................................................................................................... 47
Philippines................................................................................................................................. 48
Spain ......................................................................................................................................... 49
Sri Lanka................................................................................................................................... 50
United States............................................................................................................................. 51
REFERENCES............................................................................................................................ 52
Box 1. Yangtze Basin Water Resource Project: Self-Financing Irrigation And Drainage Districts
In China........................................................................................................................................... 4
Box 2. Alternative Cost Allocation In Three Projects In India....................................................... 7
Box 3. The Area-Crop-Based Pricing And Its Impact In India And Egypt.................................... 9
Box 4. China: Integrated Circuit Card Automated Irrigation Charge Collection System In
Groundwater Irrigation ................................................................................................................. 15
Table 1. Selected Countries Or Regions Reporting Low Cost-Recovery Rates Or Low Collection
Rates................................................................................................................................................ 2
Table 2. Cost Recovery And Ability To Pay................................................................................ 12
Table 3. Cases Where Different Combinations Of Factors Increased Fee Collection Rate......... 17
Table 4. Cases Where Different Combinations Of Factors Helped Reduce Water Use Per Unit Of
Output ........................................................................................................................................... 23
Table 5. Country Conditions And Irrigation Practices ................................................................. 27
Appendix Tables
Table A2.1: The Case Studies....................................................................................................... 32
This discussion paper is part of the World Bank Agriculture and Rural Development (ARD)
Water for Food team work program, "Re-Engaging In Agricultural Water Management." K.
William Easter is a professor in the Department of Applied Economics of the University of
Minnesota and Yang Liu is a student research assistant. They would like to especially thank Ariel
Dinar and the other World Bank staff members listed in Appendix 1 for their ideas and very
thoughtful suggestions and comments for this paper.
Water pricing
and recovery of the costs of irrigation investment, operation, and maintenance
have been contentious issues for many decades. The low charges for irrigation water are
questioned, as well as, the small percentage of farmers who actually pay the charges. In some
projects, fee collection rates are near zero, even when water charges are well below the cost of
project operation and maintenance (O&M) (Ahmand 2002; Easter 1993; Govt. of People’s
Republic of Bangladesh 2000; Svendsen et al. 1997). This creates serious problems both for
irrigation agencies and, in the long run, for farmers. If the fees collected do not cover the costs of
an irrigation project, its sustainability, without continued government subsidies, may be at risk.
Water fees are collected from farmers for two main reasons. The first is to cover the O&M cost
so that the project is financially sustainable. In many cases, fees will also need to include a
charge for the cost of capital required to construct the project. This charge for capital is important
for future irrigation investments. The second objective involves pricing to encourage farmers to
use less water per unit of output or produce greater net economic returns per unit of water, or
both. Historically, the first objective has been paramount, but as water scarcity increases, the
water use efficiency objective is likely to grow in importance and be given a higher priority.
Efficiency is used in the economic sense: maximizing benefits subject to technical and physical
limitations (Prato 1998). Efficiency in water use means maximizing society’s benefits over time
from the water and technology available. In practical terms, improving water use efficiency
means increasing the value of crop output per unit of water consumed through evapotranspiration
(ET) by plants.
To illustrate the problem, consider several countries that have historically had a poor record of
collecting water fees. In 1984, collection rates of assessed water fees were only 20 percent and 8
percent, respectively, in Nepal and Sri Lanka, while it was 38 percent in Turkey in 1989–94
(Easter 1993). In his 1995 study, Jones rated cost recovery in 105 of the World Bank’s 208
irrigation projects and found 68 percent unsatisfactory and only 32 percent satisfactory (see
Table 1 for further examples). Considering this poor record, it is not surprising that the World
Bank and others have been trying to identify the reasons for these low collection rates and ways
to improve them.
The objective of this paper is to develop guidelines for improving cost recovery and reducing
water use per unit of output. The guidelines were developed from a review of studies of irrigation
reforms and interviews of 20 World Bank staff members with responsibility for irrigation sector
reforms in countries from Asia to Latin America (names listed in Appendix 1). From these
studies and interviews, we distilled specific reforms that are important in improving cost
recovery or reducing water use, or both. Reforms needed by individual countries or projects will
depend on their institutional arrangements as well as the type of irrigation and its physical
The second section of the paper begins by listing some reasons for low collections and provides
an overview of some of the reforms that have raised collection rates. In the next section, we
review cost-recovery principles and provide some examples of what different countries have
done to recover project costs and collect water charges from users. The focus in the fourth
section is on designing water charges or water markets that will give farmers an incentive to
make better use of their water by reducing the amount of water used per unit of output. In the
fifth section, we use case studies to determine what reforms can help improve cost recovery and
increase collection rates. In addition, water pricing reforms are identified that will encourage
farmers to reduce their water use per unit of output. The final section provides a summary of the
reforms that are important for increasing cost recovery and encouraging farmers to improve their
use of water.
In developing countries as well as in many developed countries, there are many different reasons
for low water fee collection rates including the following:
No link between fees collected and funds allocated to an irrigation project
Lack of farmer participation in project planning and management
Poor communication and lack of transparency between farmers and irrigation
Poor water delivery service (timing, duration, or quantity inadequate) and no penalties for
managers and irrigation project personnel who provide poor service
No user penalties for nonpayment of water charges
Low priority given to fee collection, efficient water use, and system O&M
Small size and very low incomes of irrigated farms
Corruption of irrigation officials.
The list illustrates that the cost-recovery problem is, at least partly, an assurance problem:
assurance regarding what water users will do, as well as assurance concerning what water
managers and their staff will actually do as opposed to what they say they will do (Easter 1993).
Table 1. Selected Countries Or Regions Reporting Low Cost-Recovery Rates Or Low Collection Rates
Country/region Collection rate Percentage of cost recovered Comments
Argentina 1997
(Svendsen et al.
70 percent 12 percent of O&M
Water charges are very low and based on area: fixed area fee of
$70/ha/year. Fee collections are managed jointly by the government
and the water user associations.
Bangladesh 1998
(Government of
People’s Republic
of Bagladesh 2000)
3 to 10 percent Low
In 1997–98, water charges were levied in only 6 of the 15 major
irrigation schemes.
Botswana 1994–95
(Thema 1997)
44 percent of the O&M in 1995;
cost-recovery rates have been
between 35 and45 percent since
1988. Government pays capital
Increasing block pricing system. By the end of 1996–97, revenue was
scheduled to recover O&M costs, but by 1995 charges were too low to
cover these costs.
Jaiba Project ,
Brazil 1995
(Azevedo 1997)
66 percent 52 percent of total costs
The two-part water charging system is well designed, but collection
rates are too low.
Columbia 1996
(Svendsen et al.
76 percent 52 percent of O&M
Responsibility for fee collection has been shifted to water user
associations. The transfer was too quick, with too little time and effort
invested in clarifying water users’ rights and responsibilities.
Maharashtra, India
1984 (Easter 1997)
58 to 67 percent n.a.
There is no link between fees and funds allocated for O&M. There are
penalties for default payments, but neither user participation nor
incentives for service providers to collect fees.
Italy 1997 (Destro
n.a. 60 percent of total costs Water charges are too low and based on area.
Jordan 1999
(Rupert and Urban
n.a. 50 percent of O&M
Water fees are too low, services are not related to water charges,
meters are broken, and the volume of water used is deduced from an
assumed discharge rate instead of using the meters.
Macedonia 2000
(Hatzius 2000)
42 percent n.a.
There are no incentives for service providers to collect fees, and user
penalties are not effectively enforced for nonpayment.
Nepal 1984 (Easter
20 percent n.a.
There is no link between water fees and O&M. Fee collection is not
given high priority. There are no incentives to collect fees and no
Country/region Collection rate Percentage of cost recovered Comments
enforcement of penalties for nonpayment.
Pakistan 2001
(Ahmad 2002)
30 to 35 percent n.a.
Revenue from water charges is pooled with other taxes and goes to
provincial treasury. There is no clear link between fee payment and
service provided.
Philippine 1995
(Svendsen et al.
58 percent 46 percent of O&M Area-crop–based water charges, US$77/year/ ha on average in 1997.
Sri Lanka 1984
(Easter 1993)
8 percent n.a.
Communication between farmers and irrigation officials is poor. There
is no clear responsibility for O&M.
Tunisia 1991
(Hamdane 2002)
National average is 70 percent of
O&M costs; ranging from 44
percent in the Central region to
76 percent in the Northern
Water charges are too low and the public agency managing irrigation is
not financially autonomous.
Turkey (Dinar and
Mody 2004)
76 percent (1998)
32 percent in 1991
37 percent in 1998, and 50
percent in 1985, the highest rate,
Two-part pricing system. O&M fees are area-crop–based. Capital cost
recovery is low and does not start until 10 years after project
completion. There is no inflation adjustment despite a 70 percent high
inflation rate.
Note: O&M = operation and maintenance
Source: listed in table by country/region
The main reasons for low cost-recovery rates are likely to vary among countries and even among
projects within one country. The good news is that steps can be, and have been, taken to correct some
of these problems. For example, more and more countries have started to encourage water user
participation by establishing water user associations (WUAs). In many cases, system management
turned over to farmers. Both Turkey and Sri Lanka have used management turnover as a means of
improving performance. For Turkey, it helped increase cost recovery to 76 percent in 1998 (Table 1).
Sri Lanka did not experience any significant improvement in productivity with turnover, however,
except where it was combined with project rehabilitation (Samad and Vermillion 1998). In addition,
there is no direct cost-recovery system in Sri Lanka. Instead, the government feels that it gets indirect
cost recovery by transferring management to farmer organizations that have full responsibility for
O&M below the head of the distributional canals
Other steps taken to improve cost recovery include improved irrigation service and more transparent
decisionmaking. The success of these steps depends on government policies as well as institutional
arrangements including the basic legal system. The emphasis should be on getting farmers to accept
the idea that they should pay for irrigation water and agree on what the cost should include. Farmers
will not want to pay if service is poor and the fees paid bear no relation to system O&M. Incentives are
also needed to encourage better service delivery and raise farmers’ willingness to pay. For many
countries, a combination of actions will be needed, as illustrated by the World Bank–supported
Yangtze Basin Water Resource project (Lin 2003).
The goal of the Yangtze reform was to establish a self-financing and self-managed system consisting
of two integrated parts: the water supply corporation (WSC) supplying water from the headworks and
the water user associations operating at the local level. Farmers at a water users’ conference elected a
WUA executive committee. Villages helped mobilize community human resources for both the WUA
conference and the executive committee election, which helped improve farmer participation and
WUA operations. The WSC charter also requires farmer representatives from the WUAs to be on the
WSC board of directors, so that farmers are directly involved in WSC management and
decisionmaking. Irrigation project authorities and local water bureaus constantly held training
programs for farmers, which helped enhance the WUAs’ operational capacities after transfer of the
local irrigation systems. As a result, farmers are actively trying to improve their system through strong
user participation, although in some areas village leaders have tended to dominate associations (see
Box 1).
Box 1. Yangtze Basin Water Resource Project: Self-Financing Irrigation And Drainage Districts In China
Control of local irrigation by water user associations (WUAs) has saved both water and labor, as well as shortening the
irrigation time cycle. Water is now delivered on time and in the right amounts. This increased availability of irrigation
water has reduced farmers’ incentives to steal water. WUAs improved system maintenance in both the main canal and
the lower distribution network. Farmers are investing labor and funds in the projects, because they now view them as
their own. Irrigation costs have been reduced in multiple ways since the introduction of WUAs and volumetric pricing.
Farmers are using less water per hectare, and delivery losses have dropped. On average, each WUA has saved about 1.18
million m
of water annually, and productivity has improved. After introducing the WUAs, the average crop yield
increased by 6 percent, of which 2.5 percent was purely due to irrigation improvements. Improved delivery of irrigation
service has also helped reduce poverty. In WUA areas, even poor farmers are now able to secure their harvest regardless
of the weather.
However, Zhang et al. (2003) have pointed out some problems concerning the representation of users in some WUAs
and decisionmaking.
First, in terms of user participation, farmers are largely involved in construction and maintenance of irrigation
infrastructure, but not in the decisionmaking process of water allocation. Also, there are no checks and control
mechanisms among the WUAs. The majority of participants in the WUA elections are village officials instead of
ordinary farmers. In most places, village officials are still responsible for collecting water charges. About 80 percent of
the farmers sampled think that WUAs, not village officials, should have the legal right to collect fees, because some
village officials divert water fees for other purposes. WUAs finances are not transparent to most farmers. Finally, few
farmers understand the basic WUAs concepts. They still think the WUA is just an extension of government and do not
feel it is their own organization.
Source: Lin 2003.
Improving cost recovery clearly involves more than just charging higher fees or spending more on fee
collection. However, which water costs are to be recovered and what mechanisms can be used to
recover them have to be specified. The full costs of providing irrigation water can be divided into three
categories: direct project costs, environmental costs, and marginal user costs. Direct project costs are
the easiest of the three to measure, and most projects take only direct costs into account in determining
cost recovery. Direct costs refer to costs stemming from the process of capturing and delivering
irrigation water, which can be broken into fixed costs and variable costs. Fixed costs include all
investments in irrigation infrastructures such as building reservoirs and canals and installing meters
and pumps, plus depreciation and interest payment on the investment. Higher level administrative
costs and some operational and maintenance costs not involved with actual water delivery are also
considered fixed costs because they do not vary with the amount of water delivered. Variable costs
consist of the operational and maintenance costs of water delivery, lower level administrative costs
(usually temporary labor costs during the time of water delivery), and costs of supplying water, which
include conveyance costs, groundwater extraction costs, and costs due to water loss. These costs vary
with location, water delivery method, irrigation technology, and season (Massarutto 2002).
Environmental costs include soil erosion and damage to the surrounding ecosystem during and after
the construction of an irrigation project as well as waterlogging and salinity problems caused by the
irrigation. However, few irrigation projects in practice include environmental costs as part of their full
cost to be recovered. Environmental costs could substantially raise the total costs of many irrigation
projects. South Africa is developing a system of charges that will reflect and recover direct and
indirect costs associated with the discharge or disposal of waste. The charges will include a load-based
charge proportional to the waste load. Initially, this charge will relate to salinity, nitrates, and
phosphorous in the water discharged. An extra charge will apply if the waste load exceeds the
maximum permissible level. Also, rebates will be provided for returning water to the source at a higher
quality than when it was abstracted (Republic of South Africa 2004).
In South Australia, the government has agreed to cover the costs of salinity management caused by
pre-1988 irrigation development, while farmers (irrigators) will be responsible for the costs associated
with all the post-1988 irrigation development. In addition, the current two-part price structure can be
adapted to accommodate environmental externalities. When infrastructure has to be renovated or built
to reduce water quality–related externalities, the fixed costs can be captured in the fixed portion of the
two-part price. Quantity-related externality costs can then be included in the volumetric portion of the
two-part price (Bueren and MacDonald 2004). However, in Queensland, when the government
reviewed water resource charges in 2004, most communities rejected the idea of introducing an
externality charge. They argued that many water users already provide infrastructure to mitigate
externalities, and that further improvement should be achieved through regulatory planning
(Queensland Government 2004).
Marginal user cost is defined as the present value of future sacrifices implied by current resource use
(Howe 1979). It involves the higher costs of obtaining future water supplies because more accessible
and less expensive water resources are used up first. In an extreme case, a water resource is completely
used up in the current period. This cost is especially relevant for groundwater resources with little or
no recharge. Excluding marginal user costs in the price of groundwater often results in overuse of the
After determining which of these costs to include, the next concern is what percentage of total costs
should be allocated to farmers. In many cases, who should bear the costs of providing irrigation water
is not clear. Whether the farmers should pay the full costs depends on factors including project
objectives and the number of beneficiary groups besides irrigated farmers. Irrigation projects serve
multiple beneficiaries in two major ways. One case is multipurpose projects; the other is projects
involving indirect beneficiaries of the increased agricultural production.
Multipurpose water projects are common. Besides supplying irrigation water, projects may also supply
water for household and industrial uses as well as providing flood control and hydropower. In Asia, 90
percent of dams for irrigation are multipurpose. In these cases, different users should share the costs in
proportion to the services they receive. There are three common methods for allocating costs among
users: the use of facilities (UOF), alternative justifiable expenditures (AJE) and Separate Costs,
Remaining Benefits (SCRB) methods (Easter 2003; Young et al. 1982; Young 1985). The first
approach, UOF, allocates costs among different types of users sharing the same facility in proportion
to the water delivered to each type of user (e.g., irrigation and domestic water supply). The second
approach, AJE, allocates joint costs based on remaining benefits after subtracting specific costs, where
specific costs refer to costs directly attributable to a single purpose (e.g., irrigation) and exclude the
costs of a change in project design due to the inclusion of a particular purpose. AJE easier to calculate
than SCRB because it relies on specific
costs rather than separable
costs. The third approach, SCRB,
is similar to the second one. It assigns costs that serve a “single” purpose to the benefiting purpose,
including the costs of any project design changes required to include the added purpose. The
remaining “joint” costs are assigned in proportion to the remaining benefits derived for each type of
use after subtracting the separable costs. An irrigation project in Andhra Pradesh, India, provides a
good example of how the costs from a multipurpose water project can be allocated among different
types of uses or purposes (see Box 2).
