31I CI L 1' HE 2iSI ~ ~ I.E~ EW The World Bank Sector Policy and Research Infrastructure and Urban Development Department Report INU-OR 13 FY92 Sector Review Water Supply and Sanitaton DScr ember 1992 CYneral Opetional Review This document is published infomially by the World Bank- The vievvs and interpretations herein are those of the author and should not be attnbuted to the World Bank to its affiliated organizadons, or to any individual acting on their behaff. Copyright 1993 The World Bank 18 1 8 H Street NW Washington, DC 20433 USA Al Rights Reserved First Printing February 1993 The principal authors are Harvey A Gamn and Guillermo Yepes of the Infrastructure and Urban Developmnent Department ofthe Wortld Bank. Karen Kemperprovided valuable inputforthe special topicdiscussion of personnel costs. This document is published informaly by the Wolid Bank. Copies are available free from the World Bank at the address above. Contact Mrs. Mari Dhokai, room S-1 2-005, telephone 202 473-3970. The World Bankdoes notacceptresponsibilityforthe viewsexpressed herein, whichare theose of theauthorsand should not be attributed to the world Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bankc they do not necessarily represent official policy of the Bank. The designations employedand the presentation of the materialare solelyforthe convenience of the reader and do not imply the expression of any legal opinion whatsoever on the part of the World Bank or its affiliates conceming legal status of any country, temtory, city, area, or of its authorities, or concerning the delimitaitons of Its boundaries or nabonal affiliation. The World Bank Sector Policy and Research hfrastructuim and Urban Development Department FY92 SECOR REVIEW WATER SuPPY AND SAN1TATON December 1992 CNTETS Executive Suinmaty and Recommendations 1. The Sector in the Bank in FY92 1 Lending Operations 1 The Project Pipeline 2 Sector Work: Studies and Country Strategies 2 Staff Allocations 3 UNDP-World Bank Program 3 IH. Spedal Topics, FY92 5 Central Messages 5 The State of the Sector 5 Why Has Project Performances ween Poor? 7 How the Bank Will Move Forward 8 ANNEXES Annex I Figures and Tables 11 Annex II ARIS Data on the Water and Sanitation Sector 19 Annex III Scope and Methodology of the OED Report 21 Annex IV Specific Responses to OED Recommendations 23 FIGURES Figure 1 Bank/IDA Operations 11 Figure 2 LendingOperations 11 Figure 3 Allocation of Staff Resources 12 Figure4 Personnel Costs/Operating, Costs 12 Figure 5 Personnel Costs/Cubic Meter and Staff/1,00 Connections 13 Figure 6 Percent Real Increase in Personnel Costs 13 TABLES Table 1 Loans and Credits 14 Table 2 FY93-97 Pipeline 15 Table 3 Loans and Credits by Project Components 16 Table 4 F-imples of Operational Support Provided by the LJNDP-World Bank Water & Sanitation Program, FY92 17 Table 5 The Performance of the Water and Sanitation Sector Relative to Other Sectors 18 EXECUTIVE SUMvIARY AND RECOMMENDATIONS Sector activity (lending, supervision, sector goodandvaluing(pricing)itatitsopportunitycost work,andpolicyandresearchwork)continuedinFY92 in all uses. at about the same levels as in previous years. Sector lending was about$1 billion in Il projects. Lending for Providingservicesthatpeople wantand are willing watersupplyimprovements,asinpastyears,predorni- to pay for-i.e., tobe more responsive to economic nated relative tosanitation, drainage, and environmen- demand. tal components. sw Increasing investments in sanitation based on a Lendingprospectsfortheupcomingfiveyears demand-driven strategy. showasignificantupward trend,currentlyprojectedat about $15 billion per year. However, achieving in- '3 Rethinking institutional arrangements to press for creases of this magnitude poses significant challenges improved performance in public utilities, separat- to the Bank to step up project preparation efforts, ing provision of services from regulation, expand- identifyandsecureadequatecofinancing andincrease ing the role of the private sector, providing for the absorptive capacity of borrowers. As demon- consumerparticipation,andfocusinggovenmuents strated by OED in its recent evaluation of past Bank on what only they can do to define and enforce projects in thesector, the performanceof thesector has appropriate legal, regulatory, and administrative been poor. The major medium-term challenge will frameworks. therefore be to achieve significant sector and policy reform and improve project quality. We recommend, therefore, that the Bank pur- sue a sector improvement strategy based on two key Sector work by the Regions, and policy and points. First, it isessential totake abroader view of the research workbylNUJWSincollaboradonwithOpera- determmnants of performnance of sector orgarnzations tionsand other multilateral and bilateral organizations than has tended to prevail. Particular attention should in the sector, have confirmed the urgent need for sector begivennotordytoissuesunderthecontrolofaspecific reform to provide additional incentives for conserva- supply organization but also to the environment in tion of water resources and increase efficiency in ser- whichtheorganizationfunctions,specifically(1)regu- vice delivery. This work is sununarized in World latoryandadministrativecontrols,(2)allocationproce- Development Report 1992, Chapter5, "Sanitation and dures and practices for public expenditures, and (3) Clean Water" and in the Management Response to water resource management that treats water as an OED's study, "WaterSupply and Sanitation Projects- econornicgood,withmanagementvestedatthelowest The Bank's Experience 1967-1989." This Annual Re- appropriate level. port demonstrates, in its special topic discussion, that thekeyinstitutioWial challengeistocreateincentivesfor Second, the Bank mustacknowledge, through all stakeholderstomaximize thecongruenceofinterest itsactions, thatmore needs tobelearned-and learned arnongstakeholdersandtoperforminanefficientway. rapidly-in imnplementing approaches to sector re- In general, this implies paying more attention to creatt- form. Specifically, the Bank should develop a more ing "market-friendly" arrangements in the sector. consistent and structured way to learn from ongoing sectoractivities rather than waitforretrospectivestud- In particular, four central policy changes are ies with large time lags. This structured learning needed: process is intended to produce indicators that will enable us to say whether sector performance is ad- w Managingwaterresourcesbetterthroughtheuseof equateincomparisontootherinternationalexpenience, moreeffectivemarket-basedmethodsofallocation whether it is improving, and what are the specific and use, including treating water as an economic bottleneckstoimprovedperformance. Effortstoirnple- menttheseideasarealreadyunderwayinjointunder- refor, and produce papers investigating key themes takin between Operations and INUWS. A new, such as stinulating private-sector involvement. To related venture, the "Water Utilities Partnership" has prepare staff to gain the skills needed for this new justbeeniWtiatedtodocumentexperiences withinno- orientation, a new training plan for water and sanita- vative formal sector approaches, produce practical tonsectorstaffisinpreparation.Oneinnovativeaspect guides of best practices on reform issues, conduct of newproposalsis for jointtrainingof Bank staff with workshopsindudingbothpolicymakersandpractitio- key decision makers and sector professionals in bor- ners in specific borrowing countries with interest in rowingcounties. I. THSECS R N TE BANK IN FY92 'he Wod Delopant Rqvrt 1992: Devcop- assistance, and engireering for an additiomal 12 per- mentandtheEnironmentstates, "forthelbiBionpeople cent, and the rest (14 percent) went for miscelaneous indevelopingcontdieswhodonnthaveaccesstodean equipmentand landacquisition(Table2). Seven of the water and the 17 billion who do not have access to tenprocsaredirected tourbanareas. Oneprojectin sanitation, these are the most important challenges of Lesotho, is unique in the sense ithas been conceived to all." The Bank's activities in the water and sanitation allow the country to export water, an abundant con- sectorinFY92-endling,supervision,sectorwork,and modity, for hard curren. policy and research work-reflect this conrcen and suppM te creation of conditions necessary for sus- Inadditionto the water supplyand sanitation tamable development of the sector. These efforts (WSS) projects, six of the 15 urban projects contain emphasize the need for sector eform that includes substial water supply and sanitation components adequate cost recovery, expansion of service to the thatrangefrom7percentto99percentof talproject poo, preservation of water resources, and protection costs?2 A few projects in te agriculue and health of the envlronment. The effort and time required to sectrs contain