WATER P-NOTES 48727 ISSUE 39 jUnE 2009 Post-Construction Support and Sustainability in Community-Managed Rural Water Supply W ater projects in developing countries are as institutional and governance arrangements, gave inaugurated with great fanfare by the govern- women a larger role in decision making than was ments, lenders, and sponsors that make them customary, and required households to pay opera- possible; the projects' results, however, don't always tion and maintenance costs and at least some of receive the publicity of groundbreaking ceremonies. the capital costs of water services provided. This study reports the findings of a multi-country research project intended to discover how such Few of the RWS programs included the system- rural water supply (RWS) systems actually perform. atic provision of PCS. It was clear that communities Its emphasis was on how performance was affected needed some level of PCS, such as follow-up train- by post-construction support (PCS) to communities ing and technical assistance visits by engineers. Two after project completion. Information was collected approaches to providing PCS soon emerged. The from households, village water committees (VWCs), first was a demand-driven approach that ensured focus groups of residents, system operators, and key the availability of spare parts and technical services, informants in rural communities in Bolivia, Ghana, and Peru. Approximately 10,000 individuals regis- but left responsibility for seeking out and paying for tered their opinions. The great majority of the sys- them to the communities themselves. The alternative tems were found to be performing well; the factors approach was supply-driven, providing communi- influencing their sustainability should help policy ties with unsolicited training, technical assistance, makers, investors, and managers around the globe trouble-shooting, and repairs. who plan rural water systems. The research focused on how supply-driven post-construction support affected system sustain- Evolving Approaches to RWS ability given a well-designed community-managed Support RWS program. In all three countries, villages selected had received improved water supply Over a decade ago, a number of donors, working projects three to twelve years earlier. About half with national and regional water resources minis- the villages were treatment villages that received tries in developing countries, designed and imple- supply-driven PCS as part of the program, the mented community-managed RWS programs that other half were controls that had not received such incorporated components of a new demand-driven support. Secondary data was used to select dis- model to improve upon the record of earlier proj- tricts or regions where the village and household ects that had quickly fallen into disrepair. This new characteristics were similar to the treatment and model let households choose technologies as well control villages. This note presents important information contained in Post-Construction Support and Sustainability in Community-Managed Rural Water Supply: Case Studies in Peru, Bolivia and Ghana, edited by Alexander Bakalian and Wendy Wakeman, World Bank ­ Netherlands Water Partnership, June 2009. Readers may download the complete paper from www.worldbank.org/water. WATER P-NOTES There were unexpected challenges in all three from the improved water system; in Ghana, the countries. The research design required that the figure was 97 percent, and in Peru 95 percent. water systems in some villages be successes and others failures, in order to have variation in the The systems did occasionally break down, but dependent variable. In fact, far fewer projects were in most cases the VWCs had been able to arrange found to be failures in either treatment or control for repairs. Most villages reported one or more villages than expected, since almost all systems breakdowns in the last six months; in Bolivia, break- were working properly. Another threat to the downs were typically fixed in one or two days, in research design was the complexity of PCS provi- Peru the average was five days, and in Ghana 18 sion in the study settings. The supply-driven PCS days. Even communities that were not collecting programs were actually demand-driven in practice. enough revenue to pay operation and maintenance Only one program in Ghana received quarterly costs found the resources to make repairs. Many of visits from environmental health assistants to moni- these resources came from outside the community, tor the technical, management, and financial status as governments and NGOs often provided PCS for of their RWS systems; in the rest, when their systems maintenance. Some services were provided at the broke down, both treatment and control commu- community's request, others on the initiative of nities sought out support from nongovernmental governments, NGOs or church organizations. organizations, nearby municipalities, or even large commercial enterprises. Since most systems were working and repairs were made when needed, levels of household satisfaction were high. On average, 83 percent of Overview of Research Findings Bolivian households reported being satisfied or very satisfied with their system's operation and main- The demand-driven, community-management tenance regime, and 78 percent were satisfied or model seems to be working. Not only were the very satisfied with the performance of their VWC; in rural water systems producing water, but almost all Peru, 61 percent of the households reported over- households obtained at least some of their water all satisfaction with the improved water system; in from the systems. The vast majority of VWCs were Ghana, 88 percent of the households reported sat- functioning as planned. Community members felt isfaction with the repair and maintenance services, that they had been involved in the pre-construction and more than 80 percent of the women's focus planning of their water systems, and more than 90 groups said they were satisfied with the system. percent of the focus groups held with village leaders There were some troubling findings, however. or women reported that the community had been Almost all households reported using the new water involved in tariff design. In about two thirds of the villages in Bolivia and Peru (though less than half systems, but for some this was not their only water in Ghana), people felt they had been involved in source. In Ghana, 38 percent of households were the choice of technology, and slightly less than half still using water from unprotected sources. The of the people in each country felt they had a say in percentages were lower in Peru and Bolivia, but project siting decisions. Communities contributed still worrisome. Until households obtain their water 5-10 percent of the capital costs of the project, exclusively from improved sources, the health bene- though in many cases labor or land contributions fits of the improved systems will not be fully realized. were allowed to substitute for cash. Many of the VWCs were found to be in poor All piped systems studied in Peru, and all but financial shape. Households were paying very little one in Bolivia, were functioning at the time the for the improved water services. The programs did research was conducted. In Peru and Bolivia, 93 not require communities to finance the capital costs percent or more of households surveyed had opera- of construction or provide for capital replacement tional taps; in 55 percent of the communities in or expansion; the cost-recovery objective was simply Bolivia and 76 percent in Peru, all the household to collect enough revenue from users to pay opera- taps were working. In 90 percent of the villages tion and maintenance costs. A substantial minority in Ghana, all project boreholes were still working. of villages were not even achieving this modest Every household in Bolivia reported using water objective. 