In projects with large indirect benefits, some of the costs may be allocated to the beneficiaries. For
example, in countries where the government pursues a low food price policy, food processors and
consumers both may benefit more from irrigation improvement projects than farmers. In such cases,
subsidizing the project through tax revenue from the benefiting consumers and processors might be an
alternative to help fund the project.
Box 2. Alternative Cost Allocation In Three Projects In India
Two alternative cost allocations were calculated for the distribution of project costs. The first allocation is based on
the quantity of water delivery for each purpose or use. Since the allocation is based on water delivery, only the three
consumptive uses are allocated a share of the costs, with between 95 and 98 percent of the cost allocated to
irrigation (Table A). When the costs are allocated based on benefits generated, all five major water uses are
allocated costs, and irrigation’s share drops to between 88 and 94 percent (Table B). Thus, in multipurpose projects,
irrigation is likely to be allocated a major share of the costs but, with growing domestic and industrial demand for
water, irrigation’s share is likely to drop significantly over time. In projects that include an important flood control
component, irrigation’s cost-share would drop even more.
Cost Allocation for Three Consumptive Uses Based on Water Delivery
Three water projects
Domestic water supply
Industrial(percent) Irrigation (percent)
Nagarjursagar 2 0 98
Tungabhadra 1 4 95
Sriram Sagar 2 3 95
Cost Allocation Among Three Projects Based on Benefits
Three Water Projects
Purpose or use Sriram Sagar (percent) Nagarjursagar (percent) Tungabhadra (percent)
Irrigation 88.1 94.3 91.3
Hydropower 3.0 4.0 4.2
Domestic 3.0 1.6 2.1
Industry 4.3 0.1 2.3
Fisheries 1.6 0.1 0.1
Source: World Bank 2003b.
The Sana’a basin water management project in Yemen illustrates the idea of government cost sharing
where a major objective of the irrigation improvement program is to reduce the rate of groundwater
exploitation. The strategy is to reduce the speed of groundwater mining and extend the useful life of
the aquifers to gain time for the government to find long-term solutions. Examples of such solutions
might be to shift the focus of the basin’s economy to less water-intensive activities and to encourage
out-migration from areas with groundwater mining. In the short run, farm-level water conservation
practices are being subsidized in areas irrigated by groundwater. For example, the government
introduced piped conveyance and distribution systems, as well as drip and bubbler technology, in the
pilot areas. Irrigation efficiency was improved from 35 percent to 60 percent. The government is
responsible for 75 percent of the investment costs and 90 percent of the installation costs; farmers are
responsible for the rest of the costs. Farmers also pay 100 percent of the O&M costs. Such
cost-sharing rates have encouraged more farmers to participate in the project and reduced the amount
of water used per hectare (World Bank 2003a). Senior Water Resource Management Specialist M.F.
Abu-Taleb of the World Bank (co-task team leader on the project) reported that WUA involvement in
the implementation of the irrigation project was overwhelmingly positive. Users have signed up to pay
for and install the subsidized water conservation equipment in the pilot phase.
To achieve the two primary goals of cost recovery and reduced water use per unit of output in
irrigation water management, two key issues must be addressed: first, to design an effective pricing
mechanism based on local conditions and, second, to develop a strategy for obtaining high rates of
In this section, we focus on the three major methods for pricing water: area-based pricing, volumetric
pricing, and market equilibrium pricing. Emphasis is placed on the first two categories, including
extensions, modifications, and combinations of the two.
Area-based water charges are fixed charges, based on the area irrigated or “supposed” to be irrigated.
They are often calculated by dividing the total area irrigated into the O&M costs of providing
irrigation water, which basically follows the average cost pricing principle. Defining O&M costs is
important because the water supply entity may have an incentive to inflate the costs charged to farmers.
In addition, the use of irrigated area varies from year to year and season to season. For example, the
area irrigated during the wet season is usually much larger than during the dry season. In addition, the
project area is usually larger than the area actually irrigated. Therefore, irrigation officials will need to
estimate the area actually irrigated each season.
The disadvantage of this pricing method is that, once the irrigated area decision is made, the water
charge will have no effect on farmers’ water consumption, because the marginal cost of applying
additional quantities of water per hectare is zero. Thus, the demand for water is usually higher than it
would be under a price or charge that varied by the quantity of water used, and it is likely to lead to
overuse of water by farmers near the head of the canal.
The advantage is that it is simple to calculate, easy for farmers to understand, and the implementation
costs are lower than for volumetric pricing because water deliveries do not have to be measured. Also,
assuming 100 percent collection rates, charges per hectare, based on average direct cost, result in full
recovery of direct costs. Although it gives farmers no incentive to reduce water use per hectare, it is
still widely used in many systems throughout the world due to the simplicity of its implementation. In
Haryana, India, irrigation water is priced at US$2.50 /ha
(Cornish and Perry 2003), while in Pakistan,
prices range from $2 to $8/ha but are set to cover only part of the O&M costs (Ahmad 2002).
Pure area-based pricing is appropriate in places where water is not scarce, where crops are not varied,
and where meter installation is difficult or costly. However, pure area-based pricing systems are
becoming less and less popular, and most of the recently designed area-based pricing systems are
adopting new features. The extensions of area pricing include area-crop (the most widely used
modification), area-irrigation, area-season, and area-technology–based pricing.
Area-crop–based pricing systems vary the charge per hectare irrigated by type of crop. The water price
variation among crops depends on the policymakers’ objectives. If they want to encourage efficient
use of water, the high water-consuming crops such as rice, should have higher prices per hectare. If the
price differences are large enough, farmers are likely to switch to alternative crops. Box 3 gives
examples of how area-crop–based pricing can induce crop changes that save water. In contrast, if the
government is pursuing a low food price policy or wants to encourage production of commercial crops,
the water price for these crops could be set lower than for other crops. However, care must be taken in
subsidizing inputs such as water to increase crop production because it often leads to inefficiencies and
overuse of the resource, particularly with crops such as rice and sugarcane.
Box 3. The Area-Crop-Based Pricing And Its Impact In India And Egypt
Haryana, India: An Empirical Illustration of The Performance of Different Water Pricing Methods
A numerical example using data from Haryana, India, illustrates the impacts of two different pricing methods. Farmers can
choose to grow cotton or wheat; cotton is more profitable but uses more water than wheat. Two pricing methods are
considered: area-crop–based pricing, with a higher price for irrigating cotton, and volumetric pricing. For area-crop–based
pricing, the implementation cost is low, but high for volumetric pricing. An area-crop-based water charge of $231/ha for
cotton and nothing for wheat induces the farmers to switch from cotton to wheat. Farmers profit decreases by one fourth,
but the social benefit increases almost sevenfold. This example illustrates how a simple method, area-crop–based pricing,
can be more effective than an efficient but complicated volumetric pricing, when implementation costs are high. The
implementation costs include both the fixed costs of installing meters and the added variable costs required for water
delivery and monitoring.
Egypt. The Impact of Different Pricing Methods on Irrigation Water Consumption and Farm Income
Several studies in 1995 by International Irrigation Management Institute (IIMI) measured the impact of different pricing
alternatives on the agricultural sector in terms of irrigation water used and farm income. Three pricing schemes were tried.
First, a fixed rate of $52 per hectare, irrespective of crop or water use, resulted in a fall in farm income of 4.5 percent but
had no effect on the choice of crop or technology. Second, an area-crop–based charge, proportional to the calculated
average water consumption of each specific crop, resulted in a 2.4 percent fall in farm income. The demand for irrigation
was water reduced by 3.5 percent and the returns to water increased by 2.7 percent. Third, a volumetric charge based on the
quantity of water delivered resulted in virtually identical impacts as those obtained in the second case. The key factor
explaining the different responses appears to be the availability of a range of crops that farmers can choose to grow.
Sources: India, Tsur and Dinar (1998); Egypt, Perry (1995).
In the area-irrigation method, water charges usually reflect the differences in water delivery costs
among different irrigation methods. For example, most gravity-based irrigation systems have much
lower variable costs than pump irrigation. The advantage of pump irrigation is that water control and
measurement of water delivery is generally much easier than it is for most gravity flow systems. Thus,
area charges are usually higher for pump irrigation because irrigation costs and net income per unit of
water are generally higher.
Some countries also use area-season–based charges. For example, a higher price is charged during the
dry season, when water is scarce, and a lower price is levied in the monsoon or wet season, when
water is relatively plentiful. If the price is set high enough in the dry season, it will help limit the
number of hectares irrigated in that season. In France, the pricing structure was based on different
costs for off-peak and peak water use. The peak season lasts five months in the summer, and the water
price reflects the long-run marginal cost of supplying water. The long-run marginal cost is usually the
cost of future expansion. In reality, it is often difficult to estimate the cost of the next big
supply-capacity-expansion project (McNeill and Tate 1991). During the off-peak seasons, France
includes only operating costs. This pricing structure has helped reduce water use during summer when
demand is high compared to supply (Tiwari and Dinar 2003; Johansson et al. 2002).
Another possible combination is area-technology–based pricing. Although it has not received much
attention, theoretically it should promote selected irrigation technologies. The basic idea is similar to
area-crop–based charges, with farmers using water-saving technology paying lower per hectare water
charges. For example, drip and sprinkler irrigation generally allow better water control and more
output per unit of water delivered than flood irrigation. Therefore, a higher per hectare fee could be
levied on farmers not using these technologies to encourage them to switch.
If area-based charges can be established that reflect differences in water use by season, crop, or
irrigation technology, area pricing would have some of the benefits of volumetric pricing. This would
be the case if, after controlling for crop, irrigation technology, and season, there was little variation in
water uses per hectare. Problems are still likely to exist because farmers at the head of the canal tend
to overirrigate their fields when water charges are based on area. However, if farmers can be assured
that each scheduled water delivery will be on time and in the quantity demanded, they will have much
less incentive to overirrigate than with irregular deliveries. These and other incentives will be
discussed in more detail below, when guidelines for implementing effective water pricing are
With volumetric water pricing, the charge is based on the amount of water delivered. The economic
optimal pricing rule requires that price should be set equal to the marginal cost of providing the water,
and it requires accurate measurement of water through meters. The advantage of this pricing method is
that it encourages farmers to limit their water use. Also, it is easy to understand in the sense that you
pay for the quantity of water delivered to your farm. However, it has several disadvantages. First, the
implementation costs can be high because meters are required, and they have to be honestly read and
reported. Second, marginal cost pricing does not allow full cost recovery in the case of decreasing
average costs (e.g., large canal systems). Once the infrastructure is in place, the marginal project costs
will be lower than average costs, thus pricing based on the marginal cost will not achieve full cost
recovery. In contrast, for the case of pump irrigation using groundwater, the marginal project costs are
likely to be higher than average project costs, particularly when marginal costs include the marginal
user cost. Thus, for some groundwater projects, marginal cost pricing could result in overcollection as
well as high water charges relative to farm income. For example, in a deep tubewell project in Gujarat
India, the water fee is 37 percent of net farm income and does not cover O&M costs because
electricity is heavily subsidized. In contrast, for gravity-based systems the water fee is usually a much
smaller percentage of net farm income (see Table 2). To address the concerns about the impacts of
water charges on farm income, two different modified versions of volumetric pricing can be used.
Block pricing
Block pricing involves varying the water price when water use for a set time period exceeds a set
volume (e.g. 5,000 m
per hectare per season). If high water charges are a concern, an increasing block
charge can be used. The price of the first block can be set below O&M costs. The second and later
blocks are raised to higher rates that cover O&M costs and reflect the marginal cost of operations.
Israel (Yaron 1997) and Botswana (Thema 1997) both use this pricing method. The amount of the first
block is often considered the basic amount of water needed to support a farm family, so this method
also attempts to address equity issues. Farmers pay a low rate for the first block but a much higher
price for any water used that exceeds the first block. In Botswana, the price of the second block is
twice the price of the first block. This pricing method operates similarly to a quota. In fact, a quota is
an extreme case of increasing block pricing. Even when an official quota exists, farmers can still
obtain additional water by paying irrigation officials or private sources a high enough price. Both
Botswana and Awati in China call their first block a “quota” and state that farmers have to pay double
the price if their consumption exceeds the quota. In Israel, the quota includes three blocks, and charges
are agreed to in signed contracts with the water provider
If the price difference between blocks is large enough, farmers will try not to use more than the first
block of water. The disadvantage of block pricing is that it is not easy to decide the price level for each
block or the quantity range of each block (e.g., should the low price apply to the first 5,000 m
per season per hectare or the first 6,000 m
). In addition, the revenue is unlikely to cover the O&M
costs, particularly if the range for the first block is large. It is appropriate to use in cases where water is
scarce, farm incomes are low, and water charges are high relative to net farm income (see Table 2).
The advantage of the two-block pricing is that you have, at least, three instruments for influencing
water use and cost recovery: the first and second block prices and the quantity (e.g., 4,000m
/ha. vs.
/ha.) at which to start the second block price.
Two-part tariff
The second modification is a two-part charge, which is a combination of volumetric pricing and a
fixed admission charge (sometimes based on size of the area irrigated). For the block pricing methods
described above, the two objectives—full cost recovery and reduced water use—are often in conflict.
The advantage of a two-part charge is that it can reconcile the conflict. The volumetric part can be
based on marginal cost, which encourages less water use, while the fixed part can be used to make up
any deficits and ensure a certain revenue flow regardless of how much water is available and delivered.
Even for O&M costs, there is a fixed component that does not depend on the amount of water
delivered, and these fixed costs have to be paid even when water is not used for one season. The
disadvantage is that it is relatively hard to calculate and difficult for farmers to understand. In addition,
the administrative costs of a block-pricing scheme are likely to be somewhat higher than a single
charge scheme.
Table 2. Cost Recovery And Ability To Pay
Water fee as
percent of
net income
Water source Collection rate and cost recovery
Gujarat, India 37
No serious payment default problems, but electricity is heavily
subsidized. The charge for electricity is far below the full
cost-recovery level.
0.5 Dams
High collection rate (90 percent or more). The irrigation
department is able to achieve full O&M costs recovery by
allocating only 33 percent of the overall costs to irrigation and
charging a very low price (less than one-twentieth the costs to
industrial, municipal, and other users). Even if the allocation
was based on the proportion of water delivered, fees would be
only about 2 percent of net farm income.
7 Dam
The collection rates are between 60 and 70 percent. Because it is
a relatively new project, the O&M costs are as low as $30/ha/yr,
but the full cost including capital repayment and depreciation is
$54/ha/yr., almost double the O&M. Project-level data indicate
that subsidies are substantial because, on the extensive
“traditional” area, users do not pay the charge. In other areas, to
reduce their payments to 40 percent of the designated charge,
many farmers contribute their labor for cleaning the canals.
15 Dam
The collection rates are between 70 and 80 percent. Because this
is Morocco’s first and oldest irrigation project, its O&M costs
are much higher than in Haouz ($127/ha/yr), and the full costs
are ($150/ha/yr). Official statistics indicate that total collections
exceed the O&M costs.
The collection rates are poor, less than 30 percent of billings.
Even at 100 percent collection rates, the current charge is still
too low to cover the O&M costs. An assessment of ability to pay
O&M costs was made in 1995. The result suggested that, despite
the relatively low net farm income, farmers can pay for
irrigation services. The collection rates are low because farmers
are unwilling to pay due to corruption among irrigation officials
and discontent over service quality and system transparency.
Note: O&M = operation and maintenance.
Source: Cornish and Perry 2003; Ahmad 2002. For more information about those cases, see Appendix 2.
In the Jaiba project in Brazil, the pricing scheme is a revised two-part charge, consisting of two
components, K1 and K2. The first component, K1, reflecting the project’s capital cost, is calculated
based on a 50-year repayment period and a subsidized interest rate. The second component, K2, is
supposed to cover all the O&M costs and is estimated as a function of the volume of water used. The
second component is further divided into two components: one representing the fixed O&M costs and
the other, the variable costs. Farmers choosing not to grow crops for one season are still responsible
for the fixed O&M costs (Azevedo 1997).