water and sanitation comnponents. The achieve a sustainable sector program in all countries costfiguresof thewaterandsanitationcomponentsin will determine the outcome for thebillions of persons urban, agriculturl orhealth projects are notinduded who do not have ad,quate services today or who in the sector figures cited earlier. reeve them in very precarious conditionse ThenumbeofprojectssubmtidtodoieBoard LENDING OnPUMONs varies from year to year and over the last fifteen years seemstofollowafive-yearcyde. Themostrecentcycle, Expectations of a year ago were for 15 direct FYs 88-92, shows the lowest number of projects pre- operations in the water and sanitation sector totaling sented (48 prjects) in the last 15 years (72 inFYs 78-82 $1.5 billion in loans and credits, to be presented to the and 77 in FYs 83-87). In terns of annual lending BoardinFY92.However,ordylloperations(fiveloans volumes and the average size of the loan/credit, te andsixcredits)' fortotallendingof$9114Amiionwere cydicalpattemislessclearbutstlldiscernible(Figures approvedinFY92(Tablel). Comparedtotheprevious I and 2). This cyclical pattern pointst the difficulties yearthenumberofoperationsinaeasedbytwo;butthe ofmaintaningasteadypipeline;due,inpart toHimita- volumneoflendingdeceasedby$214million(Figurel). tions in the absorptive capacity of sector institutions Total project cost however, inceased by $1.89 billion and their ability for long term plaing. In addition, orby8Operentoverthepreviousyear;buttheBank's financial constraints of sectr institutions, often a re- share of total project cost decreased to22 perent fron flectionof inadequatecostrecoverypoiciesandopera- 58 percent in FY91. tional inefficiencies. Theseconstaintsoften dictate the pace of project execution and therefore of the pipeline. Cofinaning was important in FY92. It pro- videdaboutl9percentoftotalprojectcost,asignificant Over the years, Bank lending for water and increase over the 6 percent share of the previous year. sanitation averaged 4.8 percent of total lending, peak- External financing in FY92, however, was atypical asit ing in 1979 at 9.2 percent (21 projects) and reaching a included $759.2 million in external aid provided to the record low point in 1988 at 2.8 percent (5 projects). In Lesotho Highland Water Project that represents 97 FY92lendingtothesectorrepresented4.2peroentofthe percent of all cofinancing in the sector. totll Of the apvtoved projects, five focus exclu- sively on water supply, two on sanitation, one on drainage, and one on protction of the urban environ- ment Two have a blend of water and sanitation components. Water supply components continue to L. One operation, China-Beijing Environment Protecdon, predominate and represent 54 percent of total base has both IBRD and IDA funds. cost Sanitation,drminage,anenvironmentalconpo- 2. Angola,99%;China,50%;Venezuela,49%;Malawi, 14%; nents account for 20 percent of base costs, techdncal Benin, 11%; and Uganda, 7%. 1 TBE PRoJEcr PIEuNE work is also of particular importance in countries where the Bank is still not fanmliar with the sectoral Lending prospects in the sector show an as- issues or where a significant increase in the volume of cending trend for FYs 93-97 with a tentative lending lendingisanticipated. Sector work initiated in FY92 or program of 91 projects and a total lending volume of tostartinFY93isanticipatingtheseconcernsinmostof $7.97 billion (Table 3). This would represent a consid- the Easter European countries and in countries like erable increase both in terms of total lending and Turkey and Colombia. number of projects, even if, as can be expected, some projectsareultimatelydropped. Thecyclicalpatternis During FY92, specific water and sanitation also in evidence in the five-year lending program (Fig. sector studies were carried out for Costa Rica (Water 1), but here it also reflects the uncertainties associated and Sanitation Sector), Egypt (Water and Wastewater with the identificationi of projects in the outer years of Sector), nJ Thailand (Urban Sewerage and Wastewa- the planning horizon. To increase its lending in the ter Manaemnent). In China, a sector report was pre- sector, the Bank will need stepped-up efforts to initiate pared on the production of cast iron and ductile iron project preparation dialogue at an earlier stage. pipes and fittings and the problems created by lack of manufacturingstandards. Water and sanitation issues T.efive-yearpipelineposesseveralchallenges were also discussed in the context of broader sector to the Bank To realize the substantial increase in studies prepared for Nicaragua (Social Sector, Issues sectoral lending, active cooperation and a meeting of and Recommendations), Guyana (Infrastructure Sec- the minds with other donors will be required. This tor Review), and the Sahel (Construction Practices of includes the identification and securing of adequate Social Infrastructure in the Sahel). The South Asia cofinancing. Moreover, for the increased lending tobe Region prepared country profiles on water resources effective, there has to be a significant increase in the for the main countries in the region (India, Pakistan, absorptivecapacityofborrowers Theneedtomaintain Bangladesh, and Nepal). The LAC region commis- project quality while supporting sector reform adds to sioned a background paper on water resources in the thechalletigeofdevelopingastrongpipeline,atleastin region as a first step towards the development of more the medium term. OED has identified poor project detailed and issues-oriented country or subregional preparation and institutional weakness as major com- profiles,alongthelinesdevelopedforSouthAsia. LAC mon causes for poor project performance. This under- also prepared a comprehensive paper on the costs of lines the need to focus on early project preparation. inadequate maintenance on infrastructure (Power, Water and Transport). SECrOR WoRK. STunDS AND COUNM S1RAT6OES Main themes have emerged in these sector studies that are crucial to orderly, rapid development Sector work has become more focused, and of the sector policy recommendations more attuned to macro-eco- nomic policies and linkages with other sectors. Sector or The urgent need for sector reform coordinated workhasalsohelpeddefinewell-focusedstrategiesfor with other reform efforts to improve the produc- sector development in most countries. As a conse tivity of the public sector (Costa Rica, Egypt, Thai- quence, operations can now be tailored to respond to land). These reforms also call for renewed effort to needs and priorities identified in the course of sector elicit the partidpation of the private sector, greater work. Discussions with and reconmendations to accountabilityof sector institutions, and more voice governments are directed to help elicit the will to for users in the decision-making process. remove major sector constraints. The Bank has been instrumental in bringing together planning and finan- Emphasis on water-usepolicles thatpromote con- cial ministries as weU as sector institutions in these servation and better use of resources (South Asia, discussions. LAC). Most of these studies also stress the need to protect water resources from growing polution The projected increase in lending to the sector from urban areas and from industrial and agricul- highlights the importance of comprehensive sector tural activities (Egypt, Thailand). work and the need to keep our sector knowledge current. Theeffortonsectorworkreflectsthisconcern: o The need to increase financial resources to meet 225 staffweeks (sw) were allocated to sector work in rapid demand grwth for water and sanitation FY92, and 324 sw are being planned for FY93. Sector services. The provision of these resources need not 2 impose a heavy strain on national budgets if a Development Progarnme (UNDP) and other bilateri meaningful portion cornes from water and sanita- donors. The Program's strategy for the 1990s states: tion users. The need for sound cost-recovery poli- clescomestotheforefrontwhilecontinuingtostress In the coming years, the Program wiU focus on access to these services by the poor. improving the process through which large- scale sector development initiatives are forrnu- STAff AUOCAMIONS lated and implemented-helping tobuild a sys- tematic learning process into water and sanita- In FY92 the Regions allocated 3,825 sw to tion programs, including those supported by sector activities, a 20-percent increase over the previ- the World Bank and other multilaterals, the ousfiscal year (Figure4). Tis allocation is thehighest bilaterals,andotheragenciesactivein thesector. amounteverintheaggregateand forbothlendingand supervision activities. The Program with its country and regional activities, is well equipped to provide operational Sector work activities registered a slight de- support to Bank-financed projects. It also