Research Reports

IWMI’s mission is to improve water and land resources management for food, livelihoodsand nature. In serving this mission, IWMI concentrates on the integration of policies,technologies and management systems to achieve workable solutions to real problems—practical, relevant results in the field of irrigation and water and land resources.

The publications in this series cover a wide range of subjects— from computermodeling to experience with water users associations— and vary in content from directlyapplicable research to more basic studies, on which applied work ultimately depends.Some research reports are narrowly focused, analytical, and detailed empirical studies;others are wide-ranging and synthetic overviews of generic problems.

Although most of the reports are published by IWMI staff and their collaborators,we welcome contributions from others. Each report is reviewed internally by IWMI’s ownstaff and Fellows, and by external reviewers. The reports are published and distributedboth in hard copy and electronically (www.iwmi.org) and where possible all data andanalyses will be available as separate downloadable files. Reports may be copied freelyand cited with due acknowledgment.

ii

Research Report 52

Charging for Irrigation Water:The Issues and Options,with a Case Study from Iran

International Water Management InstituteP O Box 2075, Colombo, Sri Lanka

C. J. Perry

ii

IWMI receives its principal funding from 58 governments, private foundations, and

international and regional organizations known as the Consultative Group on

International Agricultural Research (CGIAR). Support is also given by the Governments

of Pakistan, South Africa and Sri Lanka.

The case study included in this report is based on data made available by the IranianAgricultural Engineering Research Institute’s Field Office at Esfahan, as part of IWMI’scollaborative program of research, funded by the Government of Iran, and benefited fromdiscussions with staff there and with Dr. M. Bybordi. Earlier drafts of the report were muchimproved following general comments from Hammond Murray-Rust, David Seckler, DougMerrey, Intizar Hussain and Amanda Perry, and particularly detailed comments from RandolphBarker.

The author: C. J. Perry is an IWMI Fellow, International Water 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. Colombo, Sri Lanka: International Water Management

Institute.

/ irrigation management / productivity / water allocation / water use efficiency / maintenance

/ operation / cost recovery / user charges / water pricing / water shortage / economic aspects

/ political aspects / case studies / salinity / Iran /

ISBN 92-9090-427-5

ISSN 1026-0862

Copyright © 2001, by IWMI. All rights reserved.

Please direct inquiries and comments to: iwmi-research-news@cgiar.org

iiiiii

Contents

Summary v

Introduction 1

The Rationale for Water Pricing 1

Water Charges for Cost Recovery 3

Water Charges to Encourage Efficient Use 3

A Case Study: The Zayandeh Rud Basin, Esfahan Province, Iran 7

An Alternative Approach to Improving Productivity 14

Conclusions 14

Literature Cited 16

iv

v

Summary

Inadequate funding for maintenance of irrigationworks and emerging shortages of water areprevalent. The use of water charges to generateresources for maintenance and to reduce demandis widely advocated. Examples from other utilities,and from the domestic/industrial sectors of watersupply suggest the approach could be effective.

In developing countries, the facilities requiredfor measured and controlled delivery of irrigationare rarely in place, and would require a massiveinvestment in physical, legal and administrativeinfrastructure.

To be effective in curtailing demand, themarginal price of water must be significant. Theprice levels required to cover operation andmaintenance (O&M) costs are too low to have asubstantial impact on demand, much less toactually bring supply and demand into balance.On the other hand, the prices required to control

demand are unlikely to be within the politicallyfeasible range.

Furthermore, water supplied is a propermeasure of service in domestic and industrialuses. But in irrigation, and especially as the waterresource itself becomes constrained, waterconsumption is the appropriate unit for wateraccounting. This is exceptionally difficult tomeasure.

An alternative approach to cope with shortagewould focus on assigning volumes to specificuses—effectively rationing water where demandexceeds supply. This approach has a number ofpotential benefits including simplicity,transparency, and the potential to tailor allocationsspecifically to hydrological situations, particularlywhere salinity is a problem.

Data from Iran are presented to support thesecontentions.

1

Charging for Irrigation Water: The Issues and Options,with a Case Study from Iran

C. J. Perry

Introduction

and water services. A primary target for theseinterventions is irrigation, because it is by far thelargest consumer of water—typically 80 percent—in most countries where shortage is aproblem.

In this report, we first set out some basicissues related to the introduction of effectivewater pricing for irrigation services. We arguethat implementation has a number of limitationsin the irrigation sector, and that the scope forrealizing benefits may be limited in relation tothe physical, economic, financial and politicalissues of implementation. In the second part ofthe report, readily available data from Iran areused as a case study to support thesecontentions.

The Rationale for Water Pricing

beneficiaries of irrigation—who typically are aprivileged group in most agrarian economies—receive their service at the expense of theeconomy in general,1 as scarce public resourcesare used first to finance project construction and

1Probably the most important benefit of irrigation is the overall reduction in food prices resulting from increased production. Thus the indirectbeneficiaries of irrigation, the consumers of cheaper food, should be happy to subsidize irrigation development through taxes. Within irrigatedagriculture, or within an individual project, however, the farmer who receives irrigation and benefits from it is clearly privileged in relation to afarmer who does not receive irrigation water, and a service charge may be appropriate to recover a proportion of the benefits or to cover thecosts of the service.

Until recently, water has been plentiful in mostcountries, and the role of water pricing as ameans to ensure efficient allocation andproductive use has attracted little attention.