Besides the above pricing methods, Johansson et al. (2002) also suggest using output-based pricing
methods. They also summarized many case studies from around the world, using different pricing
methods to allocate irrigation water.
In countries with water markets, formal or informal, companies or individuals, can trade water at a
particular market equilibrium price that will likely change throughout the season. To operate
effectively, water markets require a well-defined structure of water rights, a clear and comprehensive
set of rules for trading, an entity to manage water delivery, and a judicial body to oversee trading
activities and resolve disputes. They also require a well-developed conveyance system for transporting
water to all participants (Tsur and Dinar 1998). If these requirements are in place, market equilibrium
prices will effectively adjust supply and demand.
In Chile, analysis in two river valleys shows that water markets produced substantial economic gains
from trade (Hearne and Easter 1995). The water market that has been operating in the Northern
Colorado Water Conservatory district since the late 1950s is another example. The Northern Colorado
Water district maintains a bulletin board where potential buyers and sellers post offers (Howe 1997).
There has been a market in both permanent water rights and temporary sales just for seasonal water
use. Over time, the permanent rights have gradually gone from agricultural to municipal and industrial
use. However, agriculture is still by far the major water user (Kemper and Simpson 1999).
Localized markets have developed in a number of different settings in other parts of the world. The
water market in Siurana-Riudecanyes irrigation district in the Tarragone province, Spain, is a classic
example and is somewhat similar to the water market in Alicante, Spain (Maass and Anderson 1978).
The Siurana-Riudecanyes system serves farmers as well as municipal and other users and delivers
about 6 million m
of water annually. Water use rights, both long-term and temporary, are traded
among WUA members, which include both farmers and municipal users. In 1982, an official exchange
administered by the WUA was formed, which significantly reduced the volatility of water prices and
made the exchanges much more transparent. A system of bonuses and incentives was also established
for WUA employees to minimize water losses and reduce O&M costs.
The city council of Reus, the major city in the Siurana-Riudecanyes irrigation district, has played a
significant role both in the water market and in the original construction of the water system. The city
provided a large part of the funds to construct the dams and infrastructure. It financed 50 percent
directly and another 40 percent with a loan to be repaid by the beneficiaries. The direct beneficiaries
provided the remaining 10 percent of the construction funding upfront. Yet, it is the WUA that is the
central focus of the water system, with active user participation, transparency, and water trading
providing the flexibility needed to respond to changing water and economic conditions (Tarrech et al.
Another water market was developed by farmers in the Cariri region of the Ceara state in northeast
Brazil, based on a spring-fed river. Rights to the water were allocated to farmers by farm size. Water is
traded separately from land and enforced by farmers themselves. Generally, the trading system
provided water rights holders with a secure water supply and flexibility in the allocation of water. The
market was set up without any direct government involvement because it consists of a relatively small
number of fairly homogenous farmers. They all grow sugarcane and know each other very well
(Kemper et al. 1999).
An essential part of any cost recovery or water pricing strategy is implementation. In this section, key
factors that influence success and failure in implementing different cost-recovery and water pricing
strategies are discussed.
There are two key steps in cost recovery: the first is to design a pricing mechanism that covers the
appropriate costs; the second is to achieve high collection rates through effective water management.
The design involves working with the water supplier and farmers to determine what should be
included in the costs, and which of these costs should be collected through a water fee rather than
through other taxes (such as a land tax or a local property tax). Once this decision is made, setting the
appropriate fee level becomes an accounting problem influenced by the type of irrigation system and
ability to measure and monitor water use. As discussed above, when the volume of water delivered
cannot be measured, water charges are usually based on some measure of area irrigated. Sometimes
the area-based charges are adjusted to account for crops grown and season of the year. Even if the
appropriate water charge is determined, the more difficult step still remains: achieving high collection
A key to achieving high collection rates, suggested by both literature and field experience, is financial
autonomy (see Box 1). Without autonomy, collecting sufficient funds from users does not guarantee
improved O&M services because revenues from water charges, in many cases, do not go back to the
project. Instead, they are commingled with other taxes in the central treasury, as in India. This
probably explains why Jones (1995) found that, in many projects, there is no direct relation between
water charges and the service quality. Shifting irrigation project management to a financially
autonomous organization—it does not matter whether it is a government agency, a local water user
organization, or a private entity—will create a financial incentive for improving irrigation services.
Better services will give farmers an incentive to pay their fees as well as an increased ability to pay
because better service usually means higher farm incomes. Financial autonomy can be an important
key to improved irrigation water management by providing a positive feedback system through a
direct financial link between farmers and the water suppliers.
Financial autonomy ensures that revenue from water charges will revert to the project. Service
providers no longer receive subsidies from the central government, which means they have to collect
water fees from users to recover their costs. In such cases, they are likely to create incentives to
achieve high fee collection rates. Some suppliers strictly enforce penalties against payment defaulters
(see Table 3). In Bayi Irrigation District, China, payment defaulters’ irrigation water is cut off until
they pay their debts (Johnson et al. 1996). In Shangdong, China, the use of integrated circuit (IC)
machines insures that farmers cannot obtain irrigation water without paying. Farmers must purchase a
prepaid IC card to operate the IC machine that measures and controls the water release (Wang and Lu
1999). In this case, although financial autonomy is not mentioned, using IC machines is an innovative
way to collect charges, which gives farmers full control over water use and also effectively enforces
payment collection. This system reduced water use per hectare and achieved 100 percent collection
rates at the same time (see Box 4).
Box 4. China: Integrated Circuit Card Automated Irrigation Charge Collection System In Groundwater
Shangdong is one of the biggest agricultural provinces in north China. Irrigation water accounts for between 70 and 80
percent of total water use, but water is scarce. To improve water use, a card automated system was adopted, in which
irrigators buy prepaid IC cards. The card must be inserted into an automated server before water is released, and it stops
when the card is removed. After each irrigation, the farmer receives a receipt, stating the amount of water used, the price
paid per unit of water and the total deducted from the IC card. All servers are connected by the internet, so they are easy
to control and monitor while the administrative costs are greatly reduced. The costs of each irrigation server is 1,000
Yuan (about US$120)—about equal in value to the water saved annually under this new system. With more than 200,000
IC servers, province-wide, the province saves about 5 billion m
of water annually.
This method makes a 100 percent collection rate possible. If the pricing structure is designed
appropriately, full cost recovery will be achieved (assuming no stealing).
It greatly reduced the personnel costs of administration. People no longer have to collect fees or open
and close gates, and the end user is charged directly, reducing transactions among farmers and
intermediate bureaucrats.
The amount of water used is accurately recorded, and the charges are transparent. Therefore, it greatly
reduces arguments over possible measurement errors.
Farmers have full control over when, how, and how much water they use.
The water charge is on a volumetric basis, which encourages reduced water use.
Source: Wang and Lu 1999.
Incentives both to pay and to collect the fees help increase cost recovery. In Haryana, India, land can
be taken away from people who do not pay their water fees (Cornish and Perry 2003). Suppliers can
also create awards or penalties to encourage their staff to achieve high collection levels. In Awati,
China, staff members’ salaries are completely dependent on the water charges they collect. Since they
do not receive any government funding, they have to pay staff from revenues collected from farmers.
The collection rate reached 98 percent after an institutional reform that established the financially
autonomous management entity (Awati County Government 2002). In Bayi Irrigation District, China,
the staff members receive rewards for turning in the fees by a deadline and are fined for late payment
(Johnson et al. 1996).
User participation throughout the entire irrigation management process through local WUAs appears
to be another important factor in high collection rates. Farmers are more likely to pay if they are
involved in the decisionmaking process, and the earlier the involvement, the better. In fact, they are
more likely to be willing to pay for system improvements they help design. Coward (1980) cites the
Laur project in the Philippines where the WUA had a chance to scrutinize the irrigation agency’s
rehabilitation expenditures on the project. He found the irrigation agency gained in terms of improved
design as well as local commitment to the project.
The irrigation management transfer in Indonesia, started in 1987, also illustrates the benefits of
involving farmers in planning, especially in the preparation stage of renovation or new project
construction. Joint walk-throughs with farmers were found to be the single most effective technique
for communication and cooperation. It allowed farmers to suggest their top priorities and concerns for
improving O&M and has generated more farmer interest and contributed to better design of the
projects (Bruns and Helmi 1996). In addition, it is important that farmers are involved in cost-sharing
decisions and in decisions concerning what costs are to be recovered. In the Indonesia example,
cost-sharing appears to have provided farmers with a strong incentive to insist on higher quality
construction that better serves their needs. They began treating the project as their own. Almost every
successful cases in Table 3 involves some type of local user participation in water management,
suggesting that it is likely to be a necessary reform to improve cost recovery. Mexico is another recent
case of major improvements in water fee collections after management transfer to water users.
After experiencing serious problems with water delivery and fee collection, Mexico in 1990 began a
program to set up and turn over to WUAs management and tradable water rights. By the end of 1997,
400 WUAs were operational, and each controlled an average irrigated area of 7,600 ha. Surveys
conducted in 6 percent of the districts showed that water use per hectare had been reduced and
maintenance improved. Water charges went up in most districts due to the financial self-sufficiency
target, increasing more than 500 percent in some cases. Government subsidies, up to 1996, represented
only 15 percent of O&M costs in the transferred districts. When collected, the depreciation cost could
not be efficiently used due to restrictive financial regulations that prohibit the use of accumulated
funds. In addition, inflation and devaluation of the currency discourage holding such funds. Yet many
WUAs have made significant investments to repair or modernize the infrastructure using bank loans.
The irrigation fees serve as a guarantee to the banks. More than 90 percent of farmers paid their
assessed charges, mainly because they have to pay the irrigation charges before receiving WUA
service (Palacios 1999).
The success of WUAs is enhanced by the skills of its hired technical staff. In many districts in Mexico,
WUAs assist their members in commercializing their operations, obtaining inputs and renting
machinery. Eight Limited-Responsibility Companies, which are federations of WUAs, were operating
and providing services to the WUAs by end of 1996. These companies are expanding their services
beyond maintenance and management of the major infrastructure (Palacios 1999). One of the major
reasons for the positive Mexican experience is the commitment at the highest level of government. The
reforms have focused on the large schemes and farms; transfer in areas with small farms was started
much later and appears more problematic (Simas 2002).
A survey of two minor canals in Mula and Bhima, Maharashtra, India summarized the general benefits
of WUAs. By comparing four districts, two with WUAs already in place and two without WUAs, Naik
and Kalro (2000) found the majority of farmers in the districts who have experience with WUAs select
WUAs as their first choice for supplying water. This is a significantly higher approval rate than in the
two controlled districts without WUAs. Furthermore, in systems with WUAs, 75 percent of the
farmers are willing to pay 25 percent higher water charges because of the better service they have
received. The major reasons for choosing WUAs are: assurance of water delivery and supply, fewer
disputes among farmers, better maintenance, and no corruption.
Table 3. Cases Where Different Combinations Of Factors Increased Fee Collection Rate
Incentives to pay
Financial autonomy
Additional incentives to
Penalty for
payment default
Penalty/reward to
encourage better
User participation
Collection rate
Awati, China
Staff salaries are directly
tied to fee collection rate.
n.a. n.a.
Yes, through
n.a. 98 percent
Bayi ID
(Johnson et al.
Yes, managed by
VIMG with revenue
from water charges,
labor contributions
and sideline business
Staff members receive their
salaries and pensions from
the fees collected.
Payment defaulters
will not get water
until they pay.
Staff members who
do not deliver
water on time are
Yes, all lower level
canals are managed
by VIMG.
Two staff
members collect
fees together to
check on each
other; farmers
get receipts
stating amount
of water used
and price.
Almost 100
Nanyao ID
(Johnson et al.
Yes, managed by
The VIMG get rewards for
turning in collections early
and pay fines for late
n.a. n.a. Yes, all lower level
canals are managed
by VIMG.
n.a. 95 percent
China (Wang
and Lu 1999)
n.a. n.a.
Cannot get water if
they do not pay.
Farmers are willing
to pay because they
have full control
over water delivery
time and amount.
Yes, farmers get
a receipt each
time they use
water, stating the
amount of water
used and the
100 percent
Incentives to pay
Cases Financial autonomy
Additional incentives to
Penalty for
payment default
Penalty/reward to
encourage better
User participation
transparency Collection rate
Yangtze Basin,
China (Lin
Yes, the WSC and the
WUAs are
n.a. n.a.
Yes, WUAs
improved system
maintenance and
efficiency. They
deliver water on
time and in the
right amount.
Yes, farmers are
involved in the
process, and farmer
representatives are
in WSC.
Should improve
transparency by
separating water
charges from
other taxes.
Goal is
but collection
rate not
India (Cornish
and Perry
Yes, systems are
owned by farmer
Farmers own system and
collect charges to cover
Farmers failing to
pay cannot get
water in next
Farmer ownership
Yes, owned and
managed by 3 to 10
farmers in groups
Almost 100
Haryana, India
(Cornish and
Perry 2003)
Partly, charges are set
to cover O&M costs
Land can be taken
away from
payment defaulter.
Yes, each canal is
managed by farmer
n.a. 85–95 percent
Mexico (Zeri
and Easter
Yes, partly achieved
the goal of financial
Farmers have to
pay in advance to
receive water.
Services improved
and investments
made to modernize
Yes, managed by
local WUAs
n.a. 90 percent
Incentives to pay
Cases Financial autonomy
Additional incentives to
Penalty for
payment default
Penalty/reward to
encourage better
User participation
transparency Collection rate
Alto Rio
Lerma ID
(Kloezen et al.
Yes, O&M cost
recovery increased
from 50 percent in
the years preceding
transfer to120 percent
after transfer.
n.a. n.a.
Yes, a better match
expenditure and
farmers’ water
Yes, farmers are
involved in
through WUAs.
staff made
financial system
Almost 100
Note: n.a. = not available; WUA = water user association; VIMG = village irrigation management group; WSC = water supply corporation; O&M = operations and maintenance
Source: listed in table by case
System transparency is another key factor that has had a significant impact on farmers’
willingness to pay their water charges. System transparency means that farmers can see how
much water they received, how their payments are used, and how water charges are determined.
The IC machines in Shangdong, China (Box 4) illustrate good system transparency in terms of
water delivery and payments. Farmers interviewed said they were satisfied because they received
an electronic printout indicating how much water was released, the water price per unit, and the
total amount they paid each time they use their IC card to release water. The case in Sindh
Pakistan is a counter example. Farmers are not willing to pay because their financial system is
not transparent, and they do not see that the charges paid are used in their system due to the
corruption of irrigation officials. The farmers said that they were willing to pay for the services,
but not for “someone’s wife’s jewelry” (Cornish and Perry 2003).
In summary, Table 3 illustrates where financial autonomy and user participation combined with
transparency have been key factors in achieving high fee collection rates. A major task for
management reforms is to create incentives so that farmers have an increased willingness to pay
their water charges. Although in some groundwater irrigation systems, water charges required to
fully cover O&M may be too high relative to net farm income. In most cases, water charges are
only a small share of farmers’ net income, as shown in Table 2. Thus, low collection rates appear
to be caused mainly by a lack of willingness to pay rather than by inability to pay. Examples
from both China and Mexico illustrate what is possible if reforms are successful. These practices,
combined with incentives for service providers and farmers, are critical for high cost-recovery
rates. When the salaries of irrigation personnel depend on collecting water charges or
prepayment of water fees is required, or both, collection rates are much higher.
To encourage farmers to use less irrigation water per hectare, water charges have to be related to
the amount of water that farmers receive. Thus, volumetric water pricing should be considered
when reducing water use per hectare is the major concern. In cases of high volumetric
measurement costs, area-crop or area-technology based pricing methods can be considered as a
second best approach if they can be designed to influence water use, as discussed above.
Table 4 provides a summary of irrigation systems and factors that help reduce water use per
hectare. There are two general approaches to reducing water demand through pricing. One is to
set the per unit price high enough so that farmers use less water on existing crops, which is
essentially a movement along the negatively sloping demand curve. The second approach is to
shift the entire demand curve by inducing farmers to change crops or irrigation technology, or
both. A number of studies of individual crops suggest that irrigation water demand is quite
In Tunisia, the price elasticity of water demand was estimated to range from –0.03 in
the Northeast and –0.007 in the Center-West to –0.27 in the Northwest and –0.34 in the South.
The two former areas that have very inelastic water demands produce high-value crops under
controlled water conditions (fruit trees, vegetables, plastic-covered agriculture irrigated with
modern technologies) (Hamdane 2002). In such cases, water prices have to increase substantially
before they will significantly reduce water demand. In the process, farmers’ income will be
adversely affected. In parts of Spain, some estimates suggest that farmers’ incomes would need
to fall by 25 to 40 percent before an increase in the price of water would lead to significantly
lower water consumption (Berbel and Gomez-Limon 2000). In a case in Iran, water prices would
have to be raised from $4/1,000m
to $20/1,000m
to significantly reduce demand (Perry 2001).