has the ad- crease of 3 percent over the previous year. vantage ofa decentralized structure and field presence to respond quickly and flexibly to changing circum- Staff intensityinlendi,ngperprojectapproved stances. For example, in response to droughtand civil by the Board (192 sw/project) was slightly lower than emergencies in Ethiopia, Uganda, and Zimbabwe, the in thepreviousyear (195 sw/project) but still the third Regional Officein EastAfrica wasable to providestaff highest in the last ten years. expertise to support Bank Headquarters teams in pre- paring emergency lending operations. Totalsupervisiontimewasthehighestever,24 percent over the previous year. The intensity of super- ProgramparticipationinBankprojectsisprov- vision also increased from 11.7 to 13.9 sw/project, ing to be mutually beneficial. While the relationship again the highest in the past ten years (Annex II). helps the Program spotlight the needs of the poor and influenceinvestments, it helps the Bankin thedevelop- Thisburstofactivityinthesectorisawelcome ment and implementation of large poverty-focused occurrencethatneedstobesustainednotonlytobeable projects. Eighteen rural water supply and sanitation to deliver the proposed and ambitious lending pro- projects are in the pipeline for the next five years. gruambutalsotofosterthereformsnecessarytoprovide greater access to these services and for better and more Eachyear,attheRegionalManagers' Meetings efflcientop-* tionsthattranslateintoafiordableprices. heldinWashington,ProgramfieldmanagersandSODs meet to discuss and agree upon the assistance needed Tin UNDM-WoRW BAN PROCRM in operational support for Bank projects. This time is then recorded in the SOD work programs. However, TheUNDP-WorldBankWaterandSanitation the Program is usually called upon to provide addi- Program (the Progran) continues to be very active in tional support during the fiscal year. In FY92, for coordinating with the sector technical and operating example, the actual operational support provided, 58 divisions and with extemal agencies. The Program sw, was twice the planned amount Table 1 shows assists governments in strengthening institutional ca- examples o. sorne projects (out of a total of 26) where pacity to promote and coordinate the development of the Program provided operational support during water supply, sanitation, and waste management for FY92. The Program's planned contribution in opera- low-income communities. It is managed by INUWS tional support for FY93 is more than four years, well through major projects financed by the United Nations above the amount required to meet commitments. 3 II. SPEaAL Topics, FY92 DuringFY92, in addition to thesector activity there, the measures to be taken to improve discussed above, there were two events of major inw- perfonmance. OED's solution centers around con- portancetothesector. Oneofthesewasthepubllcation ditlonality earlier and with :nore backbone. We of the OED report '"Water Supply and Sanitation believe that the fundamental requirement is a re- Projects-The Bank's Experience 1%7-1989' and the structuring of sector institutions and the environ- submission of a management response to the report ment in which they operate so that suppliers have requested by the Joint Audit Commnittee of the Board. incentives to meet the demands of consumers It is Main elements of the response are repeated here to oniysuchrestructwingthattheimpositionofcov- encourage broad dissendnation of OED's views and enantsholdsanypromiseforsthnulatingsustaned the sectra managemnent response to them. The other imnprovement in performance. mnajoreventwastheindusionofamajorchapteronthe sector in the World Development Report 1992. Its e Third, for the last several years 51P, the TDs, and discussion antidpated nany of the c3ncerns of OED, theoperatingdivisionsof theBankhavevigorously anditsreconunedationsregardinghowthesectorcan addressed the problems underlying poor perfor- moveforwardhavebeenveryfavorablyreceivedinthe mance, using all the tools available-operations, Bankandelsewhere. Theyareals( luded heresince policy and research work, and training. they form the core of a positive t - ponse to the prob- lems in the sector. e Fourth, while the direction forward is dear, thereis much that needs to be learned about the restructur- Sector management and staff welcomed the ing process. opportunity to respond to OED's report on 22 years of Bank lending in water supply and sanitation. OED Methodological problems with the OED re- rightly showed that this is a very important sector. port have an important bearing of the credibility of Moreover, the 1992 WDR highlights the dual role of someofthecondusionsofthereporL Inthefirstround water as an essential human need and an economic of review of the document by Bank staff, for instance, good and, in turn, the role that the sector plays in questions were raised about a large number of the facilitating economic growth, alleviating poverty, and specific conclusions of the stdy. Because of the meth- improving environmental quality. odology of the study it was impossible to detmine how the conclusions were anived at, and whether, TheOEDreportwasthe mostcomprehensive indeed, more nuanced or different conclusions might review of completed Bank projects in the water and have been more appropriate. An unfortunate conse- sanitation sector yet undertaken. The report called quenceisthattheremaybeanindination,ononehand, attentiontomanyimportantissuesthatdeserveserious to dismiss the significance of the report too readily or, attention. The study reviewed 120 urban water and on the other hand, to take all of its spedfic negative sanitation projects started by 1967 and completed by statements as well-founded and condemn sector per- 1989. The condusions parallel many of those brought fonrance too readily. (Annex III conatins further dis- out in FY88 and subsequent annual Sector Reviews. cussion of methodological issues.) CQNRAL MESSAGES THE STATE OF THE SECrOR It is essential for the Bank to take the issues In spite of the mwetodological concerns, the raised in the OED report veiy seriously, and the Bank OED report rightly portrays a sector with serious has been doing so. The management response to the perfornance problems.3 The performance of water Board emphasized four central messages on which andsanitationorganizationsinmostdevelopingcoun- thereiswideagreementamongmanagementand staff tries is poor from several perspectves. From an eco- nomicperspective,availab'eresourcesareseldonused r First, we agree with OED that perfomance in this sector is inadequate. 3. Although, as shown in Table 5, the performance of Bank waterandsanitationprojectsisaboutaverageforinfrastuc- e Second, we have significant disagreenents with tureprojects,whlchpt 4orm,intum,somewhatbetterthan OEDon thereasons for this poorperfomanceand, Bank prjects overall. 5 efficiently. To cite just one example (also noted by w Theratioofpersonnelcostsonoperatingcosts(PC,I OED), the levels of unaccounted-for water in most OC)forW&SutilitiesinEngland,France,Germany, citiesindevelopingcountriesarethreetofivetimesthe and Spain and in selected developing countties is typical levelsin industrializedcountries. Fromafinan- shown in Figure 4. In spite of the relatively high cial perspective, water and sanitation utilities in devel- labor costs in this group of industrialized countr.es, oping countries seldom collect sufficient revenues to PC/OC ratios are, on average, under 40 pecent. cover their recurrent costs, let alone provide reserves Yet, as illustrated, many utilities in developing for future investrment. In part this is due to factors countriesattainlevelsofpersonnelcostsinexcessof beyond the contol of the utility (specifically, the tariff 60percent.Theaverage forBrazilandColombia,for levels); in part it is due to the inefficiency of both example,isaround55percent,considerablyhigher technical choices and commercial operations. Finally, then the European average. from an equity point of view perfonnance is poor. It is poor people and smaller industries that are either or The ratio of staff per 1,000 water connections(. J denied service or bear the brunt of unreliability, while 1,000wc) or 1,000 water+sewerage connectiors (SI te consumption of the rich and mi idle class is subsi- 1,000(w+s)c) are frequently used as indicators of dized. company performance. When related to personnel costs, the picture would be expected to be fairly As part of the preparation for this Annual straightformard, i.e., companies with high person- Review, a special study was undertaken to examine nel costs would be assumed to have a higher staff/ personnel costs as a share of total operations and connection ratio. Thus the problem would be of a maintenance costs. Although this study wl be issued technical nature and if personnel costs were per- ina forthcoming INU publication, 'Personnel Costs in ceived as too high the solution would be to reduce Water andSanitation Agencies," itisusefultohighlight staff numbers. It seems, however, that staff/con- the major findings here nection indicators alone do not give a satisfactory picture of the situation. An analysis of the Colom- w Bank projects, as well as those of other agencies in bian sample of 16 W&S utilities shows that in 75 thesector, haveaddressed the staffingand produc- percentof thesamplethestaff/connectionindicator tivity problems of public water utilities through wasimproving (decreasing) orstationarywhilethe tecnical assistance components and action plans. impact of personnel costs on operations was in- However, the results of most of these efforts have creasing. Figure 5 illustrates this point. For the last been disappointing. As noted in a recent OED 10yearsthestaff/connectionratiointhiscompany review, "public enterprises ... are key elements of inLAChasbeendecliningsteadilyandhasreached patronage systems ... overstaffing is often rife, and levels very few companies in the region can emu- appointments to senior management positions are late. However, personnel costs (in real terms) and frequently made on the basis of political connec- their incidence on the cost of water delivered have tions rather than merit" beenincreasingatanalarmingrate;tothepointthat the company has been in a delicate finandal situa- o Personnel costs constitute an important cost factor tion for some years and isnow near insolvency. As in the operation of W&S utilities. Most of these a consequence, it has also reduced the allocation of utilities operate in an urban setting or on a national funds for maintenance. orregional basis. High orrising personnel costs that are not matched by productivity gains cast doubts w It is particularly worrisome to find that salary onthesustainabilityof sectoroperations. Theyalso increases and productivity gains are totally de- raise equity questions when these high costs are at linked in many cases. Average salary levels and theexpenseofadequateservices,particularlytothe theirincidenceonthecostpercubicmeterdelivered poor. are growing very fast, in real tenns, in many W&S companies in developing countries. Based on a w When the financial statements of sector companies sampleofBankprojectsimplementedbetween1965 in developing countries am examined, it becomes and 1987, the incidence of salaries on the cost of clear that (a) the impact of personnel costs on the water sold grew in real terms (over a 6 to 10-year cost structure in most of these companies is very high, and (b) the incidence of personnel costs on operations4 varies widely, even from company to 4 Operatingcosts,astreatedhere,donotincludedepreda- company in ihe same country. tion or interest payments. 6 period) at unsustainable annual rates, as shown in Inpractce,autonomyforthesupplyorganiza- Figure 6. tion has stood for a variety of specific acons rangn from establishment of a corporate structure (with an o Thereisa need tok lin salaries to producdvitygains "independenr board of directors) to freedom from a if sectorinstitutionsaretoremainfinanciallyviable. number of constraints on the rights of the corpoation, Otherise they will severelystrain cental and local tihat would otherwise prevail-vil service rules, sala- govemnment resources. Linited as these resources ries,hliring and firingdecisions, and pricingof sevoies. are, a likely outcome will be a deterioration of Of course, whether these wll tum out tobe thecorrect services if corrective actions are not taken. answer depends on how the organization uses these freedoms, if granted. These cerrective actions include sector re- fonns that (1) inczease accountability of institutions To understand how sectoral institutions per- and managers, (2) increase participation of the private form, and to understand how perfornmnce might be sector, and (3) provide voice to consumers. Improved, it is instructive to consider the interests of the different stakeholders. First, of course, are the WM HAS PROJECr PERFORMANCE actualandpotential usersof theservice. Whatcanthey BEEN POOR? legitimately expect? The most important thing is delivery of the quantity demanded in services which The "good management practices" incorpo- meet reasonable standards of quality and reliabflity. rated into the relevant OMs, most dating from the late Users can also legitimately expect that these services 1970s, remain appropriate and consistent with those will be available ata cost that does not include prenia recommended by OED and virtually all sector profes- attnbutable to inefficient investment and operational sionals. The OED report shows that when these prac- decisions. Second, the owners of the assets (whether tices are followed, reslts are good. However, as docu- they are governments or private individuals) have a mented by OED, the problem is that this has happened legitimateinterestin themaintenanceofthevalueof the in only a handful of cases-Singapore, Korea, assetsandareasonablereturnfromtheuseoftheassets. Botswana-wherethequalityofadndnistrationis n- Theyalsohaveaninterestinavoidingunnucessaryand usually good. The key questionis, why has advocacy 1tmexpected expenses attributable to the operation, of these good management practices not worked in maintenance, and expansion of the assets. Third, umst of our borrowing countries? investors in the assets have an interest in the eun on their investment. Finally, themanagers and staff of the We disagree with OED on the answers. The organization can legitimately expect to be compen- core of the OED explanation can be interpreted as sated for their contributions to the productivity of the follows. Supply organizations in the sector have not organization. beengrantedsufficientautonomy,andtheconsequence of thishasbeen toonmuch political interference and lack This perspective is valuable in considering the of wil (including lack of will by Bank managers) to do likelylong-termsuccessof thereformswhichOEDand the right things. This has been further exacerbated, others desire. First, it is clear that there are potential accordingtotheOEmreport,sincetheBankisrelatvely conflicts between the interests of the different stake- powerless to address management issues and has holders. Second, it is obvious that a nornative ap- failed to enforce agreements (especialy financial cov- proach which does not recognize the fact of these enants)madewithbomowersattheinceptionofprojects. interests and incentives will not work More specifi- In our view, however, the causes of poor performance cally,wewouldarguethatthebasicreasonwhycondi- lie within both service organizations and the enabling tionality has not worked is that it has assumed these environment. Accordingly, although we agree that incentivescanbeovemddenbyobligatoy covenants. autonomyisdesirable,forserviceorganizationsitisan Rather than making minor revisions in the standard incornplete institutional response. conditionalitypackageofthepast,wewouldargue that the key institutional challengeistocreateincentives for Autonomy for supply organizations is not a all the stakeholders to maxiize the congruence of new recommendation. It is, in fact, the most frequent interest and to perform in an efficient way. World "institutional reform" measure recommended by the DevelopmentReport 1991, TheChallengeof Develop- Bank Thereisaneed,however,toexploretheconcept ment rmade a strong case that "market-fiendly" ar- ofautonomyas it mightapply to the water and sanita- rangemnents tend to accomplish this. tion sector in greater depth. 7 A related problem with OEDYs analysis of the w Providing Sevic People Want and are Willing problems,webelieve,isthatthereportfaulstoreflectthe to Pay For. In urban areas there is abundant profound implications for the water sector of, first, the evidencethatmostpeoplewanton-plotwatersup- increasing acceptance of participation as a basic ele- plies of reasonable reliability and are willing tpay ment of the development process and, second, the the full cost of these services. In some areas this widespread adoption of a market-based approach to standard solution will have to be adjusted and development First,considertheparticipationissue. specialeffortswillneedtoben-mdetoacomonwdate WithdemocratizationinBrazll,forexample,neighbor- poor people. It has been widely assumed that the hood associations have brought highly effective pres- demand situation in rural areas is quite different, sure on the state water companies to extend services to that there people have only a '"basic need' that can the poor. Nongovernmental organizations are simi- bemnetwithapublictaporhandpump. Butarecent larly involved in the design and execution of water multicountzy study by the Bank of rural water supply projects for the poor in Pakistan, Zimbabwe, demand found thatmostruralpeoplewwantandare Indonesia,andmanyothercountries. Second,consider willing to pay for a relatively high level of sevice the adoption of the market model, described in detail (yard taps or better). in WDR 1991. Just as the role of the private sector in the provision of water services has expanded in many or IncreasingInvestments in Sanitation. Of itslend- industriaLizedcountnes,sotoo,therearemajorchanges ing for water and sanitation, less than a fifth is fox inmanydevelopingcountries. Forexample,Argentina sewerage and sanitation. Adoption of a demand- has accepted bids for the privatization of the water driven sanitation strategy could increase this level supply of Buenos Aires, and Guinea has granted a of lending. There is abundant evidence that uman concessioncontract fortheprovision of waterservices families are willing to pay substantial amounts for in Conakry. Equally telling are a variety of data the removal of excreta and wastewater from their colected by OSP on what skills our water sector staff neighborhoods, once they have received an ad- most need. There is unanimity among our operating equate water supply. Subsidy levels can be held to division chiefs and sector staff that by far the highest manageableproportionsbyexpandingthemenuof priority is infornation and training on issues related to supply options, by involving users in the construc- pri atesectordevelopmnentandthenecessarylegaland tion and operation of "feeder' sewers, and through regulatory framwworks (See Annex IV for a detailed concentratingnon-userfinanceonmuncipal waste discussion of specific OED recommendations and the treatmentanddisposailatherthanonwatersupply management response to them). and waste collection. Fnally,webelievetherefonnagendaouftlined w RethdnkiglnsmtitutonalA angemets Fivernain in World Development Report 1992 is worth restating. elements are needed in a strategy for institutional This agenda is consistent with what we believe is the reform. The first is to improve the performance of coreof the challengeinoursector, and hasbeen widely public utilities. Second,provisionofsectorservices ascribedtobyourborrowersinthedebateslnitiatedby must be separated from regulation. Tird, the role WDR 1992. Ihe four key policy changes needed are: of the private sector should be expanded. Fourth, there is a need for mechanisms to give consums w Better Managment of Water Resources. Among voice and for community involvement. Fifth, it is our borrowers a start is being made in applying essential to create conditions forthe privatesector, innovativemehdodsformanagingwaterresources- NGOs, and consumers to enable them to play their essentiallymarket-driven methodsforreallocation parts while the govemnnt concentrates on what and promotionof conservation. Anessential tool in only governments can do, namely, to define and this process is the pricing of water as an economic enforce an appropriate legal, regulatoty, and ad- comnodity with an increasing opportunity cost ministrative framework As stated in World DevelopmentReport 1992, 'The striling features of ... market-based reallocation How TM BANK WIm MovE FORWARD methods is that they are voluntary, they yield economic benefits for both buyers and sellers, they Primary relianceoncovenantsandmorerigid reducetheenvironentalproblemscausedbyprof- enforcement of these covenantsby themselves wil rnot ligateuseof waterforirrgation,and theylessen the solve the problems facing the sector. Rather, the Bank need for more dams." will pursue a sector improvement strategy based on two key points. 8 Fu'st, Operations and OSP work, as well as the importance of this work on indicators, the office of elemnents of the OED Report, argue for a broader view the Vice President, OSP, has allocated special funds to ofthedeterminantsofperformanceofsectororganiza- INUWS for this purpose.) In the long run, it is antici- tions than has tended to prevail. In conducting sector pated that these performance indicators will become a operations, we will pay particular attention not only to basic tool used by the Bank and our borrowers for the issues under the control of a specific supply organiza- systematic monitoring of the performance of specific tion, but also to the environment in which the organi- projects and the sector as a whole. zation funcdons. Critical elements of thisenvironment include (1) regulatory and administrative controls (in- This effort is already under way, through joint cluding economic, financial, and legal controls); (2) effortsof OP and Operations. OSP staff areproviding allocationproceduresandpracticesforpublicexpendi- Operations support on some projects in which the tures; and (3) water resource management-availabil- support is specifically designed to facilitate and dis- ity, alternative uses, and allocation. To a considerable seminate learning. The UNDP-World Bank Water and extent, these features of the enabling environment Sanitation Program, which concentrates on informal determine the incentive structures of the principal institutions in rural and periurban areas, recently stakeholders in the sector. As is beginning to happen adopted a new strategy, 'Improving Services for the in a number of borrowing countries, greater attention Poor A Program Strategy for the 1990s." The core of will be given to possibilities for introducing more this strategy is the systematic learning from large-scale competition for and within markets, and greater levels investment projects and programs financed by the of private sector involvement, even while continuing Bank and others. This approach is already being efforts to improve the performance of publicly owned applied with considerable success in a number of Bank and operated utilities. projects. For example, at the invitation of LA1IN, INUWS has assumed a major role in the "strategic Second, since this raises issues that have been supervision' of the innovative, experimental, Bank- dealt with only sporadicaly in the past, the Bank will financed PROSANEAR project This involves ensur- acknowledge through its actions that more still needs ing that the lessons of successes and failures, especially tobeleamed,andleamedrapidly,inimplementingthis in Brazil itself, will be taken into account; ensuring that new approach. As described earlier in this note, a the subprojects incorporate technical and institutional lear:ng approach is essential both in adjusting a par- innovations; and setting up a monitoring and evalua- ticular project as experience with that project accumu- tionsystemforfeedingbackinformationonwhatisand lates, and in feedingback thelessonsof experience into what is not working. Similarly, INUWS staff are the design of new projects. Accordingly, the Bank has workingcloselywithOperationsstaffonincorporating initiated activities aimed at developing a more consis- the structured learning appoach into the design and tentand structured way to leam from ongoing sectoral execution of rural waterprojectsin Indonesia,Pakistan, activities, rather than wait for further .etrospective SriLanka,Nepal,Philippines,Nigeria,andZimbabwe. studieswithlargetimelags. Becauseitisbutone(albeit an important one) actor in the sector, the Bank will A similar strategy is being developed for for- involveourborrowersandothermajorfinancingagen- mal sector institutions. NU recently co-hosted a cies in this learning process. Initially the "structured workshop in Brussels with UNDP to design a broad- learning" process will focus on projects that provide based structured learning process for water utilities. particularly rich opportunities for learning. The neeting, which included multilateral and bilateral organizations active in the sector as well as sector Once an effective approach has been devel- professionals from developing countries, resulted in oped, however, the objective will be to generalize the theinitiationofajointUNDP-WorldBankBank "Water process. In thelongrunitisexpected thatBankprojects Utilities Partnership," abroad-based efforttoapply the will generate a richerand consistentsetof data, so that structured learning conceptto thefornal sector. Akey leaming is more systematic, less qualitative, and more pr-nciple behind the Partnership is that reform cannot generalizable. Early priorities will include a program be the business of the Bank only, but also requires the aimed at developing indicators (paralleling an effort development of a broad consensus among at least the NURD has started for the housing sector) that will majorfinancingagenciesandtheborrowingcountries. enable us to say whether sector performance is ad- The short-term objectives of the Partnership include: equate in comparison with other international experi- ence, whether it is improving, and what the specific * productionofpractical,up-to-datedocumentation bottleneckstoimprovedperformanceare. (Becauseof of experiences with innovative formal sector ap- 9 proaches ranging fromn concession contracts in Ar- United KingdomandFrancelearningaboutapproaches gentina, to statutory bodies in Malaysia, to the to private-sector participation in the water and sanita- incorporation of popular participation into the op- tion sector. A similar number of Bank staff will under- eration of state water companies in Brazil; take a week-long study tour of the French water re- sourcesmanagementand regulatory system this fall. In * production of practical guides of best practices on both cases the study tours have been oversubscribed. key reform issues such as the regulatory environ- The Bank is also in the process of preparing a new ment and contracts; training plan for water and sanitation sector staff. The plan focuses on institutional issues and proposes joint * production of papers investigating key themes, training of Bank staff with key decision makers and such as innovative approaches to stimulating the sectorprofessionalsinborrowingcountries. Thereport involvement of the private sector, with particular is available in draft form. etnphasisonthedomesticprivatesectorindevelop- ingcountries;and Finally, OSP has taken specific steps to strengthen the Bank's capacity to deal with the chal- e workslhops for both policy makers and practitio- lenges the sector faces. At reorganization in 1987 only ners in specific borrowing countries which have one Bank staff position was allocated to water and decided to embark upon reform. sanitationinthepolicyandresearchcomplex. Overthe past four years the situation has been substantially TheBank'strainingprogramisalreadyreflect- improved, with the current allotment being five Bank ing the new orientation described above, and there is staff positions to INUWS. Because this is still small for demand among Bank staff for training in the new skills a sector that deals with about6 percentof Bank lending that are required. Two recent training initiatives illus- and faces such enormous challenges, OSP is consider- trate this. A group of 25 Bank staff spent a week in the ing increasing the resources for INUWS. 10 ANNEX 1: FiGuES ANDTABLs Figure 1 BANK/IDA Operations US$ millons No. Operations 3000 30 2600 ../; . 265 2000 20. ........ .... . . . 20 1600 1 .6 ; ....... _ . . . 1000 ...................... . . .. ......... .. . . 10 G o o .... ... ..... ... ......... ......... ... ..... 6 0 0 1988 1989 1990 1991 1992 1993 1994 1996 1996 1997 Fiscal Year Historic =1 Lending Program Figure2 Lending Operations Total S-W No. Operatlons 2. 26 *2 [.20 I .. ... ... .. ...............- 10 0.6 .....6 0 0 1983 1984 1986 1986 1987 1988 1989 1990 1991 1992 FISCAL YEAR _ Total Staf Weeks E Number Operations 11 Figure 3 Allocation of Staff Resources Staffwoeks (Thousands) 4 a . ......... ......... ..... ...... ...................................... ........... ........ ............. 2 . ...... ... ......... ............... ................ ................ .............. ... ...... . ................. ........... ... ....... .... .... ......... X 1988 1989 1990 1991 1992 Fiscal Year WTotal 8Staiweeke E1 Lending E 8 Sup.rvision Saeotorv0 rk Figure4 Personnel Costs Personnel Costs/Operating Costs S 80 to .................................................................................................................................... .................................... 60 2| 60 §SBB,2B 13ER FflA ENG3 SPA COL BRA MEX TUNs TUR YOU THA Selected Countries 12 FiguIr 5 Personnel Costs Pere Cost/M3 sold Statf/1000 w conneG 160 .10 9 0 . .......................... ........................ ......................... ............. 6 60 s o . ....... . . . . . . . . . .. . . . .. . . . .. . . . . ... . . . . . . . . . . . . . . . . . . .4. . . . . . . . - . . . . 30 _ .. .. 2 - 1I l 1982 1983 1984 1986 1986 1987 1988 1989 1990 1991 Year - Peroon Cost/M3 sold - Staff/1000 w connec. Booata (In Pose of 1991) Figure6 Personnel Costs % Real Increase (89/90) 100 s o0 .................I............................................................................................................................................ 6 0 . ...........................................................................................I.................................................................. 40 .... . . . . Cal Med Buo Bog AcV Cue Per Dul Man Mon SAn Pop city Colombia. % Increase Per. Coats/M3 soid (a9/s0) 13 WATER & SANITATION SECTOR Loans and Credits ($ millions), FY92 Loans and credits Total project Country Project $ millions Cofinancing cost IBRD IDA Total BRAZIL Water Sector Modernization 250.0 - 250.0 -- 500.0 BURUNDI Water Supply Sector - 32.7 32.7 18.8 54.7 CHINA Beijing Environ. 45.0 80.0 125.0 1.6 304.5 CHINA Rural Water and Sanitation -- 110.0 110.0 189.1 Southeast Sewerage and CYPRUS Drainage 32.0 -- 32.0 32.0 103.3 IRAN Teheran Drainage 57.0 -- 57.0 0.0 100.0 KENYA 2nd Mombasa and Coastal Water Supply -- 43.2 43.2 0.0 48.7 KOREA Pujan and Taejon ____________ Sewerage 40.0 - 40.0 0.0 272.1 LESOTHO Highlands Water Project 110.0 - 110.0 759.2 2414.0 MAURITANIA Water Supply - 10.5 10.5 0.0 13.8 NIGERIA First Multistate Water Supply - 101.0 101.0 0.0 128.2 TOTALS 534.0 377.4 911.4 780.6 4128.4 WATER AND SANfTATION SECTOR FY93-97 Pipeline ($ millions) REGION FY93 FY94 FY95 FY96 FY97 Total No. of projects 3 7 6 3 1 20 AFRICA Amount 130.0 228.5 266.8 212.0 250.0 1087.3 No. of projects 4 1 5 I 1 12 EAST ASIA _ ______ Amount 320.0 200.0 615.0 100.0 75.0 1310.0 No. of projects 3 1 2 5 4 15 SOUTH ASIA I Amount 224.5 25.0 100.3 440.4 210.4 1000.6 No. of projects 2 1 1 3 1 8 ECA l Amount 161.7 250.0 70.0 325.0 100.0 906.7 No. of projects 3 3 6 3 15 MENA _. _____________ Amount 292.7 152.7 615.0 90.2 1150.6 No. of projects 4 4 9 2 3 21 LAC l Amount 433.0 330.0 1135.2 220.0 300.0 2418.2 No. of projects 19 17 29 17 9 91 Totals _ Amount 1561.9 1186.5 2802.3 1387.6 935.4 7973.4 Un 0) WATER & SANITATION SECTOR Loans and Credits, FY92 _3 Project Components ($ millions) 9 Country Water Sanitation T.A. and Other Base Cost Environment Engineering ____. BRAZIL 232.0 121.7 65.6 -- 419.3 BURUNDI 34.8 - 6.2 1.0 42.0 CHINA (Beijing) -- 113.4 13.8 117.2 244.4 CHINA (Rural) 120.9 5.9 12.8 6.1 145.7 CYPRUS -- 57.4 5.0 17.6 80.0 IRAN -- 72.6 3.2 0.7 76.5 KENYA 11.1 _ 6.2 1.0 42.0 KOREA -- 184.6 3.1 41.2 228.9 LESOTHO 982.0 -- 171.0 205.0 1,358.0 MAURITANIA 8.4 2.6 0.3 11.2 NIGERIA 87.5 -- 11.4 3.3 102.2 Total 1,476.7 555.6 320.0 396.0 2,748.2 % 54 20 12 14 100 EXAMPLES OF OPERATIONAL SUPPORT PROVIDED BY UNDP-WORLD BANK WATER AND SANITATION PROGRAM, FY92 COUN- BANK PROJECT MAJOR DESCRIION OF ACTIVITIES TRY _ ACTIVITY _ Mali Mali Sud III Project Preparation and Assistance in peparation of midterm evaluation repost. Design of a strategic approach for Supervision construetion of water points. Assistance also provided in community development aspects of rural water supply component. Nigeria Rural Water Supply and Sanitation Preparation Preparation of sector strategy paper for upcoming Rural Water Supply ProjeCt using lessons learned Project from the Rusafiya project. Uganda Northern Reconstruction Project Appraisal and Participated in appraisal mission to review institutional arrangements for implementing project. Preparation Ethiopia Drought Recovory Project Preparation Participated in the design of Emergency Drought Relief Project. Bangla- Urban Development Project - Sanitation Supervision Assisted Dacca City Corporation and Chittagong City Corporation in implemen-tation of sanitation desh component l component, including TOR for Low-cost Sanitation Units. Bangla- CWASA IIl Project Preparation Assisted in prep., participated in appraisal, and provided training as part of associated tech. assist. desh credit. India Karnataka Rural Water Supply and Envi- Preparation Review of engineering designs. ronmental Sanitation (RWSES) Project India Hyderbad Water Supply and Sanitation Supervision Assisted in the implementation of low-cost sanitation component, and preparation of training plan for Project Water Supply and Sewerage Board. Nepal Rural WSS Project Prepaation Participated in preparation of TOR, selection, and supervision of project preparation consultants. _ Prepared TOR and sought funding for grant-aided pilot program to test approach. Pakistan RWSS Project Supervision Provided support in training incL. study tour and community mgnt serninars for project personnel. China RWSS-II Project Preparation and Assistance in project preparation and implementation for water and sanitation programs in six Supervision provinces. Indo- Community Water and Sanitation Project Preparation Preparation of WSS program for low-income communities (WSSPLIC). Preparation of guidelines for nesia _ planning and implementation; preparation of plans involving local authorities and communities. Philip- Third Water Supply Sanitation and Preparation T.A. in prep. of specifications, quality control, and training. Prep. of TOR and supervision of pines Sewerage Project consultants to conduct a comprehensive study to improve sanitation in provincial urban areas. Bolivia Bolivia Integrated Health Project Supervision Implemnentation of two pilot projects for delivery of periurban water supply and sanitaion services Ecuador Second Social Development Project Preparation and Prepared and appraised a rural water and sanitation component __ __ __ ____ Appraisal Tunisia & Sewage Treatment and Reuse Project Preparation Formation of proposal and demonstration of two pilot projects showing innovative wastewater Morocco treatment and reuse options in urban areas. Table 5: Performance of the Water and Sanitation Sector Relative to Other Sectors Projects rated satisfactory, 1974-90 100 90 A A infrastructure e0 I . AUI sectors 70 60 . 50 40 30 20 10 0. ~~ d. ~~~C Projects likely to be sustainable, 1986-90 ~~ I- *E 100. 80 60 - All infrastructure 20~~~~~~0 Projects with substantia achievement of institutiontl objectives, 1988-90 60 - 40 1 mm1I AU infrastructure 30 - sectors 20 f 0~~~ 18 ANNEX II: ARIS DAxAONNTHEWATER AND SANATION SECIR Water and sandtation projects in the Bank's portfolio than all other sectors. The problem projects are tically range between 5 percent and 6 percent of all concentrated in thegroup of projectsin thefourth to the 4 .ojects(FY90,5.8percent;FY91, 53percent; FY92,5.1 eighth years of implementation (42 percent of projects percent). In tenns of dollar commitments, the water hi he fourth and fifth yearsof implementation, and 36 andsanitationsharehasincreasedmarginally,from4.7 percent in the sixth to the eighth years). Regionally, percent in FY90 to 4.9 percent in FY92. MNA has the largest share of problem projects in the portfolio In the sector: 22 percent of MNA's problem In terms of overall ratings, the water and sanitation projects were water and sanitation proIects in FY92, portfolioisaboutthesameastheaveragefortheBank; comparec' - 12 percent in LAC, 6 percent in ECA, 4 although it went from slightly better than the average percentinAAiicaandSouthAsia,andnoneinEastAsia. in FY90 (18 for the sector, 1.9 for the Bank) to slightly worsein FY92 (2.1 for thesector, 2.0 for theBank). The The sector has responded to the inaease in problem sector has experienced a fairly sharp increase in prob- projects and the decliningratingsof the sectorportfolio lem projects as a percentage of operations, from 17 by a substantial increase in the staffweeks of 'upervi- percentin FY90 to26 percentin FY92. On this indicator, sion per project, from 8.6SW per project in FY90 to 11.7 the sectr has a higher percentage of problem projecs S"'! in FY91 and 13.9 in FY9" 19 AmNX III: SCOPEANDMBTHODOLOGY oF TEf OED REPoRT Thwee features of the coverage of the OED report and a zation" and "project preparation and appraisal." The generalcommentonthemethodologyusedbyOEDare cwmments were further divided into subcategories in order. Regarding coverage, first, the OED chose not (under the category "sector strategy and organiza- to review eight engineering and techndcal assistance tion," the subcategories include "autonomy," "institu- projects, even though such operations are precisely tional structure," etc.) and the number of posifive and intended to address technical and institutional issues negative comments in each sub-category recorded. prior to an investment project, an activity which the The resulting tables form the basic "data" on which the OED report recognizes as critical to the quality of an text of the report elaborates. operation. There are several problems with the rnethodology and Second, although the ieport makes reference to re- the way it was applied in the report. First, the peater projecs in the same locations, this subset of translation of written comments into numbers is sub- projects was not explicitly analyzed as a group. As a jective, relying heavily on the sector knowledge and resultanopportunity totestthe effectsof Bank involve- judgementof the principal authors of the report (both ment over time was missed. of whom have worked in the sector in the Bank for many years.) Second,inmanyinstancesthereportfails Third,exceptforabriefmentioninthereport'sConclu- to infomn the reader of the link between the "quantita- sions and Recommendations (pp 62-63), the studies' tive"dataandthecondusions. Thenetresultisthatthe retrospective finings of prjects appraised more than reader is often left with doubts about the basis of tenyearsagowerewnotlinkedtomorerecenttrends,as particular conclusions and is unable to return to the have been explored in considerable depth in ARIS data to ask ancillary questions (for example, were reportsandintheOSP intenmal AnnualSector Reviews seculartrendsaconsequenceofachangeinthecompo- forFYs88,89,90,and 91. Thesereportsreflectcurrent sition of projects?). effortstoaddressimportantsectorissuessuchaswater resource management, envirtonmental improvement, These methodological problems have important bear- finance, and sector reform, and to show that Bank staff ings on the credibility of some of the conclusions of the in the sector are aware of most of the perfonrance report Inthisfirstroundofreviewofthedocumentby deficiencies cited in the OED report and are actively Bank staff, for instance, questions were raised about a engaged in doing something about them. large number of the specific conclusions of the study. Becauseof themethodology of the study itwas impos- The OED report is based on a mewthodology which sibletodetenninehowthe conrusionswerearrivedat, dependsforitsvaliditylargyonthesectorknowledge and whether, indeed, more nuanced or different con- and the judgement of the people who prepared the clusions might have been more appropriate. An report. In preparing the OED report, reviewers read unfortunate consequence is that there maybe an inch- PCRs and PPARs and recorded commnents which they nation, on one hand, to dismiss the significance of the deerned tobe "positive" or"negative"ineachof several report tooreadilyor,on theotherhand, to takeaU of its categores. These categores comprise the main chap- specific negative statements as well-founded and con- ters of the report, such as "sector strategy and organi- deann sector performance too readily. 21 ANmX IV: SPEcIFIC REsPONsES To OED RECOMMENDATIONS These conunents are organized under the headings disposal. Experlence suggests that, once the demand used in the OED reporL References are to paragraph for water is met, the demand for the removal of waste- numbers in the report. water arises organically, and that this is often the appropriate time to deal with wastewater issues. The Sector Strategy and Organization (paras. 2A3-44): sequence "water firt" is not, therefore, always unrea- OED ca11s for Bank poicies and guidelines for water sonable. resource n rmagmet. Ihe importance attached by OEDtowaterresourcemanagementiswelcomed.This Withregard tofinancing,thereisconsiderablewilling- recommendationhasbeenantidpatedbytheBoard.A ness to pay for wastewater collection services. An policy paper on water resource management is to be essential component of a demand-oriented approach presented by the Bank to the Board this fall. Further involvesofferingindividualsandconunuitiesamenu developnent of guidelines should await this policy of technical and instiutional options (often induding statement. inproved pit latnnes, simplified sewerage, and con- ventional sewerage) from which tochoose. Experience OED reconnends the use of adjustment operations inBrazil,Paldstan,andothercountrieshasshownthat, andtechducalassistanceaheadofinvestmentlendngto relative to the conventaonal provision of sewers, this achieve institutional aims. However, multiple instru- can result in large cost savings and much higher pe- nents are already in use to improve capacity. The portionalcontnibutlonsbythebeneficiailes. TWshasa specfic instrunents used should depend on country number of desirable effects, including reduced pres- cirumstabnc. We agree with the need for a broad sure on government finance for waste collection sys- approach,butdonotagreethatspedficprojectscannot tems. beeffectvetoolsfordealingwithbroaderpolicyissues. Consequenly, the use of project lending should notbe Inevitably, however, it will be essential to use some overlooked as a means of stengthening institudons, public funds for the provision of ful collection, treat- ment, and disposal of wastewater. Becausethecostsof Project Preparation and Appraisal (para 3.26): OED such services are very high, even industrialized coun- makes no new recommendations under this heading. tries typically cannot treat all of their wastewater. The suggestion that the Bank should apply OMS 2.28 While we do believe that the proportion of overaU requrements and carefully review technical designs publicspendingdevotedtowastewatertreatmentwil and forecasts is entirely appropriate. Demand fore- andshouldincrease,intheBank'sborwwingcountries casts,whichOSPandOEDhavepreviouslyfound tobe it is imperative that priorities be set on what to treat, generaly overoptmistic, should receive special atten- where, and how, so that the ma,dmum benefits canbe tionsincetheyplayamajopartinselectionof technical derived from limited public funds. capacity choices and should be reassessed if recom- mendations are made for substantial tariff increases to Institational Development (paras. 