Now, water is manifestly scarce in manycountries (Gleick 1996; Postel 1996; Seckler etal. 1998); individuals, agencies and internationaldeclarations advise that water should be treatedas an “economic good” (Briscoe 1996;Rosegrant and Binswanger 1994; ICWE 1992;Global Water Partnership 2000); in parallel withthis development, the maintenance of water-related facilities is often observed to beinadequate (Jones 1995). These two issuestogether have provided an impetus for theintroduction of various forms of pricing for water

In some irrigation projects water is provided as afree service. Elsewhere, even the low charges,supposed to be collected are, in fact, notcollected (World Bank 1986). Where charges arelow, or not collected at all, the direct

2

then to subsidize ongoing O&M. All toofrequently, the public subsidy is not forthcoming,the project deteriorates, productivity falls, andthe scope for charging is reduced by the lowquality of the service—a classic vicious circle ofdownward-spiraling performance.

Where charges are in place, the marginalprice is often zero—for example, where thecharge is fixed by crop or farm area, and doesnot vary with the quantity of water actually used.In this situation, even if the full costs of theservice are recovered, there is no incentive forthe farmers to save water—they will either takeas much as they want while the (fixed) priceallows a profit to be made, or not irrigate at all ifthey cannot make a profit.

Inter-sectoral water transfers (for example,reducing irrigation releases from a dam andallocating the water for domestic use) offer theappearance of greater simplicity and thepossibility to use “bulk” prices, but additionaldifficulties arise here too. First, it is likely thatthe value of water in domestic, industrial, andcommercial uses will be far higher than inagriculture. IWMI’s research (Molden et al. 1998)shows values of water consumed to be between$0.03 and 0.90 per m3. The value of water innonagricultural uses is typically at least doublethe maximum agricultural value and, of course,the value of water for human consumption intimes of extreme shortage is effectively infinite.

This leads analysts to believe that there isscope for enormous increases in economicbenefit by market-based transfers of water fromlow-value agricultural use to high-value new usesin other sectors. The flaw in this argument, ofcourse, is that once rather small volumes of

water have been transferred to the high-valueuses, the marginal value falls sharply intonegative territory. The difference between dyingof thirst and drowning is less than a tubful ofwater. And if inter-sectoral benefits of markettransfers are to be realized, then the entireagriculture sector must be covered byinfrastructure, and bureaucracy required for themeasurement, billing, and transfer of water fromthe least-productive parts of agriculture to themore-productive areas. The potential costs andbenefits for Egypt have been analyzed (Perry1995), and the likelihood is that costs wouldsubstantially exceed the benefits.

It is clear from this brief introduction thatthere are a number of reasons, each withdiffering purposes, for recommending watercharges. The three most common are

· to recover the cost of providing the service(either the full cost, including capitalexpenses, or the ongoing O&M cost, orsome intermediate level)

· to provide an incentive for the efficient useof scarce water resources

· as a benefit tax on those receiving waterservices to provide potential resources forfurther investment to the benefit of others insociety

Each of these three objectives requiresspecific levels and structures of pricing to beachieved. Here we concentrate on the first two—as a means of cost recovery, and a means ofproviding an incentive for efficient use.

3

Water Charges for Cost Recovery

efficient use, and on the level of chargesrequired to achieve that goal. It is useful, first, toestablish a benchmark of the level of chargesthat might be required to achieve recovery of theservice costs. Data in this area are scarce, andwe rely here on two studies that appear to beexceptionally comprehensive and credible. Thefirst (ISPAN 1993) reviewed the situation inEgypt, and concluded that the full cost ofirrigation and drainage O&M amounted toUS$52/ha in 1995 US dollars. This equates toabout $0.003 per m3 of water delivered.

The second is a study prepared for theWorld Bank’s appraisal of the Haryana WaterResources Project (1995). It found the cost ofO&M to be $10/ha—again in 1995 US dollars—which amounts to $0.002 per m3. These studiescover vastly different situations—Egypt havingan intensively managed system, delivering largequantities of water to highly varied croppingpatterns, with very extensive drainage facilities.Haryana’s system is very water-short—availability is about one-third of that in Egypt,management is extensive rather than intensive,and drainage costs are minor. What the twosystems have in common is that they are bothsuccessful and productive, and have beenoperating for many years.

Water Charges to Encourage Efficient Use

The second objective adds further challenges. Toachieve an incentive for efficient water use, theprice of water must be directly related to thevolume delivered. Conceptually, this is identicalto an electricity meter where the consumer candecide to switch off or switch on a particulardevice, and experience a directly proportionalresponse in the electricity bill. Fortunately,

electricity reacts instantly once the demand atthe end of the wire is perceived, and the flowdown the wire can vary sharply withoutoperational consequence (except at thegenerator). The water flow is rather slow incanals, and must be issued well in advance ofdistant demands. Changes in demand during theperiod of distribution will result either in

Cost recovery requires a politically sensitive choiceas to the extent of cost recovery—full recovery ofcapital and O&M costs at realistic interest rates, orpartial recovery at subsidized rates.

The assessed total cost of water supply (anddrainage) services must then be distributedamong various beneficiaries—farmers, villagesthat receive domestic water supplies, floodcontrol downstream, and sometimes hydropower.With this complex set of decisions made, aspecific target revenue from irrigators can bedefined. In fact, even this simple objective israther complex in reality because the costs ofO&M vary over time, especially when majorreplacement or modernization work becomesdue, or when dealing with major events such asfloods, which may damage infrastructure.

Thus even this simplest of chargingstructures will require political debate anddecisions, planning, predictions, and financialmanagement if a reasonably uniform pattern ofcharging is to be maintained. In the State ofVictoria, Australia, irrigation systems are obligedto project expenditures over a 100-year periodand adjust revenues to meet these with a stablerevenue stream (Langford et al. 1999).

In the next section, attention is focused onthe potential for using water prices to encourage

4

shortages—if demand increases unexpectedly—or in surpluses and spills if demand decreases.

We see already that transferring the electricityparadigm to water delivery has significantoperational implications; if pricing of irrigationwater is to be effective in reducing demand insurface irrigation systems, then specific andcomprehensive regulatory, operational andeconomic criteria must all be met.

Regulatory Requirements

An orderly system of distributing water must bein place through some existing and respectedregulatory framework for allocating water amongfarmers—rules and procedures defining rightsand responsibilities; priorities in case of shortageor excess supplies; penalties for breach of rules,and so on. If this is not the case—or ifregulations are not observed (if farmers takewater at will, manipulate gate settings, toleratesignificant interference in water, or do not payassessed charges) then there is no immediatescope for improving water distribution throughpricing, and attention should first be given toclarifying and enforcing water rights and therules of water distribution. To then move on tovolumetric supply, as required for volumetricpricing, additional procedures for bothmeasurement of the quantity delivered,acceptable to the farmer and the billing agency,and accounting for partial deliveries, misseddeliveries, excess deliveries, or late deliveriesmust be in place.

Operational Requirements

If pricing is to encourage changes in the patternof demand at the level of the individual farm,then system operations must be such as topermit differentiated deliveries at this level. Theoperational implications of delivering andmeasuring a differentiated service at the level of

the individual farm poses significant difficulties inmany irrigation schemes, especially large surfacesystems (which comprise the major users ofwater). Such irrigation schemes comprisehundreds, if not thousands or even millions ofindividual consumers.

Measurement and charging at the farm levelwill require substantial investments in equipment,and an associated administrative bureaucracy tocollect and collate data on farm-level deliveries,and undertake the billing process.

We recognize that there are procedures andtechnical approaches to resolve this problem.The approach has proved feasible in countriessuch as Australia and the USA, but it isimportant to recognize that farm sizes in boththese countries are typically 10 to 100 timeslarger than those in many developing countries.This directly indicates the associated additionalcomplexities of billing and also the far greaterhydraulic complexities of delivering (or notdelivering) incremental water supplies toindividual farmers. The hydraulic complexityresults from the fact that changes in flow ratesat one point in an open surface system generateimpacts on all downstream flows, which can bedealt with, but using rather different infrastructurethan most systems now have. In Israel, whilefarm sizes are often comparable to those indeveloping countries, the infrastructure is mostlysuch as to allow precise and verifiable deliveryof water to individual farms—and the concept ofscarcity was dominant in the original systemdesign rather than emerging over time (as is thecase in many existing systems). We see veryfew large-scale irrigation systems in developingcountries where the providing of infrastructure forsuch service would not involve effectivelyreconstructing the system.

An alternative approach in the context ofdeveloping countries, to avoid the “small farm”problem, is to deliver water to an intermediatepoint—a farmer organization, a lateral, or avillage—on the basis of volumetric pricing, andallow the farmers to distribute the water

5

“internally.” This has the attraction, from theagency’s viewpoint, of devolving the most difficultpart of the operation (the actual interfacebetween “supply” and “demand”) to others—acommon feature in many participatorymanagement approaches. But the overallmanagement task of delivering differentiatedsupplies to individual farms remains, as indeeddoes the required regulatory framework, the needto measure and bill, and so on. If these aspectsare missing then the direct link between serviceand payment is lost, and the efficiency incentivethat pricing is designed to produce is neutralized.

Economic Requirements

If the charging system is to have an impact onconsumption, then the system of payment mustbe such as to induce the desired economicresponse. We can usefully distinguish betweentwo desired responses; the first is simply tomake the user aware of an incremental chargerelated to incremental use, and to encourageavoidance of waste. This is analogous to thesmall cost incurred in leaving a light on—we areconscious of it, we will tend to avoid leavinglights on for long periods during daylight hours,but may leave lights on at night for theconvenience of finding a stairway well lit ratherthan having to seek the switch in the dark.

The second objective of marginal pricing isactually to achieve a balance between supplyand demand through the pricing mechanism—inother words, to find that price that will optimizedemand for and allocation of any givenavailability of water. This, of course, is a farmore complex pricing objective because thevalue of water varies very sharply by crop, bystage of growth, and by soil type, and mostdifficult of all, depending on whether it hasrained, is raining, or will rain soon.

Which objective of pricing is sought—tobalance supply and demand or simply to reduce

demand—must be very clearly understood in theprocess of designing a pricing system.

If the objective of marginal water pricing isto balance supply and demand, then the pricecharged must be significant in relation to thebenefits derived from using water. As notedabove, IWMI’s research has shown that thevalue of water consumed in agriculture rangesfrom $0.05 to 0.90 per m3, with the greatmajority of observations falling in the order of$0.10–0.20 per m3. We can assume, on thisbasis, that prices of less than a few cents percubic meter will have little effect on demand.Indeed, research has shown in the case ofEgypt (Perry 1995) that the price required toinduce a 15 percent fall in demand for waterwould have reduced farm incomes by 25percent.

If the objective is simply to reduce demandto some unspecified degree, then two furtherissues must be considered: first, it will benecessary to have additional quantitativecontrols on supply (because the proposed pricewill not achieve the necessary balance); andsecond, the question of the balance between thecost incurred to induce the intermediate fall indemand and the actual water “saved” should beconsidered. In this case, the water savedthrough pricing is zero (because the level ofconsumption is determined by the quantitativecontrols, which would work perfectly well in theabsence of pricing).

The Issue of “Saving” Water

One final problem plagues the achievement ofefficiency benefits through irrigation waterpricing, even if infrastructural, operational,political and bureaucratic hurdles can becrossed. This issue concerns the nature of“efficiency” in water resource systems.

In the municipal and industrial sectors ofwater supply, the agency providing the service

6

appropriately considers the volume and quality ofwater delivered as measures of service. This isappropriate because the agency will have paid tocapture the water, treat it to meet specifiedstandards, and deliver the water through itsdistribution network. Reductions in demand fromusers will save the agency money on each ofthese activities. Also if users waste water bytaking more than they need to fulfill a particularobjective, the agency (or another agencydownstream) would again have to capture, treat,and distribute this water—incurring further costs.

But when basin-level resource scarcitybecomes the framework for analysis (rather thansector-specific cost accounting) the picturechanges dramatically: we must focus not onwater diversions, but on consumption, andrecognize that where “wasted” water isrecaptured downstream, the resource cost of the“wastage” is essentially zero unless substantialpollution has occurred in the recycling process.An approach to addressing “efficiency” and saltpollution has been suggested elsewhere (Kellerand Keller 1995). In the context of overall watershortage it is not diversion, but consumption—through evapotranspiration, pollution, or loss—which is the concern.

We have already argued that it is difficult toenvisage a system of accurate measurement ofdeliveries to farms in large-scale irrigationsystems in developing countries. Themeasurement of consumption is inherently muchmore complex because it involves spatiallydistributed return flows to drains, rises and fallsin local water tables, rainfall, and so on.

In some cases, moreover, excess waterdelivered during one season recharges anaquifer, which provides water in a subsequentdry season when there may be no alternativesupplies, and indeed the water is considerably

more productive. Such situations are common: inparts of China, India and Pakistan, the areairrigated by wells is comparable to (and in someareas, more than) the area irrigated by surfacesupplies—but many of these wells are insurface-irrigated commands and thus dependsubstantially on recharge from surface “losses.”In all heavily developed basins, with more thanone diversion point on the river, a proportion ofthe “losses” from upstream projects formsinflows to downstream projects.

In any of these circumstances, an increasein “efficiency” will reduce the flows to the aquifer,drain, or river, by roughly the same amount as“losses” have been reduced. This is entirelydesirable where the aquifer is saline, or so deepas to be beyond economic recovery, or wheredrainage flows cannot be recaptured fordownstream use. In some cases, the effectiveloss in terms of productive potential maysometimes, on further examination, turn out tobe far greater than initially estimated whenexcess water picks up local salt from the soiland returns this extra polluting load to theriver—reducing potential productivitydownstream.

We believe that this complex situation makesit virtually impossible to formulate a pricingstructure that can serve the multiple objectivesof water charges, and the various hydrologicalsituations where it may be desirable todecrease, stabilize, or increase the watersupplied to irrigation systems. Certainly chargingfor water can give incentives, but broaderobjectives of balancing supply and demand orstabilizing environmental impacts are unrealistic.Even the introduction of incentives will requiremassive improvement in the infrastructure,management and regulatory framework thatcharacterizes most irrigation systems.

7

Water Charges and WaterConsumption

As outlined above, to assess the potentialimpact of pricing water, a number of issues mustbe addressed:

· the nature and extent of water “losses”and potential savings

· the cost of “saving” water and the cost ofsurface supplies

· the cost of water and its value to farmers

· the impact of charges on water demand

· salinity and salinization

This disaggregation is attempted below.

The Nature and Extent of Water“Losses” and Potential Savings

How much water can reach and be consumedby the crops out of what has been stored,diverted, conveyed, delivered and applied to thefield is central to project planning, and the day-to-day operation of irrigation systems. “Losses”in this process—the quantities of water releasedfrom storage that do not reach the crop—are thelegitimate concern of the operators andbeneficiaries.

A Case Study: The Zayandeh Rud Basin, Esfahan Province, Iran

The Zayandeh Rud basin (figure 1) has an areaof about 4 million hectares, of which 8 percent(300,000 ha) is irrigated—about 3 percent(130,000 ha) under surface irrigation in large,government-constructed schemes drawing fromthe Zayandeh river, and the rest in smallgroundwater or kanat-supplied areas that areprivately owned and operated.

Annual rainfall in the irrigated areas is 100–200 mm and annual ET is in the range of1,500–2,000 mm, so irrigation is essential foragriculture. Presently available surface waterresources from the Zayandeh river, augmentedby interbasin transfers, total 14 billion m3. Theseresources are essentially fully consumed by theexisting facilities, and residual flows reaching theGavkhuny marsh are limited in quantity, and ofunusable quality. With the exception of smallareas with positive recharge from nearby hills,the groundwater balance is marginal, and insome areas consumption exceeds recharge.

The Zayandeh Rud basin is thus extremelyshort of water, and the distribution of availablesupplies is unequal in terms of both quantity andquality. Projects drawing from the river in theupper reaches are relatively plentifully suppliedwith good quality water; rice is a common cropand irrigation intensities are high. Movingdownstream, irrigation supplies decline both inquantity and quality until the lower reacheswhere the quality of water from the riverbecomes marginal—and eventually unusable.Farmers supplement deliveries by pumping fromgroundwater and drains, wherever such water isfit for use, and in some cases by using water ofextremely poor quality.

8

But two separate and often coexistentfeatures dramatically alter the appropriateperspective once resource planning (rather thanproject planning, or local management) is theobjective. These features are reuse of “losses,”and basin-level competition.

Commonly, for surface irrigation schemessuch as those found in Iran (Bybordi 1989), theratio of water delivered to the project to thewater consumed by crops is about three toone—that is, for every 3 m3 delivered at thehead of an irrigation system, only 1 m3 istranspired by the crop. In such a case, the“efficiency” would be calculated as 33 percent,and 67 percent would be assumed to be “lost.”

Reuse of losses occurs most commonlywhen

FIGURE 1.

The Zayandeh Rud basin.

· seepage from canals and fields reachesaquifers that can be exploited

· farmers recover water from local drains

· the outflow from drains flows to a river thathas downstream diversion works

However, a critical examination of localconditions may show that the potential torecover such “losses” is limited by existingrecycling—which may be active (wells; pumpingfrom drains) or passive (drains flowing to theriver and on to downstream diversions; capillaryrise from shallow water tables to supportcrops)—in other words, the losses are alreadybeing recovered!

9

The Cost of “Saving” Water and theCost of Surface Supplies—Incentivesat the Farm Level

Water delivered to the farmer goes to one ofthree uses: it is utilized by the crop, runs offthe field into a drain, or percolates to the watertable. The farmer benefits from the first ofthese uses, and may suffer—throughwaterlogging or leaching of nutrients—from theother two.

If the price of water delivered to the farm isincreased, the farmer will have an incentive toreduce his or her demand for water by reducingthe water going to the drains and to the aquifer.

First, it is important to note that thefarmer’s incentive to reduce losses is linkedonly to those losses that are within his or hercontrol while overall irrigation system lossesmay amount to 60–70 percent. In Iran aselsewhere, typically less than half of theseoccurs at farm level, the rest being operationallosses, and seepage from main and distributarycanals. Thus the farm-level losses may amountto 30 percent of the water supplied to theproject—even less if some of this excessdelivery is recovered for further use.

Table 1 calculates the cost per m3 ofreducing required water deliveries by investingin improved irrigation technology at $1,000 perha—a typical figure in Iran for the installation ofsprinkler systems. The cost per m3 saved isshown in the second column ($0.11 per m3 foryear-round, and $0.14 per m3 for single-seasonirrigation), based on the assumptions aboutCrop Demand, Losses, and Costs detailed inthe columns to the right.

Year-round irrigation has a higher total costof operation ($348 compared to $220), butlower cost per unit of water saved (about $0.11

per m3 compared to 0.14 per m3) because thefixed costs are distributed over a larger numberof units. For both situations (year-round andsingle-season, the sensitivity of the cost of watersaved is tested by increasing and decreasingeach parameter by 20 percent. For example, ifthe losses before the improvement are 20percent higher (i.e., 36%) in the single-seasonsituation then the cost of water saved falls to$0.10 per m3; while if the losses are lower by 20percent, then the cost of water saved increasesto $0.21 per m3.

The calculations indicate that the cost to thefarmer of water saved through improved on-farmtechnology will be in the range of $0.08–0.21 perm3 depending on the total volume delivered andthe costs of investment and operation.

Taking the minimum cost ($0.08 per m3),which applies for the case of maximum waterdeliveries, it is interesting to compare this to thecurrent price of irrigation water, which is 20 Rialper m3, or $0.004 per m3. If volumetric pricesare to be used to induce farmers to invest inimproved on-farm technology to save water, thenwater charges would have to exceed $0.08 perm3—a 20-fold increase—for the investment inon-farm water management to be profitable.This, in turn, would result in water charges ofabout $400 per ha for a crop such as wheat(with current yields of around 5 t per ha and aprice of $120 per t, water charges wouldaccount for two-thirds of gross revenues).

These calculations indicate that for basicfield crops, the cost of saving water throughinvestment in improved irrigation technology isunattractive, and the cost of surface water wouldhave to be increased enormously to make suchinvestment an attractive alternative to purchasingadditional water from a volumetrically pricedsurface source.

10

TABLE 2.

Gross and net values of production for major crops.

Crop Land Irrigation Harvest Total Yield Price Gross Gross Net Netpreparation pesticides costs costs (t/ha) (R/ton) value of value of value of value of

seed (R/ha) (R/ha) (R/ha) production production production production(R/ha) (R/ha) ($/ha) (R/ha) ($/ha)

Wheat 350,000 500,000 150,000 1,000,000 6.5 650,000 4,225,000 704 3,225,000 538

Barley 350,000 350,000 150,000 850,000 6.0 500,000 3,000,000 500 2,150,000 358

Maize 350,000 500,000 200,000 1,050,000 6.0 500,000 3,000,000 500 1,950,000 325

Rice 2,500,000 500,000 1,000,000 4,000,000 5.5 1,750,000 9,625,000 1,604 5,625,000 938

Note: R = Iranian Rial. US$1.00 = R6,000 (in 1993).

TABLE 1.

Cost of reducing water deliveries by improved irrigation technology.

Cost of Crop Losses Losses Reduced Capital Interest Life Capital Operational Totalwater demand before after delivery cost % (years) cost/yr. cost cost

($/m3) (m3/ha) (%) (%) (m3/ha) ($/ha) ($) ($/ha/yr.) ($/ha/yr.)

Two season 0.11 10,000 30 10 3,175 1,000 10 10 148 200 348Cost/m3 for...+20% 9 8 12 12 11 11 14-20% 14 17 10 10 11 12 8

Single season 0.14 5,000 30 10 1,587 1,000 10 15 120 100 220Cost/m3 for ...+20% 12 10 15 15 15 13 16-20% 17 21 13 12 14 15 13

11

The Cost of Water and Its Value toFarmers

The value of water to farmers is an extremelycomplex issue. The marginal value of watervaries sharply through the season—water tocomplete the development and harvest of anearly mature, high-value crop will have a veryhigh value; additional water after an irrigation, orrainfall may have a negative value.

The average value of water (value of cropdivided by total water used) is more stable, andis a useful indicator of the general value of waterto farmers. Even here there are manycomplexities:

· The net value of the crop depends onvaluation of the crop itself and cash inputs,any or all of which can be significantlydistorted by market imperfections or taxes/subsidies.

· The valuation of noncash inputs such asland, family labor and draft animals, isextremely difficult.

· The basis for computing water consumptionmay be water diverted to the project; waterapplied to the field; or water actuallyconsumed by the crop (values that, asindicated above, may vary by a factor ofthree).

Table 2 summarizes data from Esfahan ontypical yields of major crops. Gross values ofproduction range from about $400 to about$1,600, while net values are generally some 30percent lower.

IWMI has made extensive studies of thegross value of water per unit of water appliedand per unit of water diverted and consumed.These values range from $0.03 per m3 for rice(water applied) to $0.50 per m3 (waterconsumed) for vegetables and fruits. Typicalgross values per unit of water consumed range

from $0.10 per m3 to $0.20 per m3, and thesevalues compare well with data from ZayandehRud for a range of crops, as shown in table 3,which is based on the crop budgets from table2, and the calculated water requirements for thearea. (IWMI uses international prices to valuecrops; here, local prices have been used, whichappear to be somewhat below world prices).

The values calculated are well within therange of other values obtained in IWMI’s studies,and are consistent with the observed trendselsewhere. Rice, for example, shows poorreturns per unit of water delivered due to thehigh requirements for field preparation andpercolation, but high values per unit of waterconsumed (emphasizing how essential it is toevaluate “losses” in assessing the desirability ofrice cropping).

Present water charges, applied volumetricallyat the current rate of Rial 20 per m3, wouldamount to $30–40 for wheat, barley and maize,and $90 for rice. In each case, this value isabout 5 percent of total revenue, and less than10 percent of net revenues—significant, butsmall in relation to the overall crop budget.

The Impact of Charges on WaterDemand

These values provide a useful guide to evaluatethe importance of water charges at present, theattractiveness of new technologies, theincentives facing the farmer, and the implicationsfor water use.

To simplify the discussion, we focus ontypical values, rather than the ranges andextremes, which have been calculated above.We assume that

· the present price of water is $0.004 per m3

· the cost of reducing water deliveries throughimproved technology is $0.10 per m3

12

· the net value of water delivered to farmers is$0.05 per m3

· the net value of water consumed by the cropis $0.08 per m3

The first two values indicate that volumetricpricing in any form, in the absence of muchhigher water charges, will have very little impacton farmers’ choice of crop or choice of irrigationtechnology.

More detailed consideration of the last twovalues provides important insights into what canhappen if water charges are substantially raised,or if water allocations to farms are physicallylimited below potential demand (for example, ifwater deliveries were restricted to 15,000 m3 perha in rice areas, or 5,000 m3 per ha in non-riceareas).

Note that the value of water consumed iscomparable to the cost of reducing deliveriesthrough improved technology. Put another way, ifthe farmers switch to (say) drip irrigation, andavoid the flows to drains and groundwater thatcurrently occur, they can consume a higherproportion of the water delivered to his farm.Each farmer’s benefit is $0.08 per m3 at presentlevels of productivity, and the cost to each is$0.10 per m3. But conversion to drip or sprinkler

may also be accompanied by higher yields(better water control) and perhaps a shift tohigher-value crops such as fruits and vegetables.

The impact of this would be a desirableincrease in productivity of land, an increase inthe volume of water consumed at the farm level,and a decrease in return flows to drains andaquifers. The latter points are crucial. Improvedcrop growth and yield are almost invariablyassociated with an increase in transpiration, sowe can be quite sure that one of the “uses” ofwater will increase. If the return flows were notrecoverable, the implications for downstreamusers of the increase in transpiration will beneutral—deliveries to the upstream area were inany case a reduction in flows available to thedownstream users. If, however, the return flowswere recoverable downstream, the new situationis that upstream consumption increases, returnflows decrease, and downstream usersexperience reduced availability of water as aresult of improved upstream “efficiency”!

Of course, the new situation offers thepossibility of reducing deliveries to the upstreamusers (who are now getting increased productionby “capturing” more beneficial use of the waterdelivered to their farms). The question thenarises as to whether upstream users will wish toinvest in improved technology if their water

TABLE 3.

Value of water for major crops.

Crop Gross Net Total Net Gross value (¢/m3) Gross value ($/m3)value of value of irrigation water delivered consumed delivered consumed

production production delivered consumed($/ha) ($/ha) (m3/ha) (m3/ha)

Wheat 704 538 8,286 6,800 0.085 0.104 0.065 0.079

Barley 500 358 8,286 6,800 0.060 0.074 0.043 0.053

Maize 500 325 11,214 7,900 0.045 0.063 0.029 0.041

Rice 1,604 938 21,714 9,200 0.074 0.173 0.043 0.102Note: The water delivered is calculated as: (ET - Effective rainfall)/70 percent for maize, wheat and barley, and (ET - Effective rainfall + landpreparation and percolation)/70 percent for rice. The ratio of 70 percent accounts for farm-level efficiency. This is the relevant figure, asinvestments in on-farm improvements can only effect efficiency at this level.

13

allocation is in consequence reduced, and thesavings from their investments are passed toothers.

The Issue of Salt

The water in the Zayandeh river contains anatural salt load. As the water is diverted forirrigation, and any excess water returns to theriver, the downstream salt load increases,reduces, or stays constant, as described below.Whether changes in the salt load are desirableor not depends on local and basin-levelobjectives. In considering this issue it isessential to distinguish between the salt load,and the salt concentration at any given point ofmeasurement.

Suppose the annual water supply releasedfrom the dam for downstream use is 100 millioncubic meters (Mm3), and the salt concentrationis 500 ppm. Then the total salt load deliveredfrom the dam is 50,000 t per yr., and the firstdownstream irrigation project diverts 50 Mm3.

Case 1If 80 percent of the diverted water is consumedby crops, and the unconsumed water is justsufficient to carry out all the incoming salt to theriver in the return flow, then:

· The downstream flow is 60 Mm3.

· The downstream concentration is increasedto 833 ppm.

· The downstream salt load is 50,000 t per yr.

· No salt is accumulating in the irrigated area.

Case 2If the entire volume of water is consumed byirrigation, so that there are no return flows to theriver, then

· The downstream flow is 50 Mm3 per yr.

· The downstream concentration is unchangedat 500 ppm.

· The downstream salt load is 25,000 t per yr.

· In this area, 25,000 t per yr. of salt areaccumulating.

Case 3As in Case 1, except that the soils in theirrigated area are saline, and the return flowspick up a further amount of salt equal to thatdelivered in the incoming irrigation water:

· The downstream flow is 60 Mm3.

· The downstream concentration is increasedto 1,250 ppm.

· The downstream salt load is 75,000 t per yr.

· From the project area 25,000 t per yr. of saltare being removed.

There are, of course, an infinite variety ofsuch possibilities, and there is likely to bevariation within individual projects. Further, thesescenarios change over time; land that is irrigatedfor the first time may often correspond to Case3, but over time, as the salt is leached from theprofile, it moves towards Case 1.

These alternative local scenarios anddownstream impacts have implications forgovernment objectives in water management,and the impact of farmers’ efforts in response towater charges or changed technology.

In Case 1, the local area is kept salt-free,and the salt load in the river is constant, to thedetriment of downstream users. Case 2represents the case of more “efficient” irrigation,and leaves the downstream water qualityunchanged, but threatens the sustainability of

14

Conclusions

irrigation in the hypothetical upstream project assalt builds up in the area. Case 3 is thereverse—the upstream area is improving whilethe salt load downstream is sharply increased. Itis clear from these examples that irrigation“efficiency” and productive, sustainable irrigation

have no precise relationship, and that theoptimum “efficiency” must be locally defined inrelation to impacts in the area concerned andelsewhere. The implications for a pricing policyas a tool in this environment are obscure.

An Alternative Approach to Improving Productivity

It is striking that many of the issues identifiedabove must be analyzed in terms of volumes ofwater. The water available is a volume, and howit is used (through crop ET, to groundwater, todrains) are all volumes, and the salt balance isbased on water and salt volumes and theirspatial and temporal distribution.

It is also relatively easy to define sustainableirrigation in terms of the water that should beapplied, water that should (or should not) returnto the system, and water that should go togroundwater to maintain sustainable balances ineach area. It is also easiest to define thepolitical economy of water in terms of allocationsof water to areas, groups, or individuals.

And finally, it is important to note that theincentives to utilize water productively are madeclear to the farmer just as directly by rationingwater as by trying to establish an appropriatesystem of water charges. In northwest India, thelong-tested warabandi irrigation system (Malhotra1982) is based entirely on ensuring an equitabledistribution (over the land) of limited waterresources. Water charges are not high, and notvolumetric, but because all farmers are water-short, they experiences directly the true value oftheir water ration, and strive to save every dropand maximize its productivity.

m3, while the charge required to substantiallyaffect demand would be much higher—perhaps$0.02–0.05 per m3. This indicates that a chargedesigned to meet the cost recovery objective willhave minimal efficiency impact and that a chargethat meets the efficiency objective will recoverfar more than the costs of O&M, which seemsattractive.

However, water is a complicated naturalresource:

The apparent misuse and waste of irrigationwater, especially in the context of low andsubsidized prices for water and deterioration ofirrigation systems, suggest that charges shouldbe increased to cover the costs of systemoperation, and that pricing mechanisms shouldhave a prominent role in encouraging moreefficient resource use.

The analysis above and data from othercountries suggest that the likely charge neededto cover O&M costs would be $0.003–0.005 per

15

· It is difficult to allocate and measure,especially in large surface systems with manysmall farmers—indeed it is rare to find evensimple proportional allocations beingachieved, with head-end farmers (andprojects) usually getting more water than tailenders.

· The nature and extent of actual losses arerarely easy to assess, due to reuse andrecycling locally and downstream. The extentto which “savings” will be achieved throughwater charges or changes in technology canonly be assessed through a full accounting ofwater flows in the basin.

· The salt load that is found in every basincomplicates the analysis of water distributionand consumption considerably, due toconflicts between “on-site” objectives ofavoiding the accumulation of salt, anddownstream impacts of concentrated saltloads.

Water is also a complicated economic resource:

· Its value varies sharply across time, spaceand use.

· Its value in irrigation is generally a largemultiple of its cost, but the cost of “saving”water as a source of extra water is veryexpensive.

And water is a complicated political resource:

· Farmers are often an important politicalconstituency, and strongly resist increases inthe price of irrigation services.

· The level of price increases that would berequired to have a significant impact ondemand (for example, by a factor of 5–10)would be politically very difficult to enforce.

· The level of water prices that would inducechanges in demand would result insubstantial profits to the supplying agency,which would present a further politicaldifficulty.

In sum, introducing volumetric water chargesis difficult and unlikely to result in water savingswithin the politically feasible range of prices—andintroducing charges at a lower rate will have noimpact on demand, but will significantly increasethe costs of irrigation services. And the “correct”price structure to balance supply and demand willnot be the “correct” price structure required tomeet environmental needs, and also will not bethe “correct” price to meet socio-political needs.

Many of the assumed advantages of waterpricing can be achieved through physical rationingof water, which is also easier to implement andadminister, more transparent, and more readilyadjusted to meet local considerations such asrising or falling groundwater conditions and saltmanagement. For Iran, as in many other countries,uniform water allocations per hectare wouldachieve some very important steps in thereallocation of water from overusing upstreamareas, where groundwater is also relatively freshand abundant, to the water-short and salt-abundant downstream area. Such an approachwould also be politically easier to explain thanresorting to water charges, especially (as is all toolikely) if charges are set well below “value” so thata) farmers upstream will buy even more than theynow get and b) they will have no incentive toinvest in productivity-enhancing technologies.

This suggests that it may be more productive(at least in the short run) to set water charges soas to ensure financial sustainability of theirrigation systems—that is, to recover full O&Mcosts. The objective of increased water useefficiency will be better served by volumetricallocation of water rather than expecting marketsto achieve the desired balance among competingobjectives in production, environment, and socialequity.

16

17

Literature Cited

Briscoe, J. 1996. Water as an economic good: The idea and what it means in practice. Paper presentedto World Congress of ICID, Cairo, Egypt.

Bybordi, M. 1989. Problems in planning of irrigation projects in Iran. In Irrigation: Theory and practice,ed. J. R. Rydzewski and C. F. Ward, 115–123. London: Pentech Press.

Gleick, Peter H., ed. 1996. Basic water requirements for meeting human activities: Meeting humanneeds. Water International 21(2):83–92.

Global Water Partnership. 2000. Vision and framework for action. www.gwpforum.org

ICWE (International Conference on Water and the Environment). 1992. The Dublin Statement andReport of the Conference. WMO, Geneva, Switzerland.

ISPAN (Irrigation Support Project for Asia and the near East). 1993. Irrigation water cost recovery inEgypt: Determination of irrigation water costs. Arlington, Virginia, USA.

Jones, W. I. 1995. The World Bank and irrigation. Washington D.C.: The World Bank.

Keller, A.; and J. Keller. 1995. Effective efficiency: A water use efficiency concept for allocatingfreshwater resources. Discussion Paper 22. Water Resources and Irrigation Division. Arlington,Virginia, USA: Winrock International.

Malhotra, S. P. 1982. The warabandi system and its infrastructure. New Delhi, India: Central Board ofIrrigation and Power.

Langford, K. John; Christine L. Forster; and Duncan M. Malcolm. 1999. Toward a financially sustainableirrigation system. Lessons from the State of Victoria, Australia, 1984–1994. Washington D. C.: WorldBank.

Molden, David; R. Sakthivadivel; Christopher J. Perry; Charlotte de Fraiture; and Wim Kloezen. 1998.Indicators for comparing performance of irrigated agricultural systems. Research Report 20. Colombo,Sri Lanka: International Water Management Institute.

Perry, C. J. 1995. Alternative approaches to cost sharing for water service to agriculture in Egypt.Research Report 2. Colombo, Sri Lanka: International Irrigation Management Institute.

Postel, S. L.; Gretchen C. Daily; and Paul R. Ehrlich. 1996. Human appropriation of renewable freshwater. Science 271: 785–788. (February)

Rosegrant, M.: and H. Binswanger, H. 1994. Markets in tradable water rights: Potential for efficiencygains in developing country water allocation. World Development 22 (22): 1,613–1,625.

Seckler, David; Upali Amerasinghe; David Molden; Radhika de Silva; and Randolph Barker. 1998. Worldwater demand and supply 1990 to 2025: Scenarios and issues. Research Report 19. Colombo SriLanka: International Water Management Institute.

World Bank. 1986. World Bank lending conditionality: A review of cost recovery in irrigation projects.Washington D.C.

Research Reports

38. An Assessment of the Small-Scale Irrigation Management Turnover Program inIndonesia. Douglas L. Vermillion, Madar Samad, Suprodjo Pusposutardjo, SigitS. Arif, and Saiful Rochdyanto, 2000.

39. Water Scarcity and the Role of Storage in Development. Andrew Keller, R. Sakthivadivel,and David Seckler, 2000.

40. Using Datasets from the Internet for Hydrological Modeling: An Example from theKüçük Menderes Basin, Turkey. Martin Lacroix, Geoff Kite, and Peter Droogers, 2000.

41. Urban-Wastewater Reuse for Crop Production in the Water-Short Guanajuato RiverBasin, Mexico. Christopher A. Scott, J. Antonio Zarazúa, and Gilbert Levine, 2000.

42. Comparison of Actual Evapotranspiration from Satellites, Hydrological Modelsand Field Data. Geoff Kite and Peter Droogers, 2000.

43. Integrated Basin Modeling. Geoff Kite and Peter Droogers, 2000.

44. Productivity and Performance of Irrigated Wheat Farms across Canal Commands inthe Lower Indus Basin. Intizar Hussain, Fuard Marikar, and Waqar Jehangir, 2000.

45. Pedaling out of Poverty: Social Impact of a Manual Irrigation Technology in SouthAsia. Tushaar Shah, M. Alam, M. Dinesh Kumar, R. K. Nagar, and Mahendra Singh.2000.

46. Using Remote Sensing Techniques to Evaluate Lining Efficacy of Watercourses.R. Sakthivadivel, Upali A. Amarasinghe, and S. Thiruvengadachari. 2000.

47. Alternate Wet Dry Irrigation in Rice Cultivation: Saving Water and Controlling Malariaand Japanese Encephalitis? Wim van der Hoek, R. Sakthivadivel, MelanieRenshaw, John B. Silver, Martin H. Birley, and Flemming Konradsen. 2000.

48. Predicting Water Availability in Irrigation Tank Cascade Systems: The CascadeWater Balance Model. C. J. Jayatilaka, R. Sakthivadivel, Y. Shinogi, I. W. Makin,and P. Witharana. 2000.

49. Basin-Level Use and Productivity of Water: Examples from South Asia. DavidMolden, R. Sakthivadivel, and Zaigham Habib. 2000.

50. Modeling Scenarios for Water Allocation in the Gediz Basin, Turkey. Geoff Kite,Peter Droogers, Hammond Murray-Rust, and Koos de Voogt. 2001.

51. Valuing Water in Irrigated Agriculture and Reservoir Fisheries: A multiple UseIrrigation System in Sri Lanka. Mary E. Renwick. 2001.

52. Charging for Irrigation Water: The Issues and Options, with a Case Study from Iran.C. J. Perry. 2001.