2 ISSUE 39 · JUNE 2009 Table 1 (5.1 in full report): Profile of village water systems and management practices (Note- some numbers missing) Bolivia Peru Ghana Description of the system Average years since project completion 7 7 6 Percent of villages with private connections only 73% 100% 0% Percent of villages with public taps only 4% 0% 100% Percent of villages with private connections and public taps 23% 0% 0% Status of the system Percent of households with functioning taps 95% 95% N/A Percent of villages with all taps functioning 54% 74% N/A Percent of villages where all project handpumps are workinga n/a n/a 89% Percent of villages with functioning systems that had reported a breakdown over last six months 55% 55 % 57% Average days to repair the system (for villages that had experienced a breakdown) 1­2 5 18 Management structure Percent of villages where the committees regularly holds meetings with the community 86% 81% 72% Percent of villages where the committee members are elected 95% 63% 42% Percent of villages where the committee members are appointed 3% 15 % 43% Median number of women in the committee 0 0 3 Percent of villages with no caretaker/operator 3% 2% 18% Percent of villages with paid caretaker/operator (in villages with a caretaker) 70% 57% 1% Cost recovery Cost recovery mechanisms Pay-by-the bucket or volumetric tariff 2% 0% 39% Fixed monthly fee 89% 82% 54% Fees vary by HH size 0% 0% 7% Irregular collections 0% 7% 16% No revenue collection 9% 11% 13% Percent of HHs in full sample who use the system that reported paying for water 87% 77% 71% Median monthly expenditure for water reported among HHs that pay for water (US$) $0.55 $0.30 $0.16 Percent of committees reporting that HH collections cover operating costs n/a 50% 51% Percent of committees reporting that HH collections cover minor repairs n/a 80% 65% Percent of committees reporting that HH collections cover major repairs n/a 12% 30% a88 percent of the systems in Bolivia were gravity only; the others used pumps. N/A: not applicable 3 WATER P-NOTES The research found no statistically significant systems for managing cash. Moreover, many association between technical PCS visits to help with households have little cash to spare, and cash repairs or maintenance and a working water system. flow is irregular and highly seasonal. People are Post-construction technical training of system opera- often distrustful of the accounting and security tors or caretakers was positively associated with of cash balances, and committee members may system performance in both Ghana and Bolivia. In distrust each other or not want the responsibility Bolivia, the share of households who were satisfied of securing cash. was, on average, 15 percentage points higher if the · When the committees do accumulate cash bal- village received a PCS visit that provided financial ances, villages often want to spend them on or managerial (though not technical) assistance. other development projects, so funds are often There was no evidence that free repairs, techni- raised only when the need arises. cal assistance, or an intensive supply-driven PCS program improved either technical sustainability or · Future capital and repair subsidies are fre- household satisfaction. quently provided to villages; in the projects studied, a significant number of VWCs had obtained donations and free spare parts and Implications for Sustaining and repairs from a variety of sources. Expanding RWS Systems These findings suggest that communities can and The Next Step should take full responsibility for their systems. The unsolicited PCS activities that appear most promis- The water supply systems studied are not financially ing are those that help communities to renew and sustainable without infusions of capital to replace existing infrastructure and provide for growth. It may increase capacity, such as post-construction training be that the sector's current capital financing model for system operators and non-technical support and the post-construction activities of NGOs and visits to help village water committees with adminis- other actors creates a dependancy that undermines trative functions or water use disputes. the principle of community self-reliance in the post- Important puzzles remain. Even those commu- construction phase and discourages communities nities whose cost-recovery systems are meeting from making their own investments in water infra- program objectives are not financially sustainable structure. enough to replace infrastructure as needed or The demand-driven community-management expand system capacity for growing demand. Even planning model has proven its worth, bit it has lim- though RWS were intended as one-time investments, its. Financial sustainability in RWS systems requires communities still rely on government and donors for a new policy model. Methods for achieving better capital subsidies, as well as repairs. coordination of NGO policies with government and In a significant number of villages, tariffs, if with each other seem especially important. One collected at all, are inadequate to cover the costs role for NGOs in the future could be as catalysts of major repairs, much less expansion or capital for post-construction support, such as training or replacement. The research suggests three principal locally-based models for raising capital, rather than reasons why this is the case: as dispensers of subsidies for communities that can- not manage to repair their own projects. The next · Generating substantial cash balances creates step is the design of a policy framework that will problems for the committees. The communities enable communities to handle the challenges of often lack access to convenient, secure banking system rehabilitation and expansion. The Water Sector Board Practitioner Notes (P-Notes) series is published by the Water Sector Board of the Sustainable Development Network of the World Bank Group. P-Notes are available online at www.worldbank.org/water. P-Notes are a synopsis of larger World Bank documents in the water sector. 4 THE WORLD BANK | 1818 H Street, NW | Washington, DC 20433 www.worldbank.org/water | whelpdesk@worldbank.org