2 | Social Accountability Notes nication, and transparency. The program was • Capacity building of the national NGO and to be implemented by a local NGO following a community stakeholders competitive bidding process. The Norwegian • Implementation of social accountability Trust Fund, the Trust Fund for Environmentally tools and Socially Sustainable Development (TFESSD), • Improved communication to encourage and the World Bank’s Communication for Gover- good governance and cooperation among nance and Accountability Program (CommGAP) stakeholders were to provide funding. • Monitoring and evaluation to measure The WBI program aimed to promote the progress, results, and outputs, and to track use of transparency and social account- outcomes, difficulties, and lessons learned ability tools in Uganda’s water sector and to encourage effective communication among Comprehensive Assessment stakeholders. Furthermore, it sought to insti- WBI conducted a comprehensive initial tutionalize the use of these tools within the assessment of Uganda’s water sector in collabo- MWE and the Directorate for Water Devel- ration with several partners: the MWE, the opment to provide training in sustainable GGWG, the Water and Sanitation Program in social accountability practices to a national Uganda (WSP-Uganda), CommGAP, and local level NGO that would implement the stakeholders. The activities and outputs of each program, as well as to community leaders, program component are listed in Box 1. local authorities, and local providers. Two The assessment results provided a compre- surveys were conducted to track changes hensive picture of the enabling environment in public opinion about the performance for promoting good governance and effective of water service providers in the Ugandan communication in the water sector in Uganda. town of Wobulenzi, in Luwero district. In particular, the assessment helped to identify A baseline survey was done in August Luwero as a geographic region that would 2008, and a follow-up survey in December substantially benefit from the program, as well 2009. Additionally, the project included a as possible partners to support the program’s participatory monitoring and evaluation of implementation. Wobulenzi’s water providers and supported the deployment of communication tools to Partnering with NETWAS Uganda in facilitate dialogue among stakeholders about Wobulenzi. WBI and the MWE identified five water use and services. Feedback from water qualified NGOs to bid for inclusion in the users was also sent to service providers. program. WBI then formed a three-person The program had five components: committee to determine selection criteria and • Comprehensive assessment of the local choose the NGO best qualified to implement context the pilot program. Ultimately, the committee Box 1. Activities and Outputs of Program Component Assessment activities Outputs • Assess the need for governance-related capacity • A communication assessment was conducted and development in the water sector; assess communi- communication guidelines prepared. cation and dissemination needs. • NETWAS-Uganda was selected as the local partner • Identify, hire, and orient a local partner to implement after a competitive bidding process. the program. • A local consultant was found to train NETWAS on • Define the capacity development needs of the chosen social accountability. partner. • A combination of citizen report cards and community • Evaluate the needs of the area selected for the pilot score cards was supported by water quality tests and and define the tools to be applied in the pilot. communication tools.
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 3 Two private water service providers operated Box 2. What Is Output-Based Aid? in Wobulenzi: (i) Trandit Ltd (Trandit), serving Output-based aid is a performance-based urban Wobulenzi and some peri-urban areas; payment to service providers that subsidizes and (ii) Bukalasa College, serving the local the cost of providing access to safe and clean agricultural college and its surrounding house- water to poor users. Its aim is to provide or holds. The two providers operated under widen access to safe water when tariffs do not contracts. Bukalasa provided water as a private cover the full cost of connecting or providing operator, whereas Trandit functioned under water. However the subsidy payment is given to an output-based aid contract with the MWE the provider only after it delivers the pre-agreed (Box 2). Trandit’s contract with the MWE granted outputs, such as a specific number of yard-tap the service provider a subsidy if it installed connections or water service delivered for a 200 new connections within five years of the specified time. contract’s execution date. The first key feature of OBA is that it is Between April 2007 and December 2008, pro-poor. Subsidies offered to the provider WBI staff made four trips to Uganda to explain are designed to open access to water to poor the program to national and local authorities, people. The second key feature is that the water donors, and representatives from national and service has to be sustainable. Normally getting international NGOs. WBI also solicited feedback connected is what poor people cannot afford, and recommendations to reinforce the technical but after being connected poor people end capacity of NETWAS. That feedback enhanced up paying less for water than they were when the technical capacity of NETWAS. WBI incor- buying it at kiosks or from mobile tanks. OBA porated suggestions from other institutions— seeks to combine incentives with outputs. for example, representatives from other key The Ministry of Water and Environment has Ugandan NGOs were included in the training launched output-based aid projects in 13 small sessions and other key activities so as to expand towns and rural growth centers in Uganda. These capacity beyond NETWAS and to build coali- projects seek to increase the poor’s access to tions while implementing the program. water services by increasing the accountability of water service providers. The Ministry of Water Selecting Appropriate Social Accountability and Environment and the water service provider Tools. Appropriate social accountability tools execute a new contract that permits payment were needed to monitor and evaluate the to the service provider only after it has achieved performance of Wobulenzi’s water service specified delivery outcomes. For example, the providers. The tools would also be used to track selection of service providers for the contracts is improvements in service delivery throughout competitive, which further drives down costs. the course of the program. Ultimately, WBI and NETWAS chose to deploy three tools: citizen report cards (CRCs), community score cardschose NETWAS because of its experience in (CSCs), and chemical water quality tests, all ofthe water sector and its significant experience which are described below.working in peri-urban and rural areas of Luwero.3NETWAS was also well-staffed and delivered a Capacity Buildingcompetitive budget proposal. NETWAS and the Following the initial assessment, WBI developedMWE decided to implement the program in the a strategy to build NETWAS’s capacity tourban and peri-urban areas of Wobulenzi and implement the program and to train communityinvited the Wobulenzi Town Council to become members to use social accountability tools.their local implementing partner. NETWAS’s capacity building and training activ- ities, and their outputs, are described in Box 3.3. The Network for Water and Sanitation (NETWAS) is a nonprofit organization registered in Uganda in 1996 to provide services in the water supply, sanitation and hygiene sector. It is affiliated with NETWAS International, a part of an international training network for water and waste management that supports sector-related activities in developing countries.
