D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 THE REAL COST OF RURAL WATER SUPPLYAN INTRODUCTION TO THE WRC RURAL WATER SUPPLY COST AND TARIFF MODEL D.A. STILL and P.C. HOUSTON Partners in Development cc, PO Box 11431, Dorpspruit, 3206 pidpmb@iafrica.com SUMMARY In 1998 the WRC commissioned a project to research the costs of stand alone type rural water supply schemes, and the development of a cost and tariff model for planning and management purposes. This paper describes and demonstrates the model that will be available with the research report from the WRC: In particular, the paper: i)describes the key attributes required for a rural water supply cost and tariff model to be useful to planners and engineers; ii)describes the structure and features of the WRC Rural Water Supply Cost and Tariff model; iii)demonstrates the models utility by showing the results of a comparative case study for two different service levels for a community water supply project; iv)gives guideline figures for design life of system components; and v)gives guidelines figures for water consumption. Themodelhasthepotentialtoprovideacommonframeworkfortheplanning,evaluationand management of all rural water supply schemes.This has hitherto been lacking in South Africa. 1. UNDERSTANDING COSTS IN RURAL WATER SUPPLY: WHY IT MATTERS In1981theUnitedNationslaunchedtheInternationalDrinking WaterSupplyandSanitationDecade.TheaimoftheDecadewastobringaboutsafewaterandsanitationforallthepeopleoftheworld. During the Decade, some 1600 million people were served with safe water.However by the end of the Decade it was estimated that there were still over one billion people without safe water (WHO, 1997).The continent with the highest percentage of unserved people was Africa, with 43%. Now,twentyyearsafterthedecadebegan,itisestimatedthat,worldwide,1.1billionpeoplearestill without access to potable water (WHO, 2000). Some 38% of Africas population are still without access to safe water, and due to population growth the size of the backlog has actually increased.In addition a number of areas which are classified as served are known to receive water only intermittently. The coverage figures quoted above do not tell the full story, for it is in rural areas that the percentage of unserved people is highest, and that the progress made is the lowest.In 1990 it was estimated that 56% of Africas rural population were unserved.By 2000, despite major efforts, the figure had reduced by only 3% to 53% (WHO, 2000). Over the last five years, a consensus has emerged on the principles to guide the provision of rural water supply.Internationally, policies call for treating water as an economic and social good managed at the lowest appropriate level.For the provision of water supply this requires that consumers be engaged in theprocessof selecting,financingandoperatingsystemsthat meettheirdemandsandwillingnessto pay. (Cairncross, 1992) Managing water as an economic good has strong implications for the establishment of proper financial arrangements for a project.Financial policies should send out correct signals linking service levels to actual cost, should maximise cost recovery by capturing community willingness to pay, and should make efficient and equitable use of subsidies. The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa, Biennial Conference, Durban, May 2002 2 Incommonwithotherpartsofthedevelopingworld,theruralwatersectorinSouthAfricahasbeen heavilysubsidisedforsometime.Priortothetransitiontouniversalfranchisein1994,government programmesinruralareastendedtofocusonhandpumpsandwindmills.Thesemoresimplewater supply solutions (which due to institutional problems did not necessarily work) were thus associated with the old South Africa and were regarded as third-class.Since 1994 the almost universal drive in the sector has been to get a piped water supply to within reach (classified as 200 metres) of every home.This is the standard practically every community aspires to, and delivery on this standard carries a very highpoliticalpremium.However,itisbeingfoundthatintheabsenceofexternalsupportthese reticulated schemes are, if anything, even more prone to failure than the old handpumps and windmills. The governments response to the problem of the poor sustainability of rural water supply has been to give an undertaking, on behalf of local government, that every family is entitled to 6 kl of water per