Such a large increase in the price of water may not be politically acceptable. In addition, farmers
must have alternative choices to be able to reduce water use per hectare. Therefore, an increase
in per unit water price may not be an effective way to reduce demand if alternative crops and
technologies are not readily available and water price elasticities of demand are low. Yet,
Schoengold et al. (2004) found own-price elasticity of agricultural water demand ranging from
–0.275 to –0.415 in California’s San Joaquin Valley. They found the indirect effects to account
for only 17 percent of the change in water use. In other words, just reducing water use on the
existing crops was more important than changing to water-conserving crops or to improved
technology. Their study suggests that movements along the demand curve result in significant
water savings. Thus, whether a water price increase will significantly reduce demand has to be
determined and then, if it does, whether it is due to a movement along the demand curve or a
shift in the demand curve. In cases of very inelastic demand, policies and practices that shift the
demand curve to reduce water use should be used, as discussed below.
Supporting institutions
To shift the demand, alternative choices of crops or technology have to be available. A shift to a
less water-consuming crop or to a water-saving technology can move farmers to a significantly
lower level of water use. If there is a wide variety of crops to choose from, policymakers can use
either area-crop–based pricing or increase the per unit volumetric price to induce a shift to crops
that use less water (Box 3). The same strategy can be applied to an irrigation technology shift.
The pricing mechanism can be either volumetric or area-technology based. The price increase
will be even more effective if combined with other policy interventions such as providing
positive supports or taking back subsidies that encourage lavish water use. Low-interest loans for
new equipment and technical assistance will help encourage farmers to adopt appropriate
water-saving technology. In Gujarat, India, electricity used by tubewells is charged at a fixed rate
per month and is heavily subsidized (Cornish and Perry 2003). Therefore, electricity charges do
not include any charge for the marginal cost of pumping groundwater. In this case, the
government should eliminate the electricity subsidies, which have been encouraging overuse of
groundwater, and charge for electricity based on the amount of electricity used. The resulting
increase in pumping costs would encourage farmers to use less groundwater per hectare and save
more for future use. However, under certain conditions, with high enough electricity prices,
farmers may move from electricity-operated pumps to diesel-operated pumps (Dinar 1994).
Besides using pricing tools, there are several other means to reduce water demand. One is to use
a water quota. A quota system is generally used to define the quantity of water that can be used in
a given time period, by whom, and for what purpose (Morris et al. 1997). When water users are
not responsive to water price changes, a quota can be effective in reducing water consumption by
creating a high shadow price. The implementation costs of quota systems can be high because
the quantity of water that goes to each farm must be controlled.
There are different ways of implementing a quota system. First would be a fixed quota system
for groundwater pumping with a specified annual rate of extraction in proportion to the land area
of each water user. A second approach would be a fixed allocation of water shares to different
canals and water users sharing water from the same canal. The fixed shares or quotas could also
be allocated to WUAs (Tiwari and Dinar 2003). For example, in Maharastra, India, the WUA
receives 0.77, 0.86, and 0.62 million cubic meters of water during winter, dry, and summer
seasons, respectively. They can also draw on any unused water quota from the previous season in
the current season (Naik and Karlo 1998).
Service Contracts
Another way of reducing water use per hectare is to provide assurance that water will be
delivered on time and in the amount demanded. If this is done, farmers will not have an incentive
to store water on their field by overirrigating. Since system reform in Katepurna, India, farmers
no longer flood their fields in the dry season and often do not irrigate in the monsoon season
because irrigation scheduling is planned ahead according to water requirements and soil type.
Farmers no longer have to irrigate in the monsoon season just so that they will have adequate soil
moisture for the dry season crop. Farmers now have an adequate and timely water supply,
resulting in reduced water use per hectare. Not only are they saving 7.7 million m
annually, but
they also expanded the irrigated area from 2,027 to 3,646 ha, an 80 percent expansion. This case
shows a real increase in productivity (Belsare 2001). In Shangdong, China, the implementation
of IC automatic machines gave farmers full control over water use (Box 4). They were able to
obtain the right amount of water when they wanted it. The end result is a 5 billion m
saving of
irrigation water in the province annually (Wang and Lu 1999).
A third effective mechanism for reducing water use can be public education campaigns to make
farmers aware of water scarcity and convince them it should be treated like an economic
commodity. This is especially important in places where people traditionally view water as a free
good and a basic right. In many projects, public education programs, combined with price
increases, have been effective. In Katepurna, India, the formation of a WUA and the need for
efficient water utilization were promoted through newspapers, radio, exhibitions, pamphlets, and
posters. Slogans on participatory irrigation management and efficient water use were written on
compound walls, canal structures, offices, and public buildings. To motivate irrigators, cultural
groups were formed from department staff members and cultural programs (e.g., songs, drama)
were arranged at the village level (Belsare 2001). This helped motivate irrigators by improving
the community’s understanding of the value and importance of irrigation water.
Incentives can also be used to induce service providers to reduce conveyance loss. In some
irrigation projects, conveyance loss is more than 40 percent of the total amount of water
delivered. The most effective incentive is financial autonomy. If the service providers are
completely responsible for the project and fee collection, they will try to reduce water losses so
that they have more water to sell, as happened in the Yangtze Basin, China.
Table 4 provides examples of how a combination of different incentives can reduce water
demand even when the water demand for one crop, with a given technology, is inelastic
(unresponsive to price changes). When the major objective is to reduce water use, a combination
of incentives should be used, not just higher water prices. Even if water cannot be metered
effectively, other actions can be taken to help reduce overuse of irrigation water, including
crop-based water fees.
Table 4. Cases Where Different Combinations Of Factors Helped Reduce Water Use Per Unit Of Output
per unit
Switch to
measurement Pricing structure
Alternative crop
Assurance of water
Reduced water
50 percent
Yes, from area
Increasing block, with
the second block price
twice the first
n.a. n.a. n.a.
Yes, public
education to
promote water use
Reduced by
(Wang and
Lu 1999)
n.a. Yes. Constant rate n.a. n.a.
Yes, IC machine installed.
Farmers have full control
over irrigation timing and
n.a. n.a.
Reduced by 5
billion m
/ yr
China (Lin
Yes, from area
Constant at 0.032
Yu an /m
n.a. n.a.
Yes, WUA operated
system delivers water on
time and in the right
amount, which reduce
incentive to steal.
Yes, local water
bureaus constantly
hold training
programs for
Each WUA on
average saves
about 1.18
million m
Yes, from area
Constant rate Yes. Yes
Water delivery schedule is
developed earlier based on
soils, crops and farmer
Yes, public
campaign through
various local
training for
Save 7.71
million m
Yes, more
from 1991
to 2000
Always has
Constant at $65.8 /1,000
loans for
Yes, encourage
user participation
and raise
awareness through
Water users are
more efficient,
and future
supplies are
more secured
per unit
Switch to
measurement Pricing structure
Alternative crop
Assurance of water
Reduced water
Mula area,
Spain (Del
n.a. n.a. n.a. n.a.
Yes, system
financed by
public and
which greatly
reduced water
Yes, farmers have full
control over their water
share. An innovative water
teller is used so farmers
can program their water
training is
provided to
Water loss
reduced from
1.2 Mm
in 1987
to 0.14 Mm
Source: listed in table by case
There is no one easy means to improve cost recovery. However, many countries have greatly
improved cost recovery through basic irrigation reforms. The reforms varied with the irrigation
system type, management structure, government policies, and institutional arrangements (see
Table 5).
Giving farmers more authority and responsibility over water management, usually through
WUAs, is a part of most reforms. In some cases, reform will require other investments or
improvements in water management. In Sri Lanka, for example, besides creating WUAs,
infrastructure investments were also needed to improve system productivity.
A transparent process, where farmers help decide what components should be
included in the costs to be recovered from them through water charges, is an
important stepping stone toward increasing their authority. As part of this process,
farmers need to be consulted early about the design and level of service they want,
as well as the extent and type of any improvements in the system infrastructure.
This type of consultation was done in the Laur Project in the Philippines (Coward
1980). Allowing farmers to participate in decisionmaking improves their
willingness to pay water charges.
To obtain high cost-recovery rates, farmers should not only agree on the costs to be
recovered but also see that the fees collected are used to maintain and improve
“their” system. Having the fees collected go back into the general revenue fund of
the state or federal government, provides farmers with a strong incentive not to pay
fees. One good approach is to have the water supply entity or the WUA collect and
keep most of the fees for use in “their” system. This is one of the big benefits of
having a financially autonomous water supply entity, as in a number of the
systems in China and Mexico (Table 3). In addition, a general government effort to
liberalize and decentralize the economy seems to facilitate such changes in the
water sector (see Table 5).
As part of widening farmers’ responsibilities and authority over water
management, the government should provide them with training and technical
assistance, as was done in a number of the successful cases discussed above (e.g.,
the Yangtze Basin Water Project in China). Again, such reforms seem to occur after
a country has carried out a general economic liberalization (see Table 5).
More should be done to reduce and prevent corruption in water distribution. In
many cases, the “rents” extracted from farmers by irrigation officials are so large
(sometimes 10 to 20 times their normal salary) that raising or introducing official
water changes (in addition to the “informal” charges) can be very difficult.
Establishing an active WUA is definitely an effective way of involving farmers in
the decisionmaking process and improving service quality. However, extra
attention must be paid to developing checks and control mechanisms to prevent
farmers from developing the same bad habits as the former irrigation officials
. One
good option would be to establish and finance a state-level oversight board or an
auditing agency to review WUA operations.
Table 5. Country Conditions And Irrigation Practices
Management transfer
and Policies
Good Practices
structure that
water savings
Assurance of
water delivery
Water scarcity
and drought
Awati, China;
India; Tunisia
n.a. Shangdong,
Tunisia n.a. Gujarat, India Gujarat, India n.a.
Poor condition
of infrastructure
n.a. Katepuna, India;
Mula, Spain
Mula, Spain
Mula, Spain
n.a. Mexico Katepurna,
n.a. Yangtze,
n.a. n.a. Awati, Bayi,
Decentralization Shangdong,
n.a. Shangdong,
n.a. Awati, China;
Gujarat, Indai;
Mexico; Sri
Sri Lanka;
Alto Rio
conditions and
Serious financial
n.a. n.a. n.a. n.a. n.a. Bayi, Nanyao,
Awati, China;
Awati, China;
Bayi, China
Definition of
water rights
n.a. n.a. Haryana, India n.a. Siurana-Riu
n.a. Haryana,
Effective local
system for
enforcing water
use rules
n.a. n.a. Bayi, Nanyao,
Haryana, India
n.a. Siurana-Riu
n.a. Alto Rio
Lerma ID,
Rights to Awati, China n.a. Bayi, Nanyao, Bayi, n.a. n.a. n.a. n.a.
establish WUAs China; Nanyao,
Note: n.a. Not Available
Source: Authors
Another effective tool to improve cost-recovery and pricing is to make the irrigation water supply
entity (WSE) financially autonomous, similar to the supply corporation created at the Tieshan
Irrigation District in the Yangtze Basin, China (Box 1). Making the WSE financially autonomous
changes the incentives for cost recovery and pricing.
If the WSE’s are financially autonomous, they have a financial stake in using incentives
and penalties to encourage farmers to pay their water charges. Incentives could include:
providing high-quality and timely water service. Penalties could include stopping water
delivery to defaulters, charging a higher rate for late payment, making farmers pay water
charges before receiving any water,
or all these measures.
These WSEs also have a financial stake in providing their personnel with a positive
incentive to deliver water on time, and in the right amount, as well as a penalty if they do
not. For example, for failing to deliver water at the scheduled time and for the right
duration, WSE personnel would be fined as is done in the Bayi Irrigation District, China.
Alternatively, good performers would receive a bonus.
To increase its effectiveness, the water supply entity needs to consult directly with
farmers when they are developing the water delivery schedule for the next irrigation
season, as done in Katepura project in India. After the schedule is developed, it should be
widely advertised along with a statement regarding the water charge farmers are
expected to pay. In addition, any changes in the schedule should be quickly conveyed to
every farmer.
The WSE will also have a strong incentive to invest in improved infrastructure to
improve their control over water use. The improved water control will allow them to
provide better services as well as better measures of water delivered. This will, in turn,
make it easier to monitor and base fees on the quantity of water delivered. In several of
the cases reviewed, improved water control saved water and enabled the WSE to increase
revenues by selling the water they saved, as has occurred in the Yangtze project in China.
The fee structures have to be equitable, administratively simple, and easily understood by users and
those administrating the fee collection.
Part of this involves identifying the full range of services and benefits produced by the
project and allocating project costs among all beneficiaries (Box 2).
In addition, information on the costs of services and benefits derived from the project and
on the way project costs are allocated among beneficiaries should be provided to all
For a new project or any major improvement in infrastructure, users’ ability and
willingness to pay should be assessed. As shown in Table 2, if the project is economically
feasible and government agricultural price policies do not disadvantage farmers,
willingness to pay is likely to range from about 5 percent to 30 percent of net farm
income, depending on service quality.
As water scarcity increases, more irrigation projects will have to take seriously the water conservation
objective and begin using water pricing and other mechanisms to reduce water use per hectare.
One good way to measure water use is output per unit of evapotranspiration (ET). Water
removed from a basin through ET is water lost for reuse, so it is really ET that has to be
When water metering is not possible, area-crop and area-technology based water charges
should be designed to strengthen farmers’ incentive to shift to crops that need less water,
or to shift to water-saving technologies, or both (box 3). This works only if alternative
crops and technologies are available and can be adopted without a significant drop in net
farm income.
Where feasible, water markets should be encouraged as a means of improving water
allocation as well as water conservation. One clear advantage of water markets is the
flexibility they provide for moving water to higher valued uses while, in most cases,
leaving both buyers and sellers better off. A number of countries, such as Chile, the U.S.
West, Spain, and Australia have made effective use of water markets.
Public awareness, education, and training programs should be used in water-scarce
regions to make farmers fully aware of the economic value of water and the need to use it
judiciously. This has been done in a number of projects including the ones in Awati
(China), Katepara (India), Yangtze (China), and Tunisia. In other words, users need to
understand the importance of conserving water. Farmers will also need training and
technical assistance to switch to better irrigation cropping practices and technologies.
Special training programs will be needed before the irrigation water is made available
where irrigation is being introduced for the first time.
The World Bank Personnel Consulted
Guy J. Alaets
Musa S. C. Asad
Adel F. Bichana
Greg Browder
Michael Carroll
Rita Cestti
Manuel Contijoch
IJsbrand H. de Jong
Salah Dhargouth
Gerald Diemer
Ariel Dinar
Usaid I. El-Hambali
Nihal Fernando
Vijay Jagannathan
Qun Li
Doug Olson
George T. K. Pitman
Richard Rideinger
Joop Stoutjesdijk
Mona Sur
Satoru Ueda
Dina Umali-Deininger
Types of Questions Asked
Are project costs charged to farmers inflated by irrigation officials?
How important do countries consider cost recovery and water conservation?
How are water charges used once they are collected?
How are fee collections made and by whom?
Do irrigation project managers have incentives to give good service?
Have financially autonomous organizations been used to manage irrigation projects and
if so, how effective have they been?
Has project turnover to farmers been effective? Did it improve service delivery?
Is farmers’ ability to pay an important issue for cost recovery?
How much of project cost should farmers pay? Why?
What incentives are important in obtaining improved cost recovery?
Should farmers invest their money in the irrigation project as a prerequisite for receiving
irrigation services?
What incentives are there for farmers to pay water fees?
What irrigation project reports should I review?
What incentives do farmers have to form WUA such as keeping a share of fees collected
in their project, or helping make water allocation decisions?
How do WUAs support themselves financially?
What enforcement problems exist with the collection of water charges?