4 | Social Accountability Notes Box 3. Activities to Build the Capacity of NETWAS and Local Stakeholders, and Outputs of Those Activities Activities Outputs • Train NETWAS on citizen report cards. • A local consultant (Frances Nsonzi) who had partici- • Provide technical assistance to NETWAS on design of pated in the application of citizen report cards to the the sample in Wobulenzi. health sector in Uganda trained NETWAS on the use • Train NETWAS on the use of community score cards. of that instrument. • WBI hired an international expert (Jakov Svensson) to accompany and advise a local statistician (Johnson Kagugube) in designing a stratified sample for Wobulenzi. • A local consultant (Monica Kapiriri) who had partici- pated on the application of the CRC for the health sector in Uganda in the past trained NETWAS on community score cards. Box 4. Social Accountability Tools Used in the Project, and Their Outputs Accountability tools Outputs • Citizen report cards • Two surveys were conducted more than a year apart. • Community score cards • Two rounds of community score cards were adminis- • Water quality tests tered in each of the six communities selected, with six months between administrations. • The quality of the water provided by both private providers was tested twice, with more than a year between tests. Box 5. Activities Related to Citizen Report Cards, and the Outputs of Those Activities Activities Outputs • Mapping of stakeholders in Wobulenzi’s water sector. • Stakeholder analysis produced. • Household listing exercise to update and corroborate • List of households in Wobulenzi. information from the last census. • Representative sample with two different strata: core- • Sample design. urban and peri-urban households. • Citizen report card design. • Three questionnaires: one for water users, one for • Data collection and data entry. Water Board members, and one for others. • Analysis of results. • Questionnaires collected and entered into data system. • Comparison of survey results. Implementation of Social Accountability Tools coverage and civil society advocacy. The results NETWAS deployed three social accountability of CRC surveys can be used to monitor progress tools in Wobulenzi, as described in Box 4. and can affect policy design and program implementation. The CRCs for the water sector Implementing Citizen Report Cards surveyed users on the quality and availability of water services, including hours of service, CRCs are detailed surveys used to assess public problems in billing and collections, tariffs, opinion. Unlike traditional surveys, however, added costs, rent-seeking by service personnel, they are often accompanied by broad media information dissemination from the provider
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 5concerning service interruptions or repairs, and ability, and transparency. To accommodate theoverall satisfaction. participants, the workshop was conducted in NETWAS deployed the CRCs in August- Luganda, the local language.September of 2008 and again in December of The participants identified the following2009. The activities related to CRCs, and their stakeholders in Wobulenzi’s water sector—outputs, are described in Box 5. community water users, members of the Water Board, service providers, and others identified inStakeholder Mapping. On July 3, 2008, Figure 1.NETWAS convened a workshop in Wobulenzi to The participants were asked to consider theidentify the stakeholders in Wobulenzi’s water connections among the stakeholders, includingsector and their relative priorities, roles, and dependency for services, the payment of rates,responsibilities. election constituencies, and advisory roles. Twenty-nine participants (15 men and 14 Based on participants’ feedback, NETWASwomen) attended the workshop. They included developed a relationship map of the stake-members of the Town Council, members of holders as depicted in Figure 1.the Water Board (the entity responsible for The workshop participants then createdmonitoring local water issues and evaluating an “influence pyramid” to display the relativewater service operators), staff from each of the influence that various stakeholders had overtwo water service providers and the MWE, and decision making and policy formulation inrepresentatives from the local community and Wobulenzi’s water sector. The stakeholders withNGOs. A staff member from NETWAS facilitated high influence included the water authority, thethe workshop, which included a series of collab- town clerk, and the Water Board. Those with loworative break-out groups and a review of key influence included community groups, vendors,concepts, such as governance, social account- local councils, the courts, and the police. The Figure 1. Stakeholder Mapping Produced by NETWAS Workshop Participants Supervises Directorate of Minister, Supervises Technical Support Water Development Water and Environment Unit (DWD) Provides technical and ﬁnancial monitoring and supervision Signs performance agreement Water Authority with authority Reporting Wobulenzi WS NETWAS Technical Support Appoints the Board Reporting and implements authority’s and allocates funding Town Clerk Providing decisions and policies technical NGOs/CBOs advice Supplying water Payment of bills Institutions, Town Water schools, HCs Water Board Engineer Advertising/ Extension Workers announcements Health Assistants Private operator Supervising Comm. Dev. Assistants and planning Mass media, radios, newspapers Community groups Buying water Supplying water Youth & Women’s groups Reporting users’ complaints Payment of bills Providing information Water vendors Selling water Water user community Water user Mobilizing Religious institutions, committees churches and mosques Local councils LC1s Reporting complaints Mobilizing Link the community to service providers Governance and Social Accountability Project Relationships Mapping