month free of charge.In rural areas this undertaking is to be financed by local government primarily by using a portionofagrant(theEquitableShare)whichhasbeenmadeavailabletothemforthepurposeof making basic services affordable to the poor. It is expected that the more rural councils in South Africa may require at least three years to implement the free basic water policy, and it is realised that in some areas it may not be possible to meet the promise at the level of service implied by the policy (Department of Water Affairs and Forestry, 2001). In this context,withinternational practice pushing towards increasing independence from government, down to the lowest level possible, and South African practice pushing towards increasing dependence on government,itisparticularlyrelevanttoaskthequestion:howmuchdoesruralwatersupplycost, anyway?How much does it cost nationalgovernment, how much does it cost localgovernment, and howmuchdoesitcosttheactualwaterservicesprovider?Thecostishighlydependentonmany variables, such as the capital cost, the need to pump, or not, the need for ongoing support, or not, and the effectiveness of management.For example, all other things being equal, one scheme which has high transmission losses, low consumption levels and high levels of bad debt, could find that its cost of sales is ten or twenty times that of a scheme which has all those conditions optimised. 2.DEVELOPMENT OF THE WRC COST AND TARIFF MODEL FOR RURAL WATER SUPPLY Inorderforamodeltobeusefulintheruralwatersupplycontext,itneedstobefairlysimpleto understandanduse,andyetitmustbeabletoaccommodatevirtuallyanycombinationofsubsidies, service levels and tariffing options. It must be able to be used either at the planning stage to investigate theeconomicviabilityofascheme,orasatoolduringtheoperationalphasetotestdifferenttariff scenarios. 2.1. Model Features The WRC COST AND TARIFF MODEL FOR RURAL WATER SUPPLY has been written with a number of desired features in mind: I.It must be possible to run the model either at a simple level, or at a more complex level - i.eit must be possible to use the model even if accuracy is compromised, even if only the bare details regarding a scheme are known, or if the user does not have the time to fill in all the information which is known. However, the user that has the detailed information and the time to run the model at a more sophisticated and thorough level should be able to do so. II.The modelmust separatecostsintoalogicalframework,i.e.capitalcosts,assetreplacement costs, overhead costs, production costs, repair and maintenance costs, support and mentorship costs. III.No costs must be hidden.Any subsidies applicable must only be taken into account once all real costshavebeendetermined.(Oneoftheusesofthemodelisforthecomparisonofthe economics of different options - this can obviously not be done if any costs are hidden). IV.All data must be entered only once in the model, to avoid situations where a change in the value of a key variable at one point is not reflected at another point. For example, the numbers of the differenttypesofconnectionsareallspecifiedonmodelpagethreeDemandAssessment. The next time this data is reflected is on page eleven, Tariff Calculation, where it is linked to page three and thus generates automatically. V.ItmustbepossibletomodelanysimultaneouscombinationofLevelsofService.Whilea The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 3 scheme may primarily supply water via community standpipes, for example, it may also include a large number of yard connections.The Mseleni Water Scheme in northern KwaZulu-Natal, for example,hasacombinationofunmeteredpublicstandpipes,meteredyardconnectionsand unmeteredyardconnections.Eachoftheseservicelevelshasitsownwaterdemand characteristic, which needs to be separately reflected. VI.Wherepossiblethemodellayout,structureandterminologyshouldreflecttheDepartmentof WaterAffairsandForestrysOperationandMaintenancereportingsystemforruralwater schemes. During thelast twoyears a fairly comprehensive O&M reporting system has been developedbyDWAFandUmgeniWaterspecificallyforruralwaterschemes.Thisreporting system uses certain terminology and cost codes, all of which have been reflected in the relevant pages of this model. VII.The model must be able to process simultaneously a range of water demand scenarios (from low to high). This is important because the cost per kilolitre is generally