Table A2.1: The Case Studies
Country Project/area covered Reference
Batateria Spring of Ceara
Ceará, Northern Brazil, 2004
Jaiba Project, 1995
Kemper et al. 1999
Lemos and Oliveira 2004
Azevedo 1997
BULGARIA Bulgaria, 2001 öKo Inc. 2001
Awati County, Xinjiang 2002
North Plain, the Nanyao and Bayi Irrigation
Shandong 2001 Zhanghe Irrigation District,
China 2003
Awati County Government 2002
Johnson et al. 1996
Wang and Lu 1999
Lin 2003
EGYPT Egypt,1995 Perry 1995
Gujarat 2003
Haryana 2003
India Water Resources Management Sector
Katepurna 2001
Mula scheme and Bhima scheme, Maharashtra,
Cornish and Perry 2003
Cornish and Perry 2003
World Bank 1998
Belsare 2001
Naik and Kalro 2000
INDONESIA Preliminary Impact Assessment, Irrigation
Reform, Phase I, 1999–2000
Indonesia Irrigation Subsector II
Indonesia: Participatory Management
Alaerts 2003
Bruns and Helmi 1996
World Bank 1996
IRAN Zayandeh Rud Basin, Esfahan Province, 2001 Perry 2001
PHILIPPINES Laur Project Coward 1980
MACEDONIA Macedonia Cornish and Perry 2003
MEXICO Development of Water User Associations
Alto Rio Lerma Irrigation District
Zekri and Easter 2003
Kloezen Garces-Restrepo and Johnson 1997
Morocco Water Sector Review
Tadla scheme
Haouz scheme
World Bank 1995
Cornish and Perry 2003
Cornish and Perry 2003
Sunsari Morang Irrigation Project II
World Bank 1996
Cornish and Perry 2003
NIGER Niger, 2000 Abernethy 2000
PAKISTAN Pakistan, 2001
Sindh Province
Ahmad 2002
Cornish and Perry 2003
Mula Area Murcia
Del Amor Garcia 2000
Tarrech, Marino, and Zwicker 1999
SRI LANKA Sri Lanka Samad and Vermillion 1998
UNITED STATES San Joaquin Valley, California Schoengold et al. 2004
Batateria Spring of Ceara (Kemper et al. 1999)
Batateria Spring is in the south of Ceara (Northeast Brazil). This region is semi-arid but generally has higher
precipitation than the interior of the state. It is also the only region in Ceara with abundant groundwater
resources. Twelve large springs in the area produce 40.5 million m
of water per year (Mont’ Alverne et al.
1994). The Batateria Spring, the most important one, serves a relatively homogenous sugarcane-growing area,
where the farmers, who set up their own water allocation system in 1854.
The farmers developed a system of water rights using an old Portuguese measure called telha. One telha
corresponds to a volume of 64.8 m
per hour. To start, the farmers divided 22 telhas among themselves and
reserved one telha to maintain a minimum flow of water in the river. Today, however, the “extra telha” has
disappeared, and the entire flow is allocated for agricultural use. The farmers have allocated the telhas based on
farm size.
Enforcement has been done mostly by the farmers. In the past, larger local farmers had the political power to
punish water thieves. Today, each farm has a levadeiro [the local judicial system] to monitor the operations.
Outsiders can, but rarely do, bribe the levadeiro to get water. A water market has developed, and water is traded
separately from land. Anybody can buy a number of telhas from a system user without any linkage to the
original ownership right to land. Today, the yield from the spring has diminished considerably; and the water
price has been halved over the past century. The reason is that sugarcane is no long as highly valued a crop as it
once was, so the decreased price for water reflects its lesser valued use. This system has generally performed
satisfactorily, providing water security for water right holders and flexibility in water allocation.
This system of water rights and trading developed in the absence of direct government involvement. Its success
is largely due to the homogeneity of the small local community and farmers’ willingness to work together to
provide a collective service.
Ceará, Northern Brazil, 2004 (Lemos and Oliveira, 2004)
Ceará’s Water Resource Management Company (COGERH), created in 1993, was responsible for implementing
the Jaguaribe/ Banabuiú Participatory Management project. Three main factors led to the creation of COGERH:
(1) political reform, involving decentralization and modernization of local government administration; (2) a
particularly long and costly drought crisis that exacerbated the conflict over water allocation; and (3) financial
support from the World Bank that encouraged state government reform. The most innovative aspect of
COGERH’s approach was to organize user commissions to debate and decide on the use and management of
bulk water in the basin. Since 1994, the user commissions have met several times each year and more frequently
in the months right before the dry season. They use reservoir scenarios as a tool to decide how much water to
discharge from the reservoir and how to distribute it among user groups. They set aside enough water to
guarantee human consumption for at least two years and then built several alternative scenarios simulating
different levels of discharge. After developing the alternatives, they debate which scenario best fits the current
water availability and different climate forecasts for the next year.
This decisionmaking tool has been successful in three ways. First, it increased the transparency of the
decisionmaking process. Second, it allows users to make better informed decisions after developing the
alternative scenarios. Third, it helps build a trust relationship between the service provider (COGERH) and the
water users. In 2000, after a year of low rainfall, COGERH decided to reduce the amount discharged. The
amount discharged declined incrementally over the next seven months, forcing users to cut back on irrigation or
other water using activities. This decision helped avert a more serious water crisis and proved that users are
capable of cutting consumption voluntarily. However, because user commissions do not have an official
mandate, their role may be undermined by their inability to enforce decisions among users or institutionalize
their roles within the water management system. The user commissions were created during the nationwide
political reform and transition toward democracy. Therefore, further studies are needed to see whether user
commissions can survive political changes when more conservative policymakers are in power.
Jaiba Project, 1995 (Azevedo 1997)
The Jaiba project is one of the best managed public irrigation projects in Brazil. The objective is full cost
recovery, including operation and maintenance (O&M) costs and capital costs. Jaiba has a two-part water charge,
K1 and K2. K1 reflects project capital costs calculated on a 50-year repayment period and a subsidized interest
rate. K1, which was $3.69 per hectare per month in 1995, accounts for 26 percent of total costs. K2 is supposed
to cover all O&M costs and based on the volume of water used. K2 is subdivided into two components: one
representing fixed O&M costs and the other the variable costs. K2 was $10.11/m
in 1995. Farmers choosing
not to grow crops one season would still be responsible for the fixed O&M costs. K1 is paid to the sponsoring
federal agency, and K2 is paid directly to the water user districts. After the project started operations, the
collection rate was 66 percent, and the cost-recovery rate was 52 percent. Nationwide, the K1 component is
standard ($3.69 /ha). K2 is varied by whether the project is public or private and whether water is supplied by
gravity or pump.
The pricing structure design conveys the idea of both cost recovery and water revenue allocation.
The water charge for O&M has both a fixed and a variable component.
Bulgaria, 2001 (öKo Inc. 2001)
After the political and economic changes of 1989–90, the Central East European (CEE) countries introduced
fundamental changes in their agricultural policies, particularly modifications of the ownership structure.
In Bulgaria, the Ministry of Agriculture and Forestry (MOAF) is responsible for formulating and executing
national agricultural policy. MOAF holds all shares in the Irrigation Systems Company (ISC), a joint stock
company responsible for management, operation, and maintenance of all state-owned irrigation and drainage
systems. In 1999, 176 irrigation water user associations (IWUAs) were established with assistance from the ISC.
However, only a few irrigation systems have been transferred to IWUAs.
Irrigation water prices depend on the source of irrigation—gravity or pump, and there is no uniform pricing
system nationwide. Each ISC and IWUA uses a different method to calculate prices. The pricing scheme
consists of an abstraction fee and a volumetric charge. The average price was between $0.01and $0.085/m
1996–98. The prices for pumping irrigation water are usually two to three times higher than for gravity-fed
water. Revenue from water charges usually covers only part of the O&M costs and in some cases part of the
capital costs. Subsidies to ISCs make up the difference between prices and costs.
This is an example of adjustment in the period following economic and political reform.
The decline of agriculture production and irrigation activities is due to water price increases and
change in the ownership structure.
Irrigation activities are effective where farmers join forces and form IWUAs.
AWATI COUNTY, XINJIANG (Awati County Government 2002)
The governmental agency originally in charge of irrigation water services was reorganized and is now a
financially autonomous enterprise. Staff salaries are directly related to fee collection rates. Through water user
associations organized in each village, a public education campaign program is operated. Since the reform,
water prices, which are based on volume used, have increased by 50 percent Water quotas have been set for
each farm. For any water exceeding their quota, farmers must pay twice the normal price. The amount of water
used per mu (0.067 ha) decreased by 50 m
after the reform, and the fee collection rate is 98 percent.
Financial autonomy and tying staff salaries to collection rates have given service providers a
strong incentive to collect fees.
Farmers have an incentive to use water more efficiently because of the increasing block rate
pricing scheme and higher per unit prices.
Participatory management by local WUAs who involve farmers in the decisionmaking process
has improved communication between water suppliers and users.
Public education programs are also used to increase water use efficiency.
North Plain, Nanyao and Bayi irrigation districts China (Johnson et al. 1996).
In the Nanyao Irrigation District, the two top-tier canals are managed by the irrigation district, and the third and
lower level canals are managed by village irrigation management groups (VIMG). All VIMG personnel receive
their salaries from water charges. The water fees are calculated according to the estimated volume of water
delivered and then translated into an area-based charge. The charge is lower than the provincial average charge
because farmers in that area are relatively poor and a high water price is considered inequitable.
If the VIMG collects 100 percent of the fees by the end of March, it retains 5 percent of the fees; if it reaches
100 percent by the end of April, it retains 3 percent; and if it collects less than 100 percent by the end of May, it
must pay a fine of 3 percent of the uncollected amount. The fees for the year should be collected by the end of
May, 10 days before the first irrigation. Fee collection rates in 1993 were 90 percent, 95 percent, and 97 percent
for March, April, and May, respectively. Before government reform, the water fee collection rate was 100
percent because the commune paid the charges. At the beginning of the reform in 1984, when individual
farmers paid their own bills, the collection rate dropped to 85 percent and lower until 1991. By 1993, after the
irrigation management reform, the collection rate rose to 95 percent and higher.
All 67 of the Bayi Irrigation District 67 staff members receive their salaries and pension from the district—none
of them are civil servants under the county Water Conservancy Bureau. The irrigation district is an independent
public utility and not part of the government. Most villages have a surcharge on their water fee (2 to 5 percent)
to pay the VIMG staff members. The district has an irrigation management division, a diversified management
division to develop sideline business, and five technical and administrative offices to supervise farmers.
There is both a fixed area fee of 1.5 yuan/mu and a volumetric fee of 7.11 yuan /m.
A higher price may be
charged if the actual delivery costs are higher because water comes from another county with its own fee
structure. The water charges are collected through the VIMGs, 3 to 5 days before the village’s scheduled water
turn. The VIMG broadcasts announcements of pending water delivery 3 to 5 days in advance to remind farmers
to bring in their payment. At least two staff members wait at a designated location for farmers to pay. Each
farmer is given a receipt upon payment. On average, 90 percent of farmers pay before delivery. Farmers who do
not pay still get a one-time reprieve and receive water, but if they remain in arrears, they must pay all back bills
before receiving any more water. The change in the collection rate in the Bayi Irrigation District after reform
was much more striking than the one in the Nanyan ID: it dropped from 100 percent to 5 percent. Now
collections have returned to nearly 100 percent.
All VIMGs are responsible for scheduling, managing, and recording water deliveries, which are arranged to
irrigate one farm at a time along a given canal. The director of the Baiyi ID fines staff members who do not
deliver the right amount of water to branch canals on time and for the right duration. Also, the district provides a
small bonus to VIMGs that use less water than planned. Farmers who damage structures are required to repair
them and pay a fine. Farmers caught illegally opening field gates must pay a water fee twice the normal rate.
The district began to create sideline businesses in 1984. By 1994, it had 11 kinds of sideline businesses,
including designing small irrigation projects, fitting water pipes and taps, repairing farm machinery and
agricultural product stores, and constructing buildings. Profits from these businesses provide the district with 7
percent of its total revenue, and the production activities create significant employment opportunities.
Both districts report that they no longer receive central government funds for O&M. Their revenue comes from
the fixed area fee, the volumetric fee, and an annual labor contribution to system maintenance (between 10 and
20 working days a year from each farmer).
The VIMGs are financially autonomous.
The VIMGs are rewarded if water fees are paid early and are penalized for late payments or
SHANDONG (Wang and Lu 1999)
The integrated circuit (IC) card automatic irrigation collection system (CAIC) operates in Shandong’s pump
irrigated area. Under this system, irrigators must buy prepaid IC cards. The card must be inserted into a server
before water is released. The flow stops when the card is removed from the server. After each operation, farmers
receive an electronically printed receipt, stating the amount of water used, price paid per unit of water, and the
total amount of money deducted from the card. All servers are Internet-connected, so control and monitoring are
easy, which greatly reduced administrative costs. Each irrigation server costs 1,000 yuan (about US$120), about
the same value as the water saved each year. The new system pays for itself in one year of operation.
Shangdong is one of the largest agricultural provinces in north China. Irrigation water accounts for between 70
and 80 percent of the total water use. However, water is extremely scarce, and the water per capita is only one
sixth of the national average. With more than 200,000 IC card controlling irrigation servers province-wide, the
province saves about 5 billion m
of water annually.
The CAIC ensures that water fees are paid on time. This method makes a 100 percent collection
rate possible. If the water charges are designed appropriately, they will achieve full cost recovery,
assuming there is no water stealing.
The system greatly reduced administrative costs, because personnel no longer have to collect fees
or open and close gates.
The amount of water used is accurately recorded, and the charges are transparent. This greatly
reduced disputes over measurement error.
Farmers have full control over how much, when, and how to use water.
Water charges are volumetric, which encourages efficient water use
At the administrative level, water use is much easier to control and monitor through automation,
as long as stealing is not a problem.
Yangtze Basin Water Resource Project, 2003 (Lin 2003)
Before the Self-financing Irrigation and Drainage District (SIDD) program, there were several major problems.
First, funds obtained through the collection of water fees were often used for other purposes. Second,
area-based charges were used, and payments were not based on the volume of water used. Third, during the
peak irrigation seasons, farmers conspired against one another to get better access to irrigation water. Fourth,
some water user groups had been established, but they were not permanent organizations. The group leaders
were appointed by local officials and not elected by farmers. Thus, the lack of user participation and ownership
gave farmers little incentive to maintain the system.
The reform goal was to establish a self-financing and self-managing system consisting of two integrated parts:
the water supply corporation (WSC) delivering water from the headworks and the water user associations
(WUAs) operating at the local level. A WUA executive committee was elected by farmers at a water users’
conference. Villages have helped mobilize community human resources for both the WUA conference and the
executive committee election, which helped expand farmer participation and WUA operations. The WSC
charter also requires seats on WSC board of directors for farmer representatives from the WUAs so that farmers
participate in WSC management and decisionmaking. After the transfer of the local irrigation system, irrigation
project authorities and local water agencies constantly held training programs for farmers, which helped
enhance the WUAs operational capacities.
Establishing WUAs created a number of benefits. First, WUA control of local irrigation saved water and labor
and shortened the irrigation time cycle. Water is now delivered on time and in the right amount. The increased
availability of irrigation water under WUAs reduced the incentive to steal. Second, WUAs improved system
maintenance in both the main canal and the lower distribution network. Farmers are investing labor and funds
because they now view the irrigation projects as their own. Third, irrigation costs were reduced in multiple ways
once WUAs were introduced along with volumetric pricing. Farmers now use water more efficiently, and
delivery is also more efficient. On average, each WUA saves about 1.18 million m
of water annually. Fourth,
productivity is improved. In Zhanghe, after introducing WUAs, the average crop yield increased by 6 percent,
2.5 percent of it due purely to irrigation improvements. Finally, the irrigation improvements also helped reduce
poverty. In most WUA areas with improved irrigation services, even poor farmers can now secure their harvest
regardless of weather.
However, problems still exist. Water charges are too low to cover the full costs. For example, in Hunan province,
water charges were 0.032/m
and the estimated costs were 0.10 to 0.15/m.
In addition, farmers are not
willing to pay more for irrigation since they already have high taxes, fees, and compulsory labor sharing.
Furthermore, the farmers are paying the O&M costs of the WUAs plus the water fee charged by the WSC. Thus,
more effort is still needed to improve system transparency, separating water charges from other taxes.
Self-financing and self-management were the keys to improved water management. Farmers
elect their WUA representatives and are involved in decisionmaking.
Support from government, especially from local township authorities, was critical to success.
The traditional top-down, command approach was replaced by a bottom-up participatory
approach at the local level.
Current water charges still do not cover all costs, but farmers already have a heavy burden from
other fees and taxes. It is important to separate water charges from other fees. This is a critical
step towards fair and transparent pricing and will also make self-financing easier.
The SIDDs have been successful because they were established with project support, favorable
policies, and a source of steady funding.
Hubei Zhanghe Irrigation District (Zhang et al. 2003)
To investigate the effectiveness of water user associations (WUAs) in irrigation water management, a survey
was done in Hubei Zhanghe Irrigation District, the first place in China to experiment with WUAs. This
evaluation report shows that WUAs are generally successful, but some organizational changes are still needed.
Since the establishment of WUAs, all kinds of conflicts over water use have been dramatically reduced. The
survey found that 80 percent of the farmers spend much less time waiting for water and supervising irrigation
because WUAs have specialized personnel who supervise water delivery. Furthermore, 77 percent of the
farmers think that irrigation channel conditions have been greatly improved. Therefore, water delivery is much
faster and more efficient than before. The disadvantaged groups, such as disabled people or families without
male adults, are especially appreciative of the WUAs because delivery schedules are well planned and they no
longer have to fight for water. The fee collection rates are much higher, because WUAs have farmers pay part of
the water fee before delivery and the rest after delivery. Also, water use was reduced by an average of 30
percent after the establishment of WUAs, for two reasons. First, the WUAs made many improvements in the
infrastructure, which reduced water loss. Second, because farmers are well informed about the increased unit
water price, they try to conserve water during irrigation.
Some problems persist, however. First, in terms of user participation, farmers are largely involved in
construction and maintenance of irrigation infrastructures, but not in the water allocation decisions. Also there
are no checks and control mechanisms among the WUAs. The majority of people running in the WUA elections
are village officials instead of ordinary farmers. In most places, village officials are still responsible for
collecting water charges and sometimes village officials divert water fees for other purposes. The survey found
that 80 percent of the farmers think WUAs should have the legal rights to collect the fees. WUA finances are
not transparent to most farmers. Finally, many farmers do not understand the basic concepts of WUAs. They
still think WUAs are an extension of government and do not feel they “own” the organization.
EGYPT, 1995 (Perry 1995)
Several studies in 1995 by the International Irrigation Management Institute (IIMI) measured the impact of
different pricing alternatives on the agricultural sector in terms of irrigation water used and farm income.
Among the findings, in Egypt, crop-based water charges were as effective as volumetric charges.
A fixed rate of $52 per hectare, irrespective of crop or water use, resulted in a 4.5 percent fall in farm
income, but had no effect on the choice of crop or technology.
A crop-water charge per hectare, proportional to the calculated average water consumption of each
specific crop resulted in a 2.4 percent fall in farm income, while demand for irrigation was water
reduced by 3.5 percent and the returns to water increased by 2.7 percent.
A volumetric charge based on the quantity of water delivered resulted in impacts virtually identical to
those in the second case. The key factors explaining the different responses appear to be the availability
of a wide choice of crops to grow. In addition, water charges were only a relatively small percentage of
the farm income.
Water price increases, when related to water use, can have a significant impact on water use efficiency (reduce
the quantity of water demanded), when alternative crops or water-saving cropping practices or irrigation
technologies are available.
GUJARAT, 2003 (Cornish and Perry 2003)
Surface water irrigation is limited in northern Gujarat and groundwater development is extensive. Deep
tubewells are the major means of irrigation. Because groundwater use is uncontrolled, a competitive water
market has developed with no concern for sustainable resource use. The irrigation schemes are privately owned
and developed. Most are owned by a group of 3 to 10 farmers who have developed management systems that
are financially profitable under the current low government electricity rates. The water charges are volumetric
and are set to cover system operation and maintenance (O&M) costs. The owners have made the capital
investments collectively based on the area they propose to irrigate and the estimated profits from increased
agricultural production. The current water charges cover full O&M costs. The issue of payment default does not
arise because each farmer has a capital stake in a well, and farmers who fail to pay are barred from access to
water in the next season.
However, the charges for electricity used by the tubewells are a fixed charge per month and do not cover the full
costs of electricity generation, because electricity rates are subsidized. Consequently, the marginal cost of
electricity is zero, and the water charges do not include anything for the scarcity rent or marginal user cost of
the groundwater.
Ownership provides an incentive to achieve full cost recovery.
Water users have an incentive to pay water charges, because defaulters will not receive irrigation
water in the next season.
There is no incentive to prevent groundwater overdrafting.
HARYANA, 2003 (Cornish and Perry 2003)
Surface irrigation is the predominant water source in the three areas studied: the Bhakra command, the Western
Yamuna command, and the lift irrigation areas in southern Haryana. Available water is divided equally over the
entire command area through a canal system that automatically apportions the water among farmer groups (300
ha or 100 farmers), who share the supply in rigidly fixed turns. Allocation and scheduling of water among
canals is the responsibility of the irrigation department. Once the water reaches the outlets, farmers are fully
responsible for operation and maintenance (O&M). This system has proven to be one of the most productive in
India, because it provides strong peer pressure governance. A farmer who steals a turn from another farmer can
cause instability over the entire canal. The system can also have serious problems if the irrigation department
does not provide water on time and for the planned duration and quantity.
The charges are set to cover the O&M costs but not capital costs. Charges are based on area and crop. Fees are
collected by the state revenue department. Collection rates are high, between 85 and 95 percent, because the
government can take land away from defaulters.
Water charges are about 0.5 percent of net farm income. The irrigation department achieves full O&M cost
recovery by allocating only 33 percent of the overall project costs to irrigation (which uses 92 percent of the
surface water) and by charging a low price (less than one-twentieth of the price to industrial, municipal, and
other users). The price would be 3.2 times higher if the costs were allocated by share of water (92 percent) used
($8/ha instead of $2.5/ha). The actual O&M costs overall are still very low. This is due to highly centralized
management, limited staff requirements, and substantial farmer participation in O&M. Also, since the municipal
and industrial users share more costs, they receive better services and supply priority during shortages.
The system combines effective overall government supervision and participatory management
through WUAs.
Farmers have strong incentives to pay fees, because land can be confiscated for nonpayment.
System design creates peer pressure against water stealing.
Haryana is an example of cost sharing among different types of beneficiaries in the same water
Katepurna, 2001 (Belsare 2001)
The irrigation improvement and reform project in Katepurna in Eastern Maharahstra state took place in 1998.
Before the reform, farmers were reluctant to pay fees because they had no assurance of receiving water at the
right time and in the amount they wanted. Heavy water loss led to drainage problems and difficulties in
transporting products to market. There was no coordination among beneficiaries and project authorities.
The new project included the following improvements and reforms:
The leaky canal structures were repaired.
The irrigation schedule was planned in advance. Scheduling was based on crop water
requirements and soil types. Farmers are now assured of adequate and timely water supplies.
Water charges are volumetric instead of area based. Flow-measuring devices have been installed
at canal head to measure the canal discharge, which has improved efficiency.
Participatory irrigation management (PIM) was promoted by the formation of water user
associations (WUAs). A project-level coordination committee was formed with WUA
representatives to plan, coordinate, and monitor irrigation water use. Farmers were involved in
project decisionmaking and management. Incentives were given to irrigation officers as well as
to WUAs. On-farm training was conducted for farmers and project personnel. Field visits were
organized to share successful WUA experiences. Farmers were trained to adopt water-use
efficient irrigation technologies such as furrow irrigation.
A public awareness campaign was conducted to promote WUAs and improve water use
efficiency through newspapers, radio, exhibitions, pamphlets, posters, and slogans written on
compound walls, canal structures, offices, and public places. To motivate irrigators, cultural
groups were formed by department staff members, and village cultural programs (songs, dramas)
were arranged.
In 1998–99 to 2000–2001, the average area irrigated in the Katepurna project command went from 2,027 ha to
3,646 ha, with annual water savings of around 7.71 million m.
The Katepurna experiment is being tried, with
similar success, in other projects in the Akola district.
The project provides economic incentives to conserve water because area-based water charges
were changed to volumetric charges.
Public education campaigns were conducted through different local media and on-farm training,
and technical support was provided to promote water use efficiency.
India Water Resources Management Sector Review (World Bank 1998)
Problems and constraints found in the World Bank–Government of India Water Resources Management Sector
Review include:
Poor maintenance, poor design and construction quality, and delayed maintenance.
Ineffective control structures, meaning that canal operators cannot deliver water on time and in
the right volume.
Lack of public sector accountability because there is no incentive linking farmers and service
Minimal involvement of farmers in irrigation management.
Poor cost recovery due to excessively low water charges.
Inadequate funds for operation and maintenance (O&M).
No incentive for water-use efficiency and conservation due to high public subsidies and
area-based water charges.
The World Bank and the Government of India worked together to devise the reform agenda. Plans include:
Water charges will be substantially increased to cover O&M costs. Full cost recovery will be
achieved through annual increases in a time-bound program not to exceed three years.
The direct collection of water charges will be shifted to WUAs. They should collect water charges
and use the revenue to finance their operations. They will be responsible for ensuring high
collection rates.
Volumetric pricing will be introduced where possible.
Cost-effective O&M will be emphasized.
Service improvement and provision will become client-driven.
Efforts will be made to persuade farmers to share in investment costs to help create a sense of
“ownership” for long-term project sustainability.
Transparent billing and cost monitoring will be practiced by the WUAs.
An independent price regulatory agency will be established to supervise and oversee water pricing.
Maharashtra: Mula scheme and Bhima scheme (Naik and Kalro 2000)
After management of two minor canals in Mula and Bhima, Maharashtra, India, was transferred to water user
associations (WUAs), a survey was done to assess the impact on farmers.
In the Mula scheme, 82 percent of the farmers ranked water user associations (WUAs) as their first choice for
supplying water. 74 percent of the farmers in the Bhima scheme indicated the same preference. This implies that
WUAs generally have performed satisfactorily since the management transfer. In contrast, 69 percent and 36
percent of the farmers in the two control groups without the management transfer in Mula and Bhima,
respectively indicate WUAs as their first choice. The differences in preferences were statistically significant. In
the systems with WUAs, the majority of farmers (more than 75 percent) were willing to pay 25 percent higher
water charges for the better services they received. The nontransferred canal in Bhima was the only one with the
majority of farmers (60 percent) indicating government provision as their first choice. The preference for a
private agency as water suppliers in both schemes was small (less than 5 percent). The survey also considered
location in choosing farmers at the head, middle, and tail of the canal. The variations in preference in different
locations were small. The major reasons farmers gave for selecting the WUA as their first choice were:
assurance of water delivery and supply, less dispute among farmers, better maintenance, and lack of corruption.
The survey results indicate very high preferences for WUAs in places where management
transfer has taken place. The survey also indicates a willingness to pay higher prices for better
services. This empirical evidence demonstrates willingness to pay is linked to better service.
The reasons given by farmers for choosing WUAs typically reflect their major concerns about
inadequate irrigation water service. Therefore, these concerns should be the focal point of any
irrigation management improvement project.
Preliminary impact assessment, irrigation reform, Phase I, 1999–2002. (Alaerts 2003)
Water management in Indonesia has been decentralized and transferred from the central government to district
governments and from the district governments to water user associations (WUAs), which are large,
autonomous, legal entities. Decentralization was facilitated by the general administrative and fiscal
decentralization that started in 2000. The local WUAs represent 2,000 to 5,000 farmers, who elect their own
leaders who are responsible for water management and fee collection. Farmers participate in the selection,
design, implementation, and supervision of any physical works. They can even engage in construction activities.
Before the reform, the central government financed almost all water projects. Now WUAs retain revenue from
water charges to run the projects and organize farmers to contribute labor for routine operation and maintenance
(O&M). District governments finance some infrastructure construction. During a transitional period, the
national government still provides a subsidy to the district governments.
Among benefits from the program are:
Overall transparency of government decisions and actions has improved substantially.
WUAs are developing their skills, capacities, and internal governance quality. Also, WUAs are
collecting increasing amounts of cash and labor from members.
Overall productivity has improved due to better and more equitable water allocation. The
irrigation water supply is more reliable and secure for farmers.
Irrigation Subsector II (World Bank 1996)
The World Bank Irrigation Subsector II project in Indonesia strengthened cost recovery by updating areas
registered for the rural land and building tax and by increasing the area covered and amount of irrigation service
fees collected. Major achievements include a general acceptance by many communities in many districts that
service fees should be paid and that WUAs should be responsible for collecting and paying this fee.
In the first season of 1994, a complete record of collection rates in the 35 districts monitored by the irrigation
improvement project was available. Twenty districts had 100 percent collection rates, 4 had a record of more
than 90 percent, and one had 80 percent. The other 10 districts had an average collection rate of 34 percent.
Overall, the average amount collected was 70 percent, with 7 percent exempted and 23 percent unpaid.
Sustainability of schemes turned over to WUAs will depend on farmers’ paying O&M costs and provincial
water resource services’ providing assistance for major or emergency repairs. Provinces are now responsible for
all irrigation O&M. Irrigation service fees were successfully collected where WUAs were responsible for
collecting the fees, users were confident that charges were determined fairly in consultation with farmer
representatives, and the fees were used for maintaining a reliable irrigation water supply.
Participatory Management (Bruns and Helmi 1996)
Irrigation management transfer in Indonesia started in 1987. It was based on policies including: gradual transfer
of irrigation schemes smaller than 500 hectares to water user associations (WUAs), implementation of irrigation
service fees (ISF) in systems larger than 500 hectares, and starting an on-farm water management development
project combining training and subsidies to support improvements.
Turn over in small systems. A 1994 study in West Java and East Nusa Tenggara showed that, on average, yields
increased one quarter ton per hectare per season after the management turnover. Water supplies and deliveries
were also more reliable and equitable. However, about one quarter of sites still had poor returns, indicating that
further improvements were needed in water management. After the turnover, government support was still
important, including, for example, an insurance program providing costs and cost sharing and capital for major
repairs and improvements in case of natural disasters.
Large systems with ISF. WUAs are involved in collecting fees and in identifying priority needs for improving
operation and management (O&M). Originally, service fees were to be used within the irrigation system where
they were collected. However, it appears that, in many districts, use of service fees is not linked to a particular
irrigation system, which makes the fees no different from an irrigation tax, without specific accountability and
incentives. In addition, farmers rarely receive information about how service fees are being used.
On-farm management development project. Although implementation is still restricted mainly to pilot projects,
it provides important lessons about improving the use of participatory management in irrigation systems.
Practical training for farmers is provided through a series of two-year sessions. After the training, improvements
were made in irrigation infrastructure. Each scheme received a government subsidy of $1,000, and farmers were
expected to make up the difference by contributing labor, materials, and cash. This arrangement is financially
transparent and should stimulate local contributions because governmental aid had clear limits on it from the
very beginning. Farmers benefited mainly through improved water availability, which increased both production
and farm income after the project. Lack of continuing support after the project is problematic because trained
officials often returned to their normal jobs.
The most noticeable benefit of participatory management is the ability to involve farmers in planning,
especially when preparing for a renovation or new construction. Participatory management evoked
farmer interest and contributed to better project design. Joint walk-throughs with farmers were the
single most effective technique for communication and cooperation because they gave farmers an
opportunity to explain their top priorities for improving O&M.
Another benefit of participatory management comes from involving farmers in cost sharing. In most
turnover areas, farmers pay fees in cash or contribute labor and materials. Sharing costs give farmers
strong grounds for insisting on high-quality construction and design that serves their needs. Farmers
have an incentive to contribute in the hope that a large contribution may raise their status in the
community or in the fear that failure to contribute might deprive them of needed assistance in the
Farmers now have more equitable and timely water delivery. Previously, whoever had the most political
or economic power received the water. Currently, WUA rules and regulations make for more equitable
distribution of water.
Zayandeh Rud Basin, Esfahan Province 2001 (Perry 2001)
In 2001, the water price for the gravity-fed surface irrigation system in the Zayandeh Rud Basin was
. This price allowed for nearly full recovery of operation and maintenance (O&M) costs. Water
charges account for less than 10 percent of net agricultural revenue.
To significantly reduce demand, however, prices have to be raised to between $20 and $50/1,000m
. Perry
(2001) shows that the current water prices would likely have to be increased twentyfold before farmers would
invest in field technologies to improve water use efficiency. At this high level, water charges will be equivalent
to two-thirds of the gross farm revenue, which would be difficult to implement in practice. Perry (2001)
suggests using water charges as a way of covering O&M costs fully and using rationing as a separate tool to
restrict the amount of water used, instead of using high water prices.
One option Perry (2001) did not consider is use of an increasing block pricing system, with the first block equal
to the quota set under the rationing. Another alternative would be to raise water rates and subsidize water-saving
irrigation technologies, as is being done successfully in Yemen.
Sometimes the two objectives in designing irrigation pricing schemes—full cost recovery and
reducing demand—cannot be achieved with just one price. However, an increasing block price
can achieve both objectives.
To deal with water shortages, rationing may be another effective way of reducing demand in the
case of inelastic demand curves.
MACEDONIA (Cornish and Perry 2003)
Most schemes have reservoir storage but much of the irrigation infrastructure is in very poor condition. Most
irrigation services are provided by one of the 20 Water Management Organizations (WMOs) which are overseen
by a central Public Water Management Enterprise (PWME). Water charges are set to cover full O&M costs and
capital depreciation. Charges are levied on an area basis. The charges include a 10 percent (it used to be 50
percent before 1998) fixed charge, calculated as 10 percent of predicted O&M and capital depreciation costs per
hectare. The remaining 90 percent depends on the type of crop grown. Practices show that the fixed charges are
too low since most of their expenditures on irrigation are fixed. The variable charge fluctuates greatly from
season to season in this semi-humid climate. Under the Tikvesko Pole Kavardaci WMO income exceeds the
expenditure on O&M by $ 41/ha, if expenditures on past debt and capital depreciation is excluded. If bad debt
and capital depreciation are included, expenditure exceeds income by $ 165 /ha.
The collection rates were 42 percent in 2000, since collection enforcement has not been effective. In the past,
defaulters were often brought to court, but as more and more people fail to pay, the court costs became too high
for the WMOs, since it takes 3 to 5 years to process these cases. There are not many alternative means to
sanction the defaulters unless they have a permanent source of income to confiscate, but the average farm
holding is very small. An alternative is to suspend supply. However, this is not very practical since more than
one farmer shares the same outlet, valve, or sprinkler system, which makes it difficult to shut off water to just
one farmer.
Effective mechanisms for fee collection are equally important or even more important than
designing the appropriate pricing policy. Lack of incentives for farmers to pay or penalties for
failure to pay have resulted in serious collection problems.
Alternatives need to be developed to solve the bad debt problem so that the WMOs can become
financially sustainable.
Development of Water User Associations (Zekri and Easter 2003)
In 1990, after Mexico experienced serious problems with water delivery and fee collection, it began a program
of establishing and turning over management and tradable water rights to water user associations (WUAs). By
the end of 1997, 400 WUAs were operational, each controlling an average irrigated area of 7,600 ha. Surveys
conducted in 6 percent of the districts showed that water use efficiency and maintenance had improved. Water
charges went up in most districts due to the financial self-sufficiency target, increasing over 500 percent in some
cases. Government subsidy, up to 1996, represented only 15 percent of the operation and maintenance (O&M)
costs in the transferred districts. If collected, the depreciation cost could not be efficiently used due to restrictive
financial regulations prohibiting the use of accumulated funds. In addition, inflation and currency devaluation
discourage holding such funds. Yet many WUA have made significant investments to repair or modernize their
infrastructure using bank loans. The irrigation fees serve as a guarantee to the banks. More than 90 percent of
farmers pay their assessed charges, primarily because they have to pay in advance for WUA services.
The success of WUAs is enhanced by the skills of a hired technical staff. In many districts in Mexico, WUAs
assist their members in commercializing their outputs, obtaining inputs, and renting machinery. Eight
limited-responsibility companies, which are federations of WUAs, were operating and providing services to
WUAs by the end of 1996. These companies have expanded their services beyond maintenance and
management of major infrastructure (Palacios 1999). The reforms focused on larger schemes and farms. The
transfer in areas with smaller farms was started much later and is more problematic (Simas 2002). Commitment
at the highest level of government is a notable reason for the positive Mexican experience.
Because farmers must pay in advance for water, they have a strong incentive to pay.
WUAs’ technically trained staff provides a wide range of services to members.
ALTO RIO LERMA IRRIGATION DISTRICT (Kloezen Garces-Restrepo and Johnson 1997)
A two-year field research project started in 1995 in the Alto Rio Lerma Irrigation District studied the impact of
irrigation management transfer. The study was based on data from 1982 to 1996, comprising 10 years of
pre-transfer information and four years of post-transfer information.
The study found that:
The transfer had very little impact on actual surface water allocation and distribution.
Active farmer involvement in decisionmaking and control has increased managerial accountability.
Farmers are particularly satisfied with the improved service provided by ditch tenders because farmers
now have more control over the ditch tenders’ work and rent-seeking behavior.
The most positive impact has been the considerable improvement in maintenance, especially at lower
system levels.
Financial self-sufficiency increased from around 50 percent before the transfer to 120 percent in the
post-transfer period. WUAs achieved 100 percent collection rates and also made the financial system
more transparent.
The transfer has not resulted in an increase in farmers’ water costs. WUAs find it difficult to convince
farmers that irrigation fees should be increased to keep up with inflation.
In the Mexican case, the irrigation management transfer did not come alone, but was part of a
wider set of economic reforms.
They created financial autonomy in the irrigation management as one of the key institutional
Morocco Water Sector Review (World Bank 1995)
The objective of water charges in Morocco is to recover 40 percent of the capital costs in addition to operation
and maintenance (O&M) costs. The water charge is volumetric and the rate is high by international standards.
The volume of water billed is typically 80 percent of available water, which suggests that the system loss in
delivery is relatively low.
The public irrigation schemes have reservoir storage, and water is distributed through overhead concrete
channels with sophisticated division structures at control points. Farmers also rely heavily on private wells to
provide additional, flexible water, which helps correct for some of the deficiencies in the public system.
Tadla Scheme (Cornish and Perry 2003)
The collection rates in the Tadla scheme are between 70 and 80 percent on average. As the first and oldest
irrigation project in Morocco, it costs much more to operate and maintain ($127/ha/yr, full costs of $150/ha/yr)
than the Haouz project. Water charges account for 15 percent of net farm income.
Every farmer in Tadla has a “checkbook” to help track water consumption. At each water delivery, the farmer
gives a “check” to the ditch rider, detailing the volume of water consumed, and keeps a copy for himself. Both
the ditch rider and the farmer sign the monthly water delivery document to make sure there is no error in billing.
The total water available for the season is defined in the checkbook, but the delivery schedule varies by
individual farmer demands.
Haouz Scheme (Cornish and Perry 2003)
Because Haouz is a relatively new project, the operation and maintenance (O&M) costs are as low as
$30/ha/year, but the full cost including capital repayments and depreciation is $54/ha/yr, almost twice the O&M
costs. Project-level data indicate that there are substantial subsidies because extensive “traditional” areas do not
pay the charge and many farmers contribute labor for canal cleaning to reduce their payments to 40 percent of
the designated charge. The collection rates range from 60 to 70 percent. For farmers paying full charges, the
water fees are about 7 percent of net farm income.
Bills are prepared by the local Office Régional de Mise en Valeur Agricole (ORMVA), the organization
responsible for agriculture development. If a farmer contests a bill, he can request a review, which is done
jointly by the farmer, the ditch rider, other farmers using the same canal, and the president of the local water
user association.
The “checkbook” approach combines rationing and flexibility. The total water available for the
season is set in the checkbook. The delivery schedule can vary based on farmer demands.
Both schemes are good examples of system transparency because farmers can verify water
delivery and water charges.
EPAL (Cornish and Perry 2003)
Most public irrigation schemes in Nepal are gravity-fed run-of-river diversions, but there are also some pumped
irrigation schemes.
There are three different management types:
Farmer managed schemes account for 70 percent of all irrigation projects and are concentrated in
the hill areas. Farmers are responsible for all management activities. Government does not levy
service fees on these systems.
The public irrigation schemes are managed by the department of irrigation (DOI). Water fees
from these systems are retained by the national treasury. In 1996–97, the average recovery rate of
operation and maintenance (O&M) costs was 1.3 percent. The fee collection rate was generally
lower than 30 percent.
In systems jointly managed by both local water user associations (WUAs) and DOI, collection
rates were 58 percent. WUAs collect the fees, retaining part of the funds in proportion to their
responsibility. The remaining funds are passed on to the national treasury. On average, water fees
represent less than 2 percent of the O&M costs.
In government irrigation systems, water charges are designed to cover O&M costs and in theory, some capital
costs. In practice, they do neither.
Niger, (Abernethy et al. 2000)
The government objective is to transfer responsibilities for irrigation operation and maintenance (O&M) to
water user cooperatives. Pricing is designed to cover all O&M costs and part of the initial capital costs. From
1992 to 1996, the average seasonal fee in three rice-growing systems was $124/ha/season—high by
international standards. The irrigation fees are equivalent to between 12 and 25 percent of gross crop value.
Fee collection rates are high, between 90 and 100 percent, but farmers are often quite late in paying their bills.
The major objective of the water pricing system is to reduce water use, but it has not been successful because
water charges are not related to water use or delivery. Farmers have little incentive to improve water use
efficiency since it would not change what they have to pay. Conveyance losses run between 40 and 60 percent,
and these losses are an extra cost that is divided among members of the user cooperative of perhaps a thousand
members. Therefore, farmers have little incentive to provide labor for improving the system, considering that
they receive only one thousandth of the potential benefit from reduced water loss. Irrigators in two of the three
systems are dissatisfied since their cooperatives do not include them in the decisionmaking process.
Effective incentives are needed to encourage farmers to pay their water fees on time.
An incentive system needs to be put into place that forces both farmers and service providers to bear
some of the costs of their water losses.
Participatory management needs to draw more farmers into the decisionmaking process.
For the water fee to reduce water use effectively, it has to be related to water use (crop-area charge or
volumetric charge)
Pakistan, 2001 (Ahmad 2002)
The large gravity flow irrigation schemes in Pakistan have not been successful in terms of either water use
efficiency or cost recovery. Area and crop-based flat rates have not encouraged efficiency in water use because
neither is related to actual water use. The charge in 2001 per acre-inch of water was Rs. 2.04 for rice, Rs. 3.09
for wheat, and Rs. 2.36 for sugarcane. Rational producers maximize water use, even though water may be
scarce. The variations in water charges by crop imply that irrigation water is subsidized more for certain crops.
Experiments show that using different irrigation methods can improve water use efficiency. However, because
there is no relation between the quantity of water used and water charges, farmers have no incentive to try better
The overall efficiency of the Indus irrigation system has been estimated at only 35 to 40 percent. Fee collections
are inadequate due to ineffective enforcement of charges and poor irrigation services. Thus, revenue collections
financed only 30 to 35 percent of the operation and maintenance (O&M) expenditure of the irrigation
department in 1998–99. Continued inadequate expenditure for O&M will cause further water loss and
inefficiency. The revenue from water charges is commingled in the provincial treasury with other tax revenues,
further reducing the incentives to pay water charges. This means that there is no relation between funds
allocated for project O&M and fees paid by farmers.
There is no link between water charges and services provided since the revenue collected from
water charges does not go to improve the irrigation system. Thus, farmers have little economic
incentive to pay water charges.
In addition, service providers have no incentives to collect fees.
Because the above crop-based water charges were designed to ignore differences in water use by
crop, they do not give farmers an economic incentive to grow water-efficient crops.
SINDH PROVINCE (Cornish and Perry 2003)
Sindh province, located along the lower Indus river, is the second largest province in Pakistan. Wheat, rice, and
cotton are the three dominant crops in that area. The irrigation management system is undergoing reform,
including changes to the pricing system, because the government subsidies can no longer sustain the required
operation and maintenance (O&M) expenses.
The major water source for irrigation is through barrages. Sindh also has serious drainage and groundwater
salinity problems. Located at the tail-end of the Indus irrigation system, the province often does not get its
agreed-on share of water. There are also enormous management problems in the system. The Irrigation and
Power Department is in charge of irrigation water in Sindh, but its performance has not been supervised. It does
not generate its own revenue or collect water charges.
An assessment of ability to pay for the O&M costs was made in 1995. The result suggested that, despite the
relatively low net farm income, farmers can pay for irrigation services, and the current water fee accounts for
only 2 percent of the net farm income. Collection rates are low (less than 30 percent) because farmers are not
willing to pay due to the corruption among irrigation officials, poor service quality, and lack of system
In 1997, a major reform program took place, and a new agency, the Sindh Irrigation and Drainage Authority
(SIDA), was established. Its goal was to achieve financial self-sufficiency, defined as paying full O&M costs in
a 10-year period. O&M responsibilities were transferred to farmer organizations that serve an average area of
3,000 ha.
Water prices, determined by the provincial government, are regarded as a tax to be paid as a single bill along
with land and other taxes. The water tax is area based and differentiated by crop and by irrigation system. Even
assuming 100 percent fee collection, current charges are still too low to cover O&M costs.
Laur Project (Coward 1980)
An important aspect of user participation is full farmer involvement in the decision-making process. For new
projects or rehabilitation, farmers should be consulted before the projects begin. The Laur project of the
National Irrigation Administration (NIA) in the Philippines is a good illustration of active farmer involvement.
NIA assisted in the rehabilitation of a community system, and local farmers actively participated in the whole
process. They assisted in the preconstruction engineering survey, commented on the initial construction designs,
and maintained a careful record of construction expenditures. The water user association was particularly
serious about controlling expenditures because it had to repay a loan to NIA for the work. Farmer supervision
made the financial system transparent and prevented any corruption by irrigation officials. The benefits of this
active user involvement are: (1) design and construction fit specific local needs because the project was done
based on farmers’ suggestions; (2) local commitment to the project was created; (3) because the project was
constructed to serve farmers’ needs, the farmers were more willing to pay for the improved services.
This case provides evidence that user participation is important, especially when it involves
farmers in the decision-making process in advance of project rehabilitation or construction.
It shows that user participation can improve system transparency and increase farmers’
willingness to pay for the services.
Mula Area Murcia (Del Amor Garcia 2000)
As part of an irrigation improvement project, a water teller system was installed. Each farmer receives an
annual share of water in cubic meters, his own water account, and a report of the water remaining after each
irrigation. Farmers can also use the water teller to program the opening or closing of their irrigation gates and
even program the release of fertilizer into their fields. The water teller is located outside community
Automation of the irrigation system in the Mula Area was accompanied by major infrastructure investments,
including water storage and a pressure irrigation network. The traditional system, which is being modernized,
dates back to the Muslim period of the 9th and 10th centuries. Before the modernization, the system was
characterized by high water losses and fragmented land holdings in which 68 percent of the fields were smaller
than one hectare. The systems modernization and automation resulted in an 88 percent reduction in water losses
from the distribution system, a decline in losses from 1.2 million m
in 1987 to 0.14 million m
in 1998.
The automation system has merits similar to the integrated circuit card system in Shangdong,
China. It provides effective enforcement of fee collection and also gives farmers full control over
their water use.
No information was provided concerning water fees or fee collection rates.
Siurana-Riudecanyes (Tarrech et al. 1999)
The Siurana-Riudecanyes irrigation district in Tarragona province delivers approximately 6 million m
of water
each year. The hydraulic infrastructure consists of two dams, an inter-basin channel, two main distribution
channels, and corresponding networks. It supplies water for the farmers in the irrigation district and water for
municipal uses in the city of Reus.
The local residents have a long history of raising funds from both public and private sources. Contributions
from direct beneficiaries (farmers, municipal users from the city of Reus, and other users) were raised by
issuing water rights for the added water capacity in the system. Operation and maintenance (O&M) costs have
always been covered by the water users through their contribution to the water user association (WUA)
. Since
1972, the WUA has operated with enough funds to support further improvements in the infrastructure.
The right to use water is obtained through regional water authorities. Water titles are distributed to association
members based on their contribution to construction and the irrigable land owned. Water titles have always been
traded among the WUA members, including farmers and municipal users. Both long-term and temporary
transactions take place in the water market. In 1982, an official exchange, administrated by the WUA, was
formed, which significantly reduced water price volatility and made the exchange more transparent. A system of
bonuses and incentives was also established for the WUA workers to minimize water loss and reduce O&M
costs. All the workers are hired from private companies, which equip the workers well for their jobs. For
example, they have a radio-telephone system so that a prompt response is guaranteed to any emergency in the
The WUA represents a central force in the system with active user participation, transparency in its functions,
and flexibility to adapt to specific circumstances. The city council of Reus also has had a significant impact on
the water market. It provided a large part of the funding for the original work (50 percent contribution, another
40 percent as a loan to be repaid by the direct beneficiaries) and played an active role in water management. In
addition, direct beneficiaries contributed 10 percent of the funding.
This is a good example of achieving financial autonomy through costs sharing among different water
system beneficiaries, particularly when a new project is to be built.
Water rights, in this case, are closely tied to financial contribution, which helps create a sense of
ownership and achieve cost recovery.
The water user association plays an important role in the system
SRI LANKA (Samad and Vermillion 1998)
From 1988 to 1998, the government of Sri Lanka formally implemented a policy of transferring to farmer
organizations (FOs) full responsibility for operating and maintaining irrigation facilities below the head of
medium-size and major schemes’ distributional canal. The FOs are organized along hydrologic boundaries.
Government expenditure on irrigation began to decline before the irrigation management transfer (IMT).