6 | Social Accountability Notes relative influence of other stakeholders fell holds received their questionnaires. between the two extremes. Interviewers were trained to select house- holds according to the target sample Training Interviewers to Use CRCs. WBI hired demographics specified by NETWAS and to a consultant with experience in the health replace a household, if necessary, without sector to train NETWAS members on the use of altering the sample’s representation. The CRCs. NETWAS then selected a team to receive following information would be collected: training and administer the CRCs to the local water users and water service providers. The • Characteristics of household members team comprised 15 people: 10 interviewers, • Access to water services 2 supervisors, and 3 data entry officers. They • Availability of water (hours of service) were drawn from staff from NETWAS and other • Household water usage Ugandan NGOs and community members from • Cost of water Wobulenzi. • Water pressure Team members were encouraged to • Quality of service delivery recognize the importance of CRCs in identi- • Quality of water (such as color, smell, and fying problems in the water sector and taste) improving water service delivery by both • Gender issues (for example, the dispro- private and public providers. Data entry officers portionate share of women responsible for understood the importance of quickly and collecting water) accurately entering data. • Community participation in service delivery NETWAS organized a four-day training • Information dissemination on water services session (August 21–25) for interviewers and field • Characteristics of respondents supervisors. WBI hired an international expert to facilitate the training and develop an interview Designing a Representative Sample. A repre- manual for participants. Training modules sentative sample of stakeholders in Wobulenzi’s included (i) an overview of CRC processes; (ii) water sector was selected to receive CRCs. roles of the interviewer; (iii) selection criteria for NETWAS divided the stakeholders into two households and replacements; (iv) instructions categories: (i) core urban and (ii) peri-urban. The for conducting interviews; (v) fieldwork exercises core urban segment comprised households in to help interviewers become comfortable Wobulenzi East, Wobulenzi West, and the zones with the process; and (vi) guidance on how to of Luzzi and Katale in Wobulenzi Central, the process feedback from field practice. The three most densely populated areas of the town. The individuals hired for data entry received training peri-urban segment comprised households in from August 26–29, ensuring that data entry Katikamu, Bukalasa, and the zones of Kikoma could begin immediately after the first house- and Kikasa in Wobulenzi Central. Box 6. What Constitutes a Household? According to the Government of Uganda, a household definition. For example, if the head of a household is defined as a group of people who normally live has more than one spouse, with more than one line of and eat together. These criteria are important: the children, but the entire group lives and eats together, government uses this definition as part of its census. the groups comprises a single household. However, if Households include families and are generally each spouse—each with a line of children—lives and comprised of some combination of a head (male or eats separately (regardless of whether the head of the female), a spouse, children, and perhaps relatives and household travels back and forth between them), then visitors. Where two or more people with separate living this family comprises more than one household. A single arrangements simultaneously occupy the same dwelling, household may also consist of one person who lives they are treated as separate households. Cultural varia- and eats alone, or may otherwise comprise a group of tions have also been addressed by the household unrelated people who live and eat together.
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 7 Credible survey results depend on a reliable sponding to the peri-urban segment, includedrepresentative sample. WBI therefore ensured 36 households per zone or village.that NETWAS relied on two experts to design Each trainer received a copy of two letters ofthe sample. NETWAS hired one expert, a local introduction: one from NETWAS Uganda to thestatistician with significant experience in census Wobulenzi Town Council; and another from theadministration. As for the other expert, WBI Wobulenzi Town Council to the local communitycontracted Jakob Svensson, a Swedish expert chairpersons of selected wards informingwith experience designing samples in Uganda. them of the purpose of the survey. To ensureHe also ensured that the sample design household privacy and prevent discrimination,complied with international standards. House- neither the local community chairperson nor hisholds were defined as described in Box 6. or her representative was permitted to partic- ipate in the household interviews (unless in theirListing Wobulenzi Households to Populate own household).the Sample. At the start of the program, 14villages in Wobulenzi were randomly chosen by Designing the CRC Survey Questionnaires.the statisticians to participate. The supervisors While the interview team compiled a repre-and interviewers were tasked with developing a sentative sample of Wobulenzi’s households,comprehensive list of households in each of the NETWAS and WBI developed separate14 villages, from which random samples would questionnaires to target the key stakeholdersbe selected to participate in CRC surveys. This in the water sector in Wobulenzi: water users,process took four days. The interviewers visited water providers, and the Water Board. Foreach household to ensure the accuracy of the example, the household questionnaire includedfinal list, which was going to be compared to more questions concerning possible feedbackthe census, and carefully screened the list for mechanisms that would enable water usersomissions and duplications. They then assigned to participate in decision-making and identifya random number to each household. Each obstacles to effective service delivery. Theinterviewer drew a series of random numbers questionnaires administered to water servicesto generate the sample. The idea behind the providers emphasized training and capacityhousehold listing was to verify the data from the building of staff and engineers, while thosecensus and to update the numbers using statis- administered to Water Board members focusedtical approximations. The final household list on communication mechanisms used to informincluded the number of households in each of water users and monitoring mechanisms used tothe 14 villages. Household data from Wobulenzi’s supervise the water provider.2003 censuswere obtained Table 1. Results of the Household Listing in Comparison to the Census of 2003for purposes ofcomparison (Table 1). Village Listed households 2003 census figures For each sample Gwafu 74 46segment (core Morden 155 162urban or peri- Katale 234 161urban), seven zones Kigulu 265 202or villages were Bukorwa Central 126 243randomly selected Upper West 34 55for sampling. For Upper East 21 North Central 15each zone or village Lutamu 150 206selected in Stratum Luzzi 351 352I, the one corre- Nakadingidi 828 533sponding to core- Kikasa 126 151urban Wobulenzi, Katikamu Proper 175 16733 households were Kitante 252 235randomly selected. Kikoma 270Stratum II, corre-
8 | Social Accountability Notes Throughout the survey design process, WBI driven monitoring and evaluation. Similar to and NETWAS held lengthy discussions with the the CRC process, the CSC seeks to foster town clerk, water engineers, community devel- social accountability and responsiveness from opment officers, members of the Water Board, service providers. The difference is that CSCs and community members to identify focus areas also encourage service providers to meet for each questionnaire. The questionnaires with members of the community to facilitate were piloted in Bombo, Luzira, Luwero town, immediate feedback and foster grassroots and Mukono. The household questionnaire was empowerment. translated into Luganda, the local language. These CSCs were used to enhance stake- holder awareness of governance challenges Conducting the CRC Survey. The household in Wobulenzi’s water sector and to promote CRC surveys were administered over a 10-day partnerships among stakeholders to respond period from August 29 to September 7 in to these challenges. Their use enabled the partnership with the Town Council, local community to provide detailed feedback on councilors, and Water Board members. The water services by both providers. Community survey team comprised nine research assistants, members were trained to select from a variety of two field supervisors, and three monitoring quality indicators and shown how to use them to specialists. From among all those who had score water services from each provider. participated in training sessions, the interviewers The interface between users, service were selected for their ability to communicate providers, and local authorities was an and their understanding of the tool. important feature of the CSC process; it allowed In total, 632 households (as defined in Box 6) community members to voice complaints, were interviewed. Five replacement households concerns, problems, and suggestions to were also selected for each village, although improve the quality and coverage of water this figure increased to 10 when field workers services. The CSCs facilitated a constructive found that some residents had either vacated dialogue among a cross-section of stakeholders their homes between the finalization of the in an effort to raise awareness of problems to be household list and the launch of the survey corrected and achieve consensus on the impor- process, or could not otherwise be located. tance of community participation in water sector The survey team also interviewed Wobulenzi’s reforms in Wobulenzi. two service providers and eight members of its Water Board. Training Participants to Use CSCs. NETWAS Data entry officers entered the results into a convened two training sessions to train the matrix between September 9 and September Wobulenzi Town Council and other water NGOs 20, and the data were subsequently analyzed on the use of CSCs. The first session lasted by NETWAS and WBI. More than a year later, in from February 7 to 13, 2009, and the second December 2009, the partnership administered from February 23 to 27, 2009. WBI contracted a second round of CRCs to track progress in an international expert on CSCs to facilitate the stakeholder actions in Wobulenzi’s water sector sessions and to support NETWAS during the and evaluate changes in stakeholder’s opinions. training. The training sessions taught selected partici- Implementing Community Score Cards pants, especially those from Wobulenzi, to understand the differences between CRCs and In addition to the 2008 and 2009 CRC surveys, CSCs and to interact effectively with diverse NETWAS deployed CSCs to facilitate dialogue stakeholders, including the local community, the among the various stakeholders identified Water Board, and water service providers. The earlier in the pilot. CSCs would bring those stakeholders together A CSC is a qualitative monitoring tool used for the first time to candidly discuss the quality for local-level monitoring and performance of local water services. NETWAS also instructed evaluation of services by communities. The the interviewers to encourage stakeholders to CSC process is a hybrid of the techniques use fieldwork and practice sessions to identify, used in CRCs, social audits, and community- implement, and sustain water sector reforms.
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 9 The first training session consisted of two Trandit or Bukalasa. During these sessions,days of classroom training, followed by three community participants (representing waterdays of practice fieldwork to acclimate partici- users) were asked to prioritize three to fourpants to the CSC process. Participants included types of water services requiring improvementseven members of NETWAS; five community within six months. Representatives from Tranditfacilitators from Wobulenzi; seven representa- and Bukalasa participated in CSC meetings.tives from the Wobulenzi Town Council; one They were asked to evaluate their respectivemanager from Trandit; six representatives provider’s service delivery and identify areas forfrom Bukalasa College; and a small number of improvement.visitors, including representatives of the MWE, Meetings brought service provider repre-the World Bank’s water sector staff in Uganda, sentatives together with community represen-WaterAid (a well-known NGO), Uganda’s tatives from Sikanusu, Upper East and WestCommunity Development Facilitation Unit, Luwero, Kikasa, Gwafu, Kitante, and Kigulu toLuwero District authorities, and NETWAS’s share perspectives and receive feedback. Priorpartners, such as the Buso Foundation. to this, Wobulenzi Town Council had no formal The second training session, held two weeks mechanisms for involving the community in thelater in Kampala, sought to increase awareness planning, implementation, and monitoring ofof water sector issues and create a climate of water services. Results from a 2008 CSC exercisetrust among water sector stakeholders. Most for the water service provider in Sikanusu areparticipants from the first training attended the presented in Table 2.second session. NETWAS therefore conducted The priority areas identified by stakeholdersfollow-up exercises designed to strengthen were next compiled into a comprehensive jointskills developed in the first training. The original action plan, which was agreed to at subsequentparticipants were joined by members of the meetings.Water Board of Wobulenzi Town Council, staff Some of the action items from the jointof both water services providers, a Community action plan were executed quickly. ForDevelopment Officer from Luwero District Local example, in the Kikasa community, membersGovernment, and staff from the MWE. A total of requested telephone contact particulars fortwenty-seven participants attended the second key personnel from the water service providerworkshop at Bukalasa. These contact particulars were promptly provided. Quarterly action-learningAdministering the CSCs. To facilitate dialogue meetings were also held to facilitate partici-among the stakeholders in Wobulenzi’s water patory reflection and learning processes,sector, NETWAS administered CSCs in six assess progress, fine-tune activities undercommunities that receive water either from development, and highlight those activitiesTable 2. Example of a Community Score Card Developed by Water Users in SikanusuAreas for improvement Desired changes Score % Reasons for the score Proposed activitiesSystem machines are old New and functioning machines 40 Frequent breakdown of To ask for new pumps from theand some pumps are non- machines ministryfunctioningUnstable power supply and Uninterrupted power supply 35 Power supply is very inconsistent Increase on budget allowancehiking price of fuel for preparedness when power is offPoor payment of water users Good and timely payment 40 High percentage of defaulters To ask water users to pay on timeTransparency Openness 60 Most information is To be transparent in all acticities communicatedSocial accountability Timely reporting to the water 45 Roles and responsibilities are Reporting and accounting of all board not clear. responsibilitiesDialogue Frequent dialogue with water 48 Limited communication and Create avenues of users dialogue between water users communication with water users and service providers
10 | Social Accountability Notes completed successfully. Other action items have to achieve development goals. Recently, seen no movement. however, two-way approaches that engage NETWAS held a second round of CSCs six stakeholders and empower them to voice months after the first. The same representatives their opinions and identify issues important from the community and the water providers to them have become the accepted form of participated to follow up on the action plan and communication for achieving sustainable devel- to determine whether the service providers had opment results. NETWAS, WBI, and the World improved water delivery. Bank’s CommGAP program jointly oversaw the development of a communication strategy for Testing Water Quality Uganda’s water program. The strategy aimed to foster trust among water sector stakeholders NETWAS conducted water quality tests and facilitate dialogue and knowledge sharing in September 2008 and December 2009, regarding the importance of social account- coinciding with CRCs. The following measures ability and transparency. of water quality were tested: A CommGAP communication specialist accompanied WBI staff on two missions to • pH, which affects the taste and corro- Wobulenzi to collaborate with a local consultant siveness of the water. on a communications assessment. NETWAS • Turbidity, which indicates the cloudiness of used the reports prepared by CommGAP to the water and affects the risk of infectious develop a strategy for informing stakeholders disease transmission. of priority issues in the water sector and solic- • Electrical conductivity, which affects the iting feedback about areas in need of further taste and freshness of the water. improvement. Furthermore, to ensure the • Fecal coliform, which indicates recent sustainability of improvements, stakeholders fecal pollution and the potential risk of were encouraged to continue to exchange ideas contracting infectious diseases. after the completion of the program. • Total coliform, which affects the general NETWAS, WBI, and CommGAP used various hygienic quality of the water. channels to share knowledge and perspectives on the program with diverse local and national Nine tap stands, four water tanks, and audiences and to disseminate and explain the the main outlet of Bukalasa were tested. In results of surveys and water quality tests. Those Wobulenzi Town Council, a total of four kiosks, channels were: two tanks, four tap-stands, and four boreholes were tested. Finally, in Sikanusu zone, one • Regular meetings with stakeholders, unprotected spring in Wobulenzi (used by a including officials from the MWE, local sizeable portion of the population when the flow government authorities, water service from tap stands is irregular) was also tested. providers, and users’ associations Water sources were first tested on-site for • A bulletin developed by NETWAS in English physical quality using portable electronic and Luganda meters. Samples were then collected in • Posters in high-traffic areas sterilized glass bottles for laboratory testing • A blog to share real-time information with supervised by NETWAS. Community focus stakeholders groups were convened by NETWAS to • A Facebook account to share program infor- raise public awareness of the importance of mation with other donors and water experts maintaining the cleanliness of water points to • A Web site created by WBI to provide easy avoid contamination, and to demonstrate how access to all materials generated by the to collect, transport, and store drinking water. project, such as surveys questionnaires and results, reports, and evaluations Communication for Governance • A video produced by NETWAS-WBI In the past, the World Bank and other devel- • Local forums to host presentations on the opment organizations relied on one-way infor- project, including during the quarterly mation dissemination and communication action-learning meetings convened by
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 11 Box 7. Monitoring and Evaluation Activities and Outputs Activities Outputs • Routine administrative data recording • Financial records of expenses • Process reporting • Quarterly reports • Process documentation • A website that documents the project over the • Activity reporting process • Team meeting • Specific report for each major activity • Regular inspections/spot checking • Minutes from each meeting • End of project report • Field visits by the M&E coordinator • Final report and case study about the program the MWE, joint sector review meetings Results and Outcomes convened each September, and water integrity workshops held in Kampala Results of CRC Survey. The results of the 2008 • Television, radio, and print media, including and 2009 surveys indicated that water provision the Uganda Broadcasting Council (a national in Wobulenzi improved after the deployment television program) and the national of social accountability tools. This was despite newspaper, to disseminate survey and test the malfunction of two water pumps, which results prevented Trandit from providing water for two months before the administration of the 2009Monitoring and Evaluation survey, until the MWE intervened to procureNETWAS and WBI collaborated to develop a replacement parts.monitoring and evaluation (M&E) strategy for Notwithstanding this difficulty, overall stake-tracking stakeholder feedback and water sector holder satisfaction improved significantlyimprovements. M&E activities enabled the between 2008 and 2009. Many households usedpublic to sustain progress after completion of more piped water than they had before, partlythe program (Box 7). because Trandit added 43 water tap access An important aspect of the M&E process was points to its piped water infrastructure. Thethe careful documentation of program activities, number of users encountering difficulties inwhich will inform subsequent analyses and accessing water—such as children harassed bypromote sustainable outcomes. adults competing for water at access points— decreased significantly. Users also reported that improved communication with service providers had increased the transparency of the costs of certain water services, such as connectionsTable 3. Improvements in Water Service, by Provider Trandit BukalasaVariables 2008 2009 2008 2009Percentage of households using piped water 22.4 34.9 40.4 50.0People encountering long queues, wet season 37.0 25.8 23.9 0.0People encountering long queues, dry season 73.1 63.9 47.5 13.8Harassment of children by the adults at the water source, wet season 7.0 3.6 0.0 0.0Harassment of children by the adults at the water source, dry season 14.7 4.8 0.0 0.0Average cost of connection to the piped water system (U Sh) 4,220 65,310 10,280 40,000Percentage of households satisfied with the quality of water services 69.9 82.2 81.7 94.6Percentage of households satisfied with the quality of water (somewhat satisfied, satisfied, and 80.2 92.1 86.2 97.0very satisfied)
12 | Social Accountability Notes to the piped system. Finally, service providers began to adjust their practices to improve Figure 2. Performance toward Sikanusu Priorities services in response to public feedback. Table 3 70 summarizes the improvements in water service provision for each of Wobulenzi’s service 60 providers based on household survey results. 50 40 Results of CSCs. The service providers quickly 30 executed several items from the joint action Baseline 20 Evaluation plan. In the Kikasa community, for example, members requested—and promptly received— 10 telephone contact information for key 0 personnel at Bukalasa. Quarterly action-learning Old Poor payment Unstable equipment of bills power meetings were also held to facilitate partici- Priorities patory reflection and learning, assess progress, fine-tune activities under development, and highlight successfully completed activities completed. Other items from the action plan have received no attention. Figures 2, 3, and 4 illustrate progress in addressing the priority areas in the water sector Figure 3. Progress against Community over the six months since the first CSCs. In Priorities in Kigulu each of the six communities participating in 25 the follow-up survey, participants received a 20 Baseline summary document detailing priority areas 15 Evaluation identified during the 2008 survey, indicators devised to measure progress towards these 10 goals, initial scores, explanations of initial 5 scores, proposed activities to improve scores, 0 progress as of the most recent quarterly action- Water Costs Old quality equipment learning meeting, and any information from Priorities more recent CSCs, such as new scores. Each community has demonstrated progress in the water sector in short time between the first CSC and the second. In the community of Sinakusu, for example, priority areas for improvement included old equipment, poor payment of bills, and unstable Figure 4. Progress against Upper West power. The results of the second CSC indicate Priorities that the community’s perception of each of 70 these indicators had improved. Similarly, in Baseline 60 Kigulu, priority areas included water quality, Evaluation cost, and old equipment; in Upper West 50 Wobulenzi, priority areas include old pumps, 40 insufficient water coverage, water treatment, 30 and billing problems. 20 In both the 2008 and 2009 CSCs, each 10 community evaluated the pillars of gover- nance—transparency, accountability, and 0 Old pumps Insufﬁcient Water Billing communication—in their local water sector. The water coverage treatment problems CSC results indicate that users saw improve- Priorities ments in each of these areas.