5.9, 5.17-18, and achieve financial and econonic objectives. 537-38): OEDrecomnmendsmoreattentionbegivento O&M, reducng unaccounted-for water, and improv- Pxojectlmplemenatlon(para.4.24): OEDargues that ing utility management. On all these counts, OED cost overruns and delays have tended to reduce the recommends demonstrated perfonnance prior to in- benefitsof projectsand haveresulted in the sacrificeof vestment. But to make progress on these fronts very sewerage components. This has tended to be true. often requires both investments and time. A more Before concluding that sanitation has been realistic solutionrecommended in theinternal Annual underfunded, however, some issues relating to se- SectorReviewfor FY90isthatimplementableplansfor quencing, financing, and priorities need to be consid- improvementinO&M and reduction of unaccounted- ered. for water should be made an explicit part of project appraisal. Morefundamentally,wewouldargue,once Withregardtosequending,inallplacesthedemandfor again, that the fundamental issue is not the more watersupplyprecedesdemandforwastewatercollec- igorousspecificationandiimplementationofcovenants, ton, let alone envirtally sound treatment and but the development of institutional arrangenents in 23 which the stakeholders have incentives to perform analysis of costs and benefits of physical investments, better. Once this is done, there may indeed be cov- policy and institutional reforms, underlying assump- enants,butnowthesensewouldbemoreofastatement tions, sensitivity/risk analysis, and impact. The OD of reasonabletargets,andlessalegalrequirementtobe will also discuss procedures and responsibilities for inposed on the institutions involved. conducting economic analyses and for monitoring the 3ffort. After the work of this task force is completed, it Financial Performance (paras 6.27-629): The OED may be necessary to review other guidance material, report convincingly establishes that financial perfor- e.g., OMS 3.72, to apply the OD fully to the water and manceofsupportedorganizationshasbeenpoor. OED sanitation sector. In particular, additional efforts will doesnotdemonstrate,however,thatthemorerigorous be required to assess the opportunity costs of water, use of selective financial covenants, in the absence of incorporate environmental effects (both benefits and broader sectol reform as discussed above, will be costs), and utilize Bank experience to assess the sensi- effective. Financial performance is a function of many tivity of rates of return to various types of risk. more aspects of the sector than supply organization autonomy, e.g., selection of technical options consis- Impact on the Environment and Poverty (paras. 821- tent with demand, efficient operation of facilities, and 23): OED has recommended that sector projects in- the administrative/regulatory structure (particularly clude environmental impacts in cost/benefitanalyses. price regulation). Good financial performnance should This issue is partially discussed in OD 9.00 and will be be construed as an outcome of effective sector perfor- discussed in the new OD 10.40. We are in agreement mance. Defidenciesinfinancialreportingandauditing with OED and other commentators that much more have been addressed in OD 10.60 and have improved must be done to improve the ways in which waste is with the introduction of ARCS (Audit Report Compli- collected, treated and disposed of in our borrowing ance System), which has been in operation for two countries. The key elements, described in detail in years. WDR 1992 and summarzed here, are now being put into practice in a number of Bank projects. Regarding Economic Issues (paras. 7.18-20, 730-7.33): On tariff povertyalleviation,OEDsuggeststhatratherthanadd ussues, OED recommends preparation of guidelines oncomponentsaddressingpovertytoexistingprojects, based on proven international experience and case "it would be more effective to have poverty relief studies in developing countries. The OED discussion factored intoaspecificaUydesignedand targeted water of the issue, however, is once again too confined to the supply and sanitation project eventuallyasa follow-on concerns of individual supply organizations and too project." Whilethisapproachmaywellbeappropriate little addressed to the broader institutional roles to be in some circumstances, we ncte that stronger sector played by pricing, economic regulation, and demand- institutions inevitably extend services to a larger num- responsiveness. The issues around tariffs and tariff- berof people, and we believe, therefore, that the key to setting mechanisms are important elements of the providing decent services at reasonable cost to poor institutional reforrn approach discussed above, and people will often be the direct result of developing material will be prepared and disseminated on these viable institutions. When coupled with participation issues as part of the Utilities Partnership work de- by the poor, this approach has, for example, been scnibed on p. 14. spectacularlysuccessfulinservingthepoorinBrazilian cities. Wealsonotethatmorerecentprojectsthanthose OED raised a number of important issues regarding reviewed by OED have made some improvements in economic rates of return for water projects. As now poverty orientation. In the OSP internal Annual Sector calculated and used, ERRs are of limited use in influ- ReviewforFY90,forexample,itwasconcludedthatthe encing project design, judging project worth at ap- majority of the approved operations in that year tar- praisal, or as a benchmark for ex post evaluation of geted the poor as the rmain beneficiaries. projectperformance. TheBankhasrecognized thatthis is a multisectoral problem, and a Task Force is prepar- Overall Recommendations (paras. 920-23): In Rec- ingnewBankguidanceonthssubject. Theresultof this omunendation 1,OED reconunends that Bank policies work is being incorporated in OD 10A0,"Economic promote balanced and sustainable strategies for water Evaluation of Investment Operations." INJUWS is resource management The forthcomingpolicypaper participating in the work of this task force and the on water resource management will address this con- preparation of material. OD 10.40 will discuss Bank cern and issues of coordination across major sectors. policy regarding investment criteria, as well as the We do not believe that there is a need for a new policy 24 statementonurbanwatersupplyandsanitation. We that some changes in supervision practices will be believe that there is now broad consensus, along the recommended by the current Task Force on Supervi- lines descried in WDR 1992, among Bank staff on the sion. The water sector, like all other sectors, will be directionsin whichweneed tomove. Thatsaid,agreat affectedby,and putsumablyimprovedby,these modi- deal remains to be done in translating these general fications. directions into practice. We therefore believe that the key challenge is to design proects taking the broad Recommendations3and4, although listed separately, institutional perspective described in this memo, and form a package recommendation to establish prior to put in place a process whereby lessons are learned conditionsforinvestmentlending(relatingtoautonomy andfedbacktotheoperationsprocess. Asdescribedin and accountability for day-to-day operations, this mnemo, we believe that a reasonable start in this unaccounted-for-water, and prices, for example) direction has been made. As more is leamed about coupled with more rigorous and rigorously enforced sectorreform, anew policy statement may bein order. conditionality onloans. Asdiscussedinthismemo,we It is premature to prepare it now. believe that the central issue is restructuring of the sector organizations and the environment in which Recommendation 2 is for the Bank to undertake a theyfunction. Itisthefleshingoutofthisnewapproach thoroughreviewofongoingprojectsinthesubstantive in the context of policy and sector work as well as areascoveredbytheOEDreportand todetermine,case projectlendingthatcanprovideabasisforhopethatthe bycase,appropriatemanagementresponses.Thispro- next OED retrospective study will be able to report posed review is a standard feature of the ARIS process significant improvements. and will continue to be handled as such. It is expected 25