Expenditures dropped in both non-IMT schemes and in schemes where transfer had occurred. The reforms have
not substantially increased farmers’ irrigation costs. Farmers generally make fewer direct payments (in cash and
kind), but contribute more labor for canal maintenance. The survey done in Hakwatuna Oya and Nachchaduwa
showed that the majority (90 percent) of farmers in both schemes claimed that there was no cash fee on
irrigation before turnover. After transfer of O&M functions to FOs, some of the organizations charged a modest
fee for canal maintenance. The survey results showed that a minority of farmers (23 percent in Hakwatuna Oya
and 16 percent in Nachchaduwa) paid the maintenance fee. In both schemes, farmers’ irrigation cost is primarily
in contributions of family labor for canal maintenance and payments in kind (a half-bushel or 11 kgs of paddy
per acre) to the person employed by the FO to distribute water.
In Nachchaduwa, nearly 60 percent of all farmers interviewed felt that the condition of the canal system was
worse after management transfer. This implies extensive farmer dissatisfaction with the rehabilitation, which
was done without farmer participation. Also, the infrastructure inspections revealed a serious under-investment
in maintenance. Neither IMT nor rehabilitation alone resulted in significant improvements in agricultural
production. However, in schemes where both management transfer and rehabilitation occurred, significant
increases were found in agricultural productivity and economic returns.
Senior Irrigation Engineer N. Fernando from the World Bank, has reported significant improvements in
communication between farmers and irrigation since the management transfer project, including: farmer
participation and decisionmaking in O&M, seasonal crop planning, water allocation and management, and
rehabilitation activities. All major and medium-size irrigation schemes now have project management
committees chaired and managed by farmers
Participatory management with the aim of full cost recovery will not necessarily increase water
charges since farmers have the option of contributing more labor for O&M rather than cash.
Participation should start before project rehabilitation is considered so that farmers can be
involved in the decisionmaking process regarding the rehabilitation.
Sometimes a combination of changes must be made before a project can operate effectively. In
this case, rehabilitation and management transfer needed to occur at the same time.
SAN JOAQUIN VALLEY, CALIFORNIA (Schoengold et al. 2004)
An agricultural water demand model based on an episode of rate reform was estimated by using a unique panel
data set from California’s San Joaquin Valley. There were three major findings: (1) Price changes do reduce the
demand for irrigation water. The estimated own-price elasticity of agricultural water demand was in the range
[–0.275 to –0.415], which is higher than indicated in previous studies. (2) Price changes influence the choice of
irrigation technology and crops. The demand elasticity can be decomposed into direct effects and indirect
effects of demand reduction in response to price increases. The study showed that indirect elasticities were all
negative and significantly different from zero, meaning that a change in the price of water induces
water-conserving activities. On the other hand, the indirect effects were much smaller than direct effects
(indirect effects account for 17 percent and direct effects account for the rest). This can be explained by the fact
that water price influences technology choices but it was not the only determinant of technology improvement.
Similarly, the water price had a relatively small influence on crop choice because the price of water is often a
small share of the cost of agricultural production. (3) Water use can be reduced significantly by switching
irrigation technology from gravity to drip or sprinkler in most crops, including: citrus, grapes, deciduous tree
crops, and truck crops. Of these, citrus shows the most striking changes. When irrigation technology switches
from gravity to drip irrigation, efficiency (water use per acre) increases by about 50 percent.
This study illustrates how water use efficiency can be improved through pricing incentives. In addition, it shows
how different prices for different crop and technology can induce technology and crop changes.
Abernethy, C. L., Sally, H. Lonsway, K. and Maman, C. 2000. “Farmer-based Financing of Operations in the Niger Valley
Irrigation Scheme.” Research report 37, International Water Management Institute, Colombo, Sri Lanka.
Ahmad, B. 2002. “Implications of Water Pricing in Pakistan.” FAO Regional Office for the near East. Cairo, Egypt.
Alaerts, G. 2003. “Indonesia: Preliminary Assessment of the Impact of the First Phase of Its Irrigation Reform
(1999-2002)”. Jakarta Office, Rural Development and Natural Resources Unit, World Bank.
Awati County Government. 2002. Report of Water Management Institutional Reform in Fengshou Irrigation District.
Awati County, Xinjiang, China. (in Chinese).
Azevedo, L.G.. 1997. “Brazil.” In Water Pricing Experiences An International Perspective, ed. A. Dinar and A.
Subramanian. Technical Paper Number 386, World Bank, Washington, D.C.
Belsare, Er. S. 2001. “Participatory Irrigation Management in Katepurna Irrigation Project: A Success Story.” ICID
Watsave Young Professional Award Winning Paper. New Delhi, India.
Berbel, J. and Gomez-Limon, J.A. 2000. “Multicriteria analysis of derived water demand functions: a Spanish case
study.”Agricultural Systems 63(1):49-72, January.
Bosworth, B. G. Cornish, C. Perry, and F. van Steenbergen. 2002. “Water Charging in Irrigated Agriculture: Lessons from
the Literature.” Report OD 145, HR Wallingford Ltd, Wallingford, UK.
Bruns, B., and Helmi. 1996. “Participatory irrigation management in Indonesia: lessons from experience and issues for the
future.” Paper presented at the Indonesian National Workshop on participatory irrigation management, November
Bueren M. and MacDonald D.H. 2004. “Addressing water-related externalities: issues for consideration.”Paper presented
at the Water Policy Workshop convened by the Australian Agricultural and Resource Economics Society,
Melbourne, Australia.
Cornish, G. A. and Perry, C. J. 2003. “Water Charging in Irrigated Agriculture: Lessons from the Field.” Report OD 150.
HR Wallingford Ltd, Wallingford, UK.
Coward, E.W. 1980. Irrigation and Agricultural Development in Asia. Ithaca, New York: Cornell University Press.
Del Amor Garcia, F. 2000. “Modernization Plan of Mula Traditional Irrigation.” ICID WatSave Award 2000: Innovative
Water Management Award. New Delhi, India.
Destro, S. 1997. “Italy.” In: Water Pricing Experiences: An International Perspective. ed. A. Dinar and A. Subramanian.
Technical Paper Number 386, World Bank, Washington, D.C.
Dinar, A. and Mody, J. 2004. “Irrigation water management policies: Allocation and pricing principles and implementation
experience” Natural Resources Forum 28 (2) 112.
Dinar, A. 1994. “Impact of Energy Cost and Water Resource Availability and Quality on Agriculture and Groundwater
Quality in California.” Resources and Energy Economics 16: 47-66.
Easter, K. W. 1993. “Economic Failure Plagues Developing Countries’ Public Irrigation: An Assurance Problem.” Water
Resources Research 29(7): 1913-22.
Easter, K. W. 2003. “Cost Allocation and the Distribution of Costs For Irrigation in Andhra Pradesh, India.” Unpublished
paper, University of Minnesota, Department of Applied Economics.
Govt. of the People’s Republic of Bangladesh. 2000. “National Water Management Plan Project.” Draft Development
Strategy, Vol. 5. Ministry of Water Resources. Dhaka.
Hamdane, A. 2002. “Irrigation Water Pricing Policy in Tunisia.” FAO Regional Office for the Near East, Cairo, Egypt.
Hatzius, T. 2002. “The case of new water fee system in the Republic of Macedonia.” Series paper of MAINTAIN-case
studies. Eschborn, Germany.
Hatzius, T. 2000. “The Case of New Water Fee System in the Republic of Macedonia: Market and Non-Market Failure in
Path Dependent Institutional Reform, Division 45-Rural Development.” MAINTAIN-Case Study no.7.
Hearne, R. R. and Easter, K.W. 1995. “Water Allocation and Water Markets: An Analysis of Gains-from-Trade in Chile.”
Technical Paper 315. World Bank, Washington, D.C.
Howe, C.W. 1979. Natural Resource Economics. New York: John Wiley and Sons, Inc.
Howe, C.W. 1997. “Increasing Efficiency in Water Markets: Examples from the Western United States.” In Water
Marketing—the Next Generation, ed. T. Anderson and P.J. Hill. London: Rowman & Littlefield Publishers, Inc.
Huppert, W. and Urban, K. 1999. “Institutional Analysis of Water Delivery and Maintenance Service Provision in
Irrigation: The Example of the Jordan Valley.” Series paper of MAINTAIN-case studies. Eschborn, Germany.
Johansson R.C., Tsur Y., Roe, T.L., Doukkali R. and Dinar A. 2002. “Pricing irrigation water: a review of theory and
practice.” Water Policy, 4: 173-199.
Johnson III, S. H.,Vermillion, D., Svendsen, M., Wang, X., Zhang, X. and Mao, X. 1996. “Management Reform and
Performance Changes in Two Irrigation Districts in the North China Plain.” Short Report Series on Locally
Managed Irrigation, Irrigation Water Management Institute. Colombo, Sri Lanka.
Jones, W. I. 1995. “The World Bank and Irrigation.” Operations Evaluation Study, World Bank, Washington, D.C.
Kemper, K, Goncalves, J. and Bezerra, F.W.B. 1999. “Water allocation and trading in the Cariri Region—Cerar, Brazil”.
In “Institutional frameworks in successful Water Markets: Brazil, Spain and Colorado, USA. Technical Paper
Number 427,” ed. M. Marino and K. Kemper. World Bank, Washington, D.C.
Kemper, K, and L.D. Simpson. 1999. “The Water Market in the Northern Colorado Water Conservancy
District-Institutional Implications.” In “Institutional frameworks in successful Water Markets: Brazil, Spain and
Colorado, USA. Technical Paper Number 427,” ed. M. Marino and K. Kemper. World Bank, Washington, D.C.
Kloezen, W. H., Garces-Restrepo, C. and Johnson III, S. H. 1997. “Impact assessment of irrigation management transfer in
the Alto Rio Lerma Irrigation District, Mexico.” Research Report 15, International Irrigation Management
Institute, Colombo, Sri Lanka.
Lemos, M.C. and Oliveira, J.L.F. 2004. “Can Water Reform Survive Politics? Institutional Change and River Basin
Management in Ceará, Northeast Brazil.” World Development 32 (12).
Lin, Z. 2003. “Participatory management by farmers - Local Incentives for Self-financing Irrigation and Drainage Districts
in China.” Environment and Social Development East Asia and Pacific Region Discussion Paper, World Bank,
Washington, D.C.
Maass, A. and Anderson, R.L.1978. …And the Desert Shall Rejoice Conflict, Growth, and Justice in Arid Environments.
Cambridge, Massachusetts: The MIT Press.
McNeill, R. and Tate, D. 1991. “Guidelines for Municipal Water Pricing.” Social Sciences Series No. 25, Environment
Canada, Ottawa, ONT.
Marino, M., and K. Kemper, eds. 1999. “Institutional Frameworks in Successful Water Markets: Brazil, Spain, and
Colorado USA.” Technical Paper 427, World Bank, Washington, D.C.
Massarutto, A. 2002. “Irrigation Water Demand in Europe: The Impact of Agenda 2000 and the `Water Framework
Directive.” Economics Working Paper Series, Università degli Studi di Udine, Italy.
Morris J., E.K. Weatherhead, J. Mills, J.A. Dunderdale, A.M. Hess, D.J. Gowing, C. Sanders, and J.W. Knox. 1997. “Spray
Irrigation Cost-Benefit Study.” Cranfield University, UK.
Naik, G.. and Kalro, A.H. 2000.“A Methodology For Assessing Impact Of Irrigation Management Transfer From Farmers’
Perspective.” Water Policy 2.
öKo Inc. 2001. “Agricultural Water Management Policies in Bulgaria, Hungary, Romania and Slovakia.” Final Report,
oKo Inc., Budapest.
Palacios, E.V. 1999. “Benefits and Second Generation Problems of Irrigation Management Transfer in Mexico.” Economic
Development Institute Participatory Irrigation Management Case Studies Series, Economic Development Institute,
World Bank and Irrigation Water Management Institute.
Perry, C. J. 1995. “Alternative Approaches to Cost Sharing for Water Service to Agriculture in Egypt.” Research Report 2,
International Irrigation Management Institute, Colombo, Sri Lanka.
Perry, C. J. 2001. “Charging for Irrigation Water: the Issues and Options, with a Case Study from Iran.” Research report
52, International Irrigation Management Institute, Colombo, Sri Lanka.
Prato, T. 1998. Natural Resource and Environmental Economics. Ames, Iowa: Iowa State University Press.
Queensland Government, Natural Resource and Mines 2004. “Water Resource Charges—submissions summary.”
Republic of South Africa, Department of Water Affairs and Forestry 2004. “National Water Resource Strategy.”
Samad, M. and Vermillion, D. 1998. “Assessment of Participatory Management of Irrigation Schemes in Sri Lanka: Partial
Reforms, Partial Benefits.” Paper presented at the Annual Conference of the International Association for the
Study of Common Property, Vancouver, Canada, June 10-14.
Schoengold, K. Sunding, D. and Moreno, G. 2004. “Panel Estimation of Agricultural Water Demand Based on an Episode
of rate reform.” Paper presented at the AAEA meeting, Denver, CO, Aug 1-4.
Simas, J. 2002. “Issues Affecting Irrigation and Drainage Sectors in Latin America: Lessons from Mexico, Argentina and
Brazil.” Technical Paper Number 524 in Institutional Reform for Irrigation and Drainage. World Bank,
Washington, D.C.
Svendsen, M. and Gladys, N. 1996. Irrigation Management Transfer in Turkey: Early Experience with a National Program
under Rapid Implementation. The Short Report Series on Locally Managed Irrigation, Irrigation Water
Management Institute, Colombo, Sri Lanka.
Svendsen, M., Trava, J. and S.H. Johnson III. 1997. “Participatory Irrigation Management: Benefits and Second Generation
Problems.” International Irrigation Management Institute, Colombo, Sri Lanka.
Tarrech, R. Marino, M. and Zwicker, G. 1999. “The Siurana-Riudecanyes Irrigation Subscribers Association And Water
Market System.” In “Institutional Frameworks In Successful Water Markets: Brazil, Spain And Colorado, USA.”
Technical Paper Number 427, ed. M. Marino and K. Kemper, World Bank, Washington. D.C.
Thema, J. M. 1997. “Botswana.” In “Water Pricing Experiences: An International Perspective,” ed. A. Dinar and A.
Subramanian, Technical Paper Number 386. World Bank, Washington, D.C.
Tiwari, D. and Dinar, A. 2003. “The Role and Use of Economic Incentives in Irrigated Agriculture.” Working Paper,
World Bank, Washington, D.C.
Tsur, Y. and Dinar, A. 1998. “On The Relative Efficiency Of Alternative Methods For Pricing Irrigation Water And Their
Implementation.” World Bank Economic Review 11.
Wang, X. and Lu, J. 1999. “The Application of IC (Integrated Circuit) Cards in Efficient Irrigation Management System.”
China Water Resources 46 (October) (in Chinese).
World Bank. 1996. “Implementation Completion Report, Indonesia, Irrigation Subsector II (O&M) project”. Agriculture
Operations Division, Country Department III, East Asia and Pacific Region, World Bank, Washington D.C.
World Bank. 1998. “India—Water Resource Management Sector Review.”Rural Development Unit South Asia Region,
World Bank, Washington, D.C.
World Bank. 1995. Morocco Water Sector Review. Washington, D.C.
World Bank. 2003a. “Project Appraisal Document for Republic of Yemen Sana’a Basin Water Management Project,”
Water, Environment, Social and Rural Development Department, Middle East and North Africa Region,
Washington, D.C.
World Bank, South Asia Development Unit. 2003b. “The Incidence of Canal Irrigation Subsidies in India, A Policy Note.”
Washington, D.C., June.
Yaron, D. 1997. “Israel.” In “Water Pricing Experiences: An International Perspective.” Ed. A. Dinar and A. Subramanian,
Technical Paper Number 386. World Bank, Washington, D.C.
Young, H.P. ed. 1985. Cost Allocation: Methods, Principles, Applications. Amsterdam :North-Holland Elsevier Science.
Young, H.P., Okada, N., and Hashimoto, T. 1982. “Cost allocation in Water Resource Development” Water Resource
Research Vol.18.(3).
Zekri, S. and Easter, K. W. 2003. “Institutional and Organizational Reforms in LDCs: Management Transfers, Private
Managers and Water Markets.” Draft paper, Dept. of Applied Economics, University of Minnesota, June 19.
Zhang, L., Liu, J. and Hu, D. 2003. “The Benefit and Problems of Water User Associations.” Issues of Agriculture
Economics Vol.2 (In Chinese).
“Price,” “fee,” and “charges” are used interchangeably in this report. In all cases, we mean the amount of money asked or
given for a good or service. A price, fee, or charge for water is money asked or given for the water itself or the service of
delivering the water, or both. This is in contrast to the term “tariff,” which