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 13Results of Water Quality Testing. The results nities to conduct routine maintenance and toof the first of the two water-quality tests clean boreholes on their own.revealed that some water outlets, notablyreservoir tanks and several tap stands, were Less Corrosion in Water Source. Althoughcontaminated with fecal matter or contained the pH of water samples in Wobulenzi was low,turbid water. During the CSC interface boreholes did not show signs of corrosion. This ismeetings, corrective measures were suggested, possibly a result of routine maintenance and thesuch as increasing the frequency of tank continuous movement of water. Water pH alsocleanings and repairs of leaking pipes. In 2009, improved between the first test and the second.the testers revisited the original water points to The follow-up tests revealed that sometrack improvements in water quality. The results water points in Wobulenzi had stoppedof the follow-up testing suggest that water functioning altogether. For example, the servicequality had generally improved. provider disconnected one users’ tap stand for non-payment of tariffs and two boreholes had Less Fecal Contamination. The first test ceased functioning.revealed that water samples from a cylindrical Table 4 summarizes the improvements inwater tank in Kikasa and from tap stands in water quality between the first and second tests.Bukalasa and Kitante were contaminated with Although tests results indicate overallfecal coliform. Although the unprotected improvements in water quality by both servicespring was still contaminated by the time of providers, communities still need to increasethe second test, the remaining water points their awareness of proper techniques for waterwere found to pump safe drinking water. transport and storage. Based on observationsImprovements were due to more frequent tank at the water sources, the jerricans that mostcleanings in Bukalasa and increased awareness people use to collect water, were contaminatedof the importance of covering tap stands to with algae.prevent contamination. Information dissemi- The success of the water program innated during the CSCs also motivated commu- Wobulenzi demonstrated how social account-Table 4. Results of Tests of Water QualityParameter Trandit BukalasapH The average pH value tended to rise toward neutrality. The There was also a rise in the average pH from 5.99 to 6.44. average pH was 6.98, compared with the previous pH value The pH value of the source outlet was 6.40, indicating no of 6.41. The neutral pH value is 7.00. The recommended pH significant change along the distribution system. range of untreated water supplies is 5.0–9.5Turbidity Average turbidity was 1.98 NTU compared with the prior value Average turbidity was 1.59 NTU. Previous turbidity was 1.88 of 2.11 NTU. This was an indicator of an improvement in the NTU, also signifying an improvement in the clarity of water. clarity of water. The maximum recommended turbidity value in untreated water supplies is 30 NTUConductivity/TDS The average TDS value was 167.8 mg/l compared with the The average TDS value was 127.4 mg/l compared to the previous value of 148.4 mg/l. However, there was no significant previous value of 112.3 mg/l. However, there was a difference change in the value from the sump, which was 164 mg/l. The with the source value of 109 mg/l. maximum acceptable DS value in untreated water supplies is 1,500 mg/lFecal coliform (e. coli) All the results were satisfactory. The maximum acceptable e. All results were satisfactory, in contrast to previous tests, where coli concentration in untreated water supplies is 50 cfu. unsatisfactory results were obtained in the cylindrical water tank and Sakaza tap-stand.Total coliform (T. coli) All water points showed satisfactory results. Previous All results were satisfactory. Previous results showed results showed some contamination in several tap-stands— contamination in the elevated cylindrical water tank and Semaganda, Florence Nabowa, and Betty Nakajubi. The Sekaza tap-stand. maximum acceptable concentration in untreated water supplies is 100 cfu. Katongole Expedito Kiosk The supply is located in Luzzi zone. All the results were satisfactory. Unprotected spring Results remained unsatisfactory. Boreholes Results from all 4 boreholes were satisfactory.
14 | Social Accountability Notes ability tools helped stakeholders to build The water program significantly improved the effective partnerships and improve communi- relationships among Wobulenzi’s various water cation. They also allowed water users to provide sector stakeholders, including the Water Board, feedback to their water service providers for the the water service providers, and the community. first time. Even though Wobulenzi’s two service Water users can now voice complaints at regular providers operated under different types of stakeholders meetings, and stakeholders are contracts (Bukalasa as a private service provider encouraged to respond to user feedback. and Trandit under an output-based aid contract The participation of Water Board members with the MWE), both benefited from social was critical to the CSC training sessions. accountability tools to significantly improve Chairman Stephen Sawa Loboowa and Ms. service provision and water quality. Sarah Nagujja—two of the participating Board In 2009, the MWE identified a list of “golden members—were very supportive, assuming indicators” for service delivery in the water facilitation roles and lobbying for transparency sector to provide a standard for evaluating and inclusion of users in the water sector. After the success of water programs. The golden the program launch, they were also instrumental indicators for Wobulenzi’s two water service in convening the first Board-driven community providers are summarized in Table 5. Data was meetings in Wobulenzi. Within six months, they gathered in interviews, the two program surveys, had visited ten of Wobulenzi’s 20 communities CSCs, and field visits. to collect feedback on water services. Table 5. Improvements Measured against Indicators of the MWE, by Provider, June 2008–August 2009 Indicator Trandit Bukalasa Access In-house connections increased from 25 in June A new pipeline is being extended in Kikasa Number of people connected to piped 2008 to 28 in August 2009. Public standpipes community to supply more households. water during the period of the program increased from 32 to 33 and yard taps from 471 to 514. Number of people with in-house 25 28 22 22 connections Number of standpipes providing water 68 61 (some closed 0 0 because not sustainable) Number of kiosks providing piped water 32 33 (one closed because 0 0 of insufficient clients) Number of yard taps providing piped 471 514 66 66 water Functionality Two back-up generators and two new batteries The college had completed repairs of one pump Water sources improved during the period were purchased. A unit called the “change-over” and was working on the second one at the “fish of the program was purchased to ease the interchange between pond.” Work was to be complete by February 2010. the national power grid and generator power The college had already procured the materials for supply. A pump to move water from the borehole to the second pump, which participants reviewed at the reservoir was repaired. the end of the Action Learning meeting. Investment More than US$2,000 spent in repairs by the Town More than U Sh 5,000,000 (US$2,688) was invested Amount of money invested in the water Council (Water Board). Trandit has invested more in repairing 2 pumps that were not functioning. The schemes during the period of the program than U Sh 20,000,000 in new connections and rest of the repairs required an additional U Sh 25 extensions to the poor communities. million (US$13,440), which the college did not have at the time. Quality of water The quality of piped water improved in general. The quality of water from all sources improved. All Improvement in quality based on samples water reservoirs that had high e. coli values were analyzed comparing first quality test with cleaned. Covers were installed and leaks plugged. second water quality test, that complies with national standards. Quantity of water During the program one pump was repaired, At least two pumps are now functioning. During the Increase in cumulative capacity of reservoirs increasing volumes. It is still operating below program two pumps were repaired, thus increasing and other storage means during the period capacity. water volumes in reservoirs. Current capacity is 220 of the program. m3 but pump supplies 190 m3, up from 10.5 m3 and 4.5 m3 per day, respectively, at project start. One pump needs to be repaired Equity Output-based aid is an approach to reach poor The villages of Kikasa, long neglected, saw Sub-county deviation from the district people. Subsidies go only to connect people who excavation of pipelines begin. Villagers hope to be average in persons per improved water cannot afford to connect. During the program, the connected to the main line. point. poor communities of Kitante, unconnected to the system for many years, obtained 16 connections.
Improving Governance in Water Supply through Social Accountability, Communication, and Transparency in Wobulenzi, Uganda | 15 The water program marked the first time that provider to a private provider. The servicea combination CRCs and CSCs were deployed provider, the local government, and the MWEin Uganda’s water sector. The first round of are still learning how to use subsidies to betterCRCs provided baseline data to monitor respond to the needs of the poorest withinimprovements in quantitative and qualitative a private sector scheme. With another year,indicators of water quality and services. The Trandit may learn to better respond to stipula-CSCs employed communication techniques tions in the OBA, and the differences betweento engage participating communities and the two providers may become clearer.Theempowered them to provide feedback and MWE may also have to provide further trainingbecome partners in efforts to improve water to service providers to increase their awarenessservices. Water service providers were likewise of the results-based terms and conditions oftrained in the use of communication tools and OBA contracts. In April 2009, for example, onecame to understand the importance of using of Trandit’s backup pumps broke, and waterfeedback from water users to improve their could no longer run through the Trandit networkservices. Local authorities, including the Water in Wobulenzi. Based on the OBA contract,Board, were encouraged to use social account- Trandit was responsible for maintaining its infra-ability tools to promote good governance. structure and replacing any broken parts. Trandit NETWAS was trained to use CRCs and CSCs did not understand its contract, however, andto monitor water quality and the provision of immediately requested that the MWE replacewater services. NETWAS is also a member the broken part. Because the water system wasof UWASNET (Uganda Water and Sanitation obsolete (more than 10 years old), the MinistryNetwork), an umbrella NGO that promotes could not find a replacement part in Uganda,cooperation and knowledge sharing among and although it was not responsible for the150 NGOs and community-based organizations repair, it was forced to buy a replacement partworking in Uganda’s water and environment from Sweden. It took more than two months forsectors. Through UWASNET, NETWAS is sharing the part to be replaced, and water did not run ininformation and lessons learned from the Wobulenzi during this time.Wobulenzi water program with other local NGOs, Despite the fact that Trandit neglectedboth online and through participatory learning its responsibility to maintain its network, theactivities, such as seminars and workshops. Ministry of Water did not levy a penalty against the service provider. OBA contracts shouldLimitations and Recommendations therefore incorporate terms that specify whatWBI and NETWAS originally expected that constitutes a breach of responsibilities, alongbecause of incentives in its OBA contract, with clear penalties for breach, such as theTrandit would improve its level of customer payment of damages. The Ministry of Watersatisfaction more than Bukalasa, which did should inform service providers of these termsnot have an OBA contract. Based on the data, before entering into an OBA contract. Servicehowever, this does not seem to be the case. providers should be sufficiently prepared andSatisfaction with both providers improved financially stable to quickly handle repairs andalmost identically. Except for the 47 new yard maintain service to users without depending ontaps announced by Trandit, which will soon be the Ministry.functioning, there is no statistical evidence so The government of Uganda should institu-far that Trandit improved more than Bukalasa tionalize the social accountability tools usedduring the implementation of this program. in the water program. Cooperation betweenThe NETWAS-WBI team believes that one year centralized government agencies and represen-is not time enough to evaluate the impact of tatives from specific service sectors would allowan OBA contract on the provider performance. the tools to be developed for a wide rangeAnother year of monitoring of Wobulenzi’s of sectors at low cost. For example, Uganda’sservice providers would provide a better basis statistical bureau could consult with a represen-for evaluation. Implementing OBA contracts tative from the MWE to design surveys, allowingrequires a change in behavior from various the MWE to avoid duplicate costs associatedstakeholders in terms of moving from a public with developing surveys in cooperation with