very sensitive to demand,because the fixed costs (capital costs, asset replacement, rental charges and salaries) tend to be the dominant factor in the pricing of water supply, particularly rural water supply. VIII.The levels of water loss and bad debt must be explicit. In combination, water losses and bad debt can affect the per kilolitre cost of sales by an order of magnitude.It is thus critical that these aspects of pricing are clearly dealt with at the relevant points in the model. IX.The model must distinguish between total population and population served.It is misleadingto base tariff calculations on the total population in a project area.The model provides data fields bothfortotalpopulation(whichisusedmerelyforbackgroundinformation,andnotin calculations), and for numbers of households actually served at each service level (which are the figures used in later calculations). For example, if the first phase of a water scheme includes onlythemainlinewithanumberofpublicstandpipes,and,say,twothirdsofthepopulation residemorethanakilometrefromthosestandpipes,thenonemightbesupplyingandsellingwater to only one third of the population.The distance people will travel to collect water from a tap and the volume they will collect from the tap are dependent on the alternative supply options available to them (e.g. rainwater tanks and springs). X.Loan and grant finance must be separately specified.Note that, in keeping with the principle of no costs being hidden, the model does not assume that grant finance is free.The user is asked tospecifytheopportunitycostofthegrantfinance(intermsofaninterestrate),andthis opportunitycostislaterallocatedtothesourceofthegrant(typicallynationalorlocal government, or even international government aid funding). XI.The model must reflect the effect of inflation on asset replacement costs. The model has data fields for the cost of the different components of the scheme, as well as the commissioning date of the schemes major components and the applicable inflation rate.The escalated replacement cost of the assets can thus be calculated at anylater date, and this is used incalculating the schemes full real costs for that date. XII.ThemodelmustallowforcostsharingbetweenWaterServicesAuthorities,WaterService Providers, National Government and other stakeholders. While no costs are hidden, the user is primarilyinterestedincalculatingthecostsandtariffsrelevanttotheWaterServiceProvider.The Monthly Cost Summary sheet allows costs to be distributed according towhich body (WSA, WSP, National Government or Other) is to budget for that expense.In this way the final screen, Tariff Calculation, deals only with the costs allocated to the Water Service Provider. XIII.Insettingtariffs,usersmustbeabletouseacombinationoffixedchargesandvolumetric charges (with rising blocks, if required) for any level of service. The user must be able to specify different tariffs and charges for each different level of service. XIV.The model must reflect costs in terms of both volume of water supplied (R/kl) and in terms of households served (R/hh).The volume of water supplied here must exclude losses.The cost per kilolitre is important indicator of the effectiveness of a water supply scheme.For example, if oneschemesuppliesonaveragefivekilolitresperfamilypermonthatR4/kl,foratotalof R20/family/month, it is far more cost-effective than another scheme which suppliestwo kilolitres per family per month at a cost of R10/kl (which would also have a monthly cost per household of R20).The monthlycostperhouseholdservedisanimportantindicatorofaffordability.Itis The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 4 generally accepted that 3 to 5% of monthly income is the amount people are prepared to spend on a reliable and adequate water supply service. TheWRCCostandTariffModelforRuralWaterSupplySchemesdoesnotcalculatecashflow.It determines the costs and revenue (and thus profit or loss) for a particular status of the scheme at a point in time.However, the model does contain three scenario options (low, medium and high consumption) which can be used to simulate changes in levels of service and consumption, and thus indirectly cash flow,overaperiodoftime.Notethatthislattermethodofdeterminingcashflowisbasedonthe assumption that no extra capital costs are incurred- i.e. customers pay for their own connections, and the scheme already has the capacity to meet the higher demand. If future scenarios require additional capitalinvestment,thenthecashflowmustbedeterminedindirectlybyrunningadditionalsimulations which include the revised capital costs. The model can be used to calculate pumping costs, and allows for a relatively sophisticated analysis in this regard.Up to five separate pumpstations can be included, and the user can specify what portion of the total demand is pumped at each station.The user specifies the current Eskom block tariffs, and the modeldifferentiatesbetweenenergycostsbelowandabovethestepinthetariff. Theusercanalso specify different pumping efficiencies at each stage, and whether the pump is powered by Eskom or a diesel engine. 2.2Model layout Figure1 shows the outline of the ten model spreadsheets. Three colours are used to denote input and output fields in the spreadsheets. Blocks coloured dark blue denote essential information, which must be entered by the user ifthemodelistoproduceanyoutput.Blockscoloured lightbluedenoteusefulinformationwhichcan nevertheless be omitted at theusers discretion. Yellow blocksdenotevalueswhicharegeneratedthoughthe working of the model. Allcellswithformulaeandtextarewriteprotected,to protect theuser from inadvertently corrupting the model.However, this write protection can be bypassed by simply saving the model under a new name. 3. MODEL DEMONSTRATION The analysis of a fairly typical project is discussed below. 3.1 Anal ysis of Piped Water Proj ect serving 5 400 The projectareahas670homesteads,withanestimated population of 5378 (8 persons per home).Aproduction borehole has been drilled, and this has sufficient water to supply the estimated future population of 8600 persons. Itisplannedtosupplythecommunitythrough20public standpipes,eachserving10homes,aswellasthrough 470yardtaps.Allowanceisalsomadefor12schools and 5 shops. Transmission losses of 40% are used in this analysis. Anal ysi s of costs The project capital cost is R8 031 089, or R1 500 per person served. Of this amount it is estimated that R295987isrequiredformechanicalandelectricalworks.A100%grantisavailablefromDWAFto construct this scheme, and thus no loan costs will be incurred.However the opportunity cost of the grant finance, at 5%, is R33 463 per month. The models default life expectancies are used for the various components, and the resulting expected assetreplacement budget is R24 331 per month. Asset replacement costMonthly cost summaryTariff calculationProject summaryOverhead costsProduction costsRepair and maintenance costsSupport and mentorship costsDemand assessmentCapital costThe Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 5 TheestimatedmonthlyoverheadcostsareR6777,mostofwhichiscomprisedofstaffcosts.The expected production costs are between R844 and R2 499 per month (according to demand).The budget forrepairsandmaintenanceisR1300permonth,mostofwhichisforlocallabour.Thelongterm support and mentorship budget is R8 000 per month. Threewaterdemandscenariosarebuiltintothemodel:low,mediumandhigh.Inthiscasethe respective consumptions analysed are 7 l/c/d, 15 l/c/d and 30 l/c/d.The projected total cost of the water supply in this case comes to between R74 716 and R76 371 per month, depending on demand (note how small the difference in total costs is between these scenarios - most of the costs are not variable but fixed).The total cost per kilolitre sold varies from R54.10 to R15.78, depending on the demand scenario. Of the above costs, R33 463 (the opportunity cost of the grant finance) is borne by national government (byvirtueofthegrantfinancingofthecapitalcosts).R30131,forsupportandmentorshipandthe replacementcostsofthecivilworks,isallocatedtotheWaterServiceAuthority.Theremainder,or between R11 121 and R12 776 is left to the Water Service Providers account. Analysed in terms of costs per unit of sales and per household per month the above costs translate as follows: Table 1 Modeled costs for Piped Water ProjectServing 5400 ScenarioDefinition of cost Total CostsWater Service Provider Water Services Authority National Govt. low demand7 l/c/d R/kl of sales54.108.0521.8224.23 R/hh/month111.5216.6044.9749.94 medium demand 15 l/c/d R/kl of sales29.604.5911.8513.16 R/hh/month112.3517.4344.9749.94 high demand 30 l/c/d R/kl of sales15.782.646.236.92 R/hh/month113.9919.0744.9749.94 With an average household income in the area of approximately R800 per month, the affordability ceiling would typically be judged at between R24 and R40/month (i.e. between 3% and 5% of monthly income).The costs per household (at WSP level) under all three demand scenarios fall beneath this ceiling, but in realitynot all householdswillbe contributing.The cost per kilolitre for the medium and high demand scenarios is fair, but the cost per kilolitre in the low demand scenario, at R8.05 per kilolitre, is high.High costsmaydiscouragepeoplefromusingthesystem,whichthensetsupanegativecycleoflower consumptionandevenhigherunitcosts.Neverthelessthesystemisplannedtosupplywatermainly throughyardtaps,andthemediumconsumptionlevelof15litresperpersonperdayisprobably reasonable. On the tariff calculation sheet, a tariff of R4 per kilolitre sold is tested, combined with a R15 per month basic charge.Users of public standpipes pay R6 per kilolitre.Bad debt of 30% from private consumers and 60% from users of public standpipes is assumed.The resulting projections are as follows: low demand:projected deficit of R2 238 per month medium demand:projected surplus of R361 per month high demandprojected surplus of R5 654 per month Ifitisassumedthatthemediumdemandscenarioisthemostprobable,thentheschemeshouldbe sustainable. The free basic needs water policy will have to be taken into account, however.With the medium demand scenario, the average water usage is only 3.6 kl/month per household. If the definition of basic needs is takentobe6kl/family/month,thentherewillbenoincomefromthisscheme,regardlessofthetariff The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTON Water Institute of Southern Africa, Biennial Conference, Durban, May 2002 6 structure. In such a case there are three options: the Water Service Authority takes over responsibility for all costs (but the WSA share of the costs is already R45 per family per month, which may be more than it can afford unless the equitable share grant is increased); or the free basic water policy is applied only to the public standpipes, from which the income is in any case very small; or the scheme should not be built as planned. If the latter option is chosen, an alternative to supply the community with water is a combination of public handpumps and protected springs.The economics of this option are discussed below. 3.2Anal ysi s of a Handpump Al ternative Assume that 45 handpumps and 10 protected springs are planned to serve the same community - this would provide a coverage of one water point to every 12 homes. The demand figures used are 5 litres per person per day for the low demand scenario, 10 l/p/d for the medium demand and 20 l/p/d for the high demand scenario. Anal ysi s of costs The project capital cost is R2 065 000, or R385 per person served. Of this amount it is estimated that R360 000isrequiredformechanicalworks.Itisassumedthata100%grantwillbeavailablefrom DWAF to construct this scheme, and thus noloan costs will beincurred.The opportunity cost of the grant finance, at 5%, is R9 179 per month. Allowance is made to replace pumps after 10 years, and boreholes after 30 years.The projected asset replacement cost is R6 942 per month.The estimated monthly overhead cost is R710, which allows for a part-timebook-keeper/administratorandalocalrepairtechnician.Thereisnoproductioncost.The budget for repairs and maintenance is R825 per month, which is evenly split between materials and local labour. The long term support and mentorship budget is R3 900 per month. The projected total cost of the water supply is thus R21 556/month.The total cost per kilolitre varies from R24.60 to R6.15, depending on demand scenario.The medium demand scenario would be realistic with a high coverage of handpumps like this - for this the cost per kilolitre is R12.30 per kilolitre. Of the above costs, R9 179 (the opportunity cost of the grant finance) is borne by national government,R10 842 per month is allocated to the Water Service Authority, and R1 535 is allocated to the community Water Service Provider. The above costs translate as follows: The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 7 Table 2 Modeled costs for Handpump Proj ect ScenarioDefinition of cost Total CostsWater Service Provider Water Services Authority National Govt. low demand 5 l/c/d R/kl24.601.7512.3810.48 R/hh/month29.942.1315.0612.75 medium demand 10 l/c/d R/kl12.300.886.195.24 R/hh/month29.942.1315.0612.75 high demand 20 l/c/d R/kl6.150.443.092.62 R/hh/month29.942.1315.0612.75 In terms of the Free Basic Water policy the Water Services Authority would probably pick up the costs of the community water services provider. The comparison between the two case studies is shown graphically in Figures 2 and 3 below. In these figures all costs are reflected (i.e. including the cost of capital, asset replacement and external support).The costs are shown both in terms of rands per kilolitre sold, and in terms of rands per household per month. Figure2: A Comparison of costs in rands per month per household served for a typical community in KwaZulu-Natal, with 670 families and 8 persons per home.Two scenarios are compared, one with full reticulation, the other with one handpump or spring protection to every 12homes. 020406080100120Pumped &reticulated - 7l/c/dPumped &reticulated -15l/c/dPumped &reticulated - 30l/c/dHandpumps -5 l/c/dHandpumps -10 l/c/dHandpumps -20 l/c/dRands/householdNationalWSAWSPThe Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 8 Figure 3:This shows the same comparison as Figure 2, but excludes the capital costs, external support costs and most asset replacement costs.In other words, this is the expected monthly operating cost, per household, at Water Service Provider level. Figure 4: This shows the same comparison as Figure 2, but in terms of rands per kilolitre sold. 0102030405060Pumped &reticulated - 7 l/c/dPumped &reticulated -15 l/c/dPumped &reticulated -30 l/c/dHandpumps - 5l/c/dHandpumps - 10l/c/dHandpumps - 20l/c/dRands/klNationalWSAWSP0510152025Pumped &reticulated - 7 l/c/dPumped &reticulated -15l/c/dPumped &reticulated - 30l/c/dHandpumps - 5l/c/dHandpumps - 10l/c/dHandpumps - 20l/c/dRands/householdThe Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 9 Figure 5:As for Figure 4, but excluding asset replacement and mentorship costs. 4. ASSET REPLACEMENT: WHO IS PROVIDING FOR THIS? Ifoneisinterestedinthelongtermcostofwatersupplyschemes,onehastoaskhowlongthe infrastructure will last.The life expectancy of an asset is the period during which the asset will continue to render the service for which it is intended.An item of infrastructure, in particular, must be replaced either when it is no longer large enough to perform its intended function (e.g. with growth in population and water demand, a pipeline which was once adequate may become too small), or when it is no longer serviceable(e.g.whenanenginewearsout,orwhenasteelpipecorrodes).Thetypicallife expectancies upon which traditional water tariffing analyses are based, are shown in Table 3 below. It is acceptable to use relatively long life expectancies for items such as reservoirs and pipelines, provided they are made of durable materials and that little change in water demand is expected.Mechanical and electrical items, however, should not be expected to last longer than ten years. In the model analyses, the asset replacement cost of fast wearing items (e.g pumps) is typically put down to the Water Service Providersaccount.However,thereplacementcostofpipelinesandreservoirs,whicharemajorcost items, are set down against the Water Service Authoritys account.In the short term the costs of asset replacement are not seen, and are often forgotten.In the short and medium term they are in fact not costs but recommended savings, to be invested at interest until the time when replacement is needed.It is a concern that local government in South Africas rural areas does not have the financial strength to set aside such savings.However in 40 or 50 years time, someone will have to pay to rebuild the water infrastructure. Who will that be? 0123456789Pumped &reticulated - 7l/c/dPumped &reticulated -15l/c/dPumped &reticulated -30l/c/dHandpumps - 5l/c/dHandpumps -10 l/c/dHandpumps -20 l/c/dRands/klThe Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 10 Table 3:Life expectancy of various scheme components(Umgeni Water, 1997) DescriptionUseful Life Expectancy (Years) Mechanical/electrical costs Pump installation10 Electric controls 10 Other10 Civil Works Borehole15 Dam50 Pumphouse40 Water Treatment Plant (excludes elec/mech)20 Bulk Main40 Primary Reservoirs40 Secondary Reservoirs20 Primary Reticulation20 Secondary Reticulation20 Bulk Meters15 Consumer Meters5 Public Tapstands20 Yard Tanks10 Yard Taps10 Connections20 Buildings30 NOTE:Itisrecommendedthattheeconomiclifeofreservoirsandreticulationinthetableabovebe extendedto40yearsifpopulationgrowthislowandtheconstructionmaterialsusedarenotproneto degradation. The design life of 15 years for a borehole can also be extended provided a non-corroding well casing and screen is used. 5.GUIDELINE FIGURES FOR WATER CONSUMPTION Expected water consumption is a critical figure in any cost and tariffing calculation.While in hydraulic design this figure should never be underestimated, in a financial analysis it should not be overestimated.Many schemes in thepast havebeen fundedbasedonwholly unrealistic sales projections. As more schemes have been built and data on actual sales is recorded, a measure of reality is being brought to bear on plans for new projects. One of the most critical determinands of cost per kilolitre is the level of consumption.Table 4 shows the default figures included in the model as a guide. The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 11 Table 4.Default consumption values, litres per person per day Service LevelLow DemandMedium Demand High Demand Public Handpump/well51020 Private handpump/well203040 Metered public standpipe2712 Unmetered public standpipe62136 Metered Yard Tank71530 Unmetered Yard Tank204060 Metered Yard Tap71530 Unmetered Yard Tap204060 Metered House Connection3060120 Unmetered House Connection60120240 6.CONCLUSIONS AND RECOMMENDATIONS There is a need for a much clearer understanding in South Africa regarding what rural water supply really costs.Without such an understanding, the sustainability of much of the work which is now taking place must be called into question.Unfortunately the understanding of costs is hampered by the number ofassumptions that must be made in the course of analysis.The value of the WRC Rural Water Supply Cost and Tariff Model is that it can provide a standard approach, and that it prompts the user to include costswhichmightotherwisebeoverlooked.Itisrecommendedthatthismodelisadoptedforthe calculation of costs and tariffs for rural water supply projects in South Africa, whether for feasibility and planning purposes prior to construction, or for the setting of tariffs during operation. ACKNOWLEDGEMENTS The authors wish to thank the Water Research Commission for funding the research and development of this model, and the Mvula Trust and the University of Durban Westville for their contributions. REFERENCES

BASSELINK, J . (1992)TechnicalandManagementaspectsofsmalldrinkingwatersupplysystemsfor deep wells.InterAction Design.The Netherlands. CAIRNROSSS.(1992)SanitationandWaterSupply:PracticalLessonsfromtheDecade:Waterand Sanitation.Discussion Paper Series No. 9. Water and Sanitation Programme.World Bank. DEPARTMENTOFWATERAFFAIRSANDFORESTRY.(1998).NewApproachtoWaterPricingin South Africa. DEPARTMENT OF WATER AFFAIRS AND FORESTRY. (1998)Study to determine a cost effective and sustainable method to provide rural communities with yard or house connections. DEPARTMENT OF WATER AFFAIRS AND FORESTRY. (2000) Water Supply Service Levels, A Guide for Local Authorities.DEPARTMENTOFWATERAFFAIRSANDFORESTRY(2001).FreeBasicWater,ASummaryfor Local Authorities. HAZELTOND.ANDKONDLOS.(1998)CostRecoveryforWaterSchemestoDevelopingUrban Communities.Water Research Commission Report No. 521/1/98. LIMARURALDEVELOPMENTFOUNDATION.(2000)AssessmentoftheAttendedCoupon-operated Access-pointCostRecoverySystemforCommunityWaterSupplySchemes.WaterResearch Commission Report.MIDDLETONR.N..(1998)WorldofWater:BeyondtheInternationalDrinkingWaterSupplyand Sanitation Decade. Lewis Publishers. The Real Cost of Rural Water Supply An Introduction to the WRC Rural Supply Cost and Tariff Model D.A. STILL and P.C. HOUSTONWater Institute of Southern Africa,Biennial Conference, Durban, May 2002 12 MVULA TRUST. (1998). Help Manual for Rural Water Credit. PALMERDEVELOPMENTGROUP.(1998).WaterSupplyServicesModelManual.WaterResearch Commission Report No. KV109/98. REPUBLIC OF SOUTH AFRICA. (1997) Water Services Act.Act No. 108, 1997.Office of the President. UMGENI WATER. (1997) Planning and Design Guidelines for Community Water Supply and Sanitation. Version 1a.WATER RESEARCH COMMISSION. (1998) 12 Successful cost recovery case studies for water services in South Africa. WATER SERVICES MANAGEMENT. (1996).GuidelinesfortheestimationofDomesticwaterdemand of developing communities in the Northern Transvaal.Report to the Water Research Commission. WORLDHEALTHORGANISATION.(2000).GlobalWaterandSanitationAssessment2000Report.www.who.int/water_sanitation_health/Globalassesment/Global2.htm