WATER GLOBAL PRACTICE WSS GSG UTILITY TURNAROUND SERIES Case Study— SISAR Ceará, Brazil Wilson Dos Santos Rocha and Maria Salvetti AUGUST 2017 Key Characteristics of Aggregation Case Study SISAR CEARA, BRAZIL Context • Upper-middle-income country • Aggregation covering rural areas • Low level of WSS performance Purpose Access to water provision, performance, solidarity, economic efficiency Scope Water functions and services Scale • Watershed limits • Localities covered: 153 for water • Population covered: 89,500 inhabitants for water • Connections: 25,548 for water • Network length: 415 km for water Process Bottom-up with financial incentives Governance • Special-purpose vehicle • Private association • Decision making: General assembly in which each affiliated association has one vote • Asset transfer: Operating assets owned by the state, non-operating assets (office, workshop, vehicles, and maintenance tools) owned by SISAR • Liability: No liability as there was no service before • Staff transfer: No staff transfer as there was no service before • Clear entry and exit rules Outcome • Positive with financial sustainability for operation Findings • Successful model of aggregated utility to provide service in rural areas which has been duplicated in other parts of the country, close relationship between communities and their water associations, community-based labor force hired part-time, performance-based monitoring 1 In 1996, when SISAR was created in the Brazilian A Community-Based Aggregation Model state of Ceará, its main purpose was to bring access to to Provide Access to Water in Rural water supply in rural areas where such service was Locations not available before. These areas had been left aside The state of Ceará spans an area of 148,920 km2, with by state water supply and sanitation (WSS) compa- 184 municipalities and an estimated population of 8.9 nies, which had predominantly focused on urban million inhabitants in 2016. CAGECE, the state water access to WSS services. From 1996 to 2016, SISAR’s and sanitation company, supplies water to 151 munic- coverage expanded dramatically, from 18 to 153 local- ipalities and sewerage services to 73. The 33 remain- ities it reached operational cost recovery in 2012. ing municipalities have their own services. The WSS This success, which has encouraged the replication sector institutional is composed of a state-level regu- of the SISAR model in other rural contexts in the latory agency called ARCE (the Regulatory Agency for Brazilian states of Piaui and Bahia, is mainly based Public Services in Ceará) which regulates energy, on a gradual improvement strategy and a specific water, and sanitation activities; the Superintendent labor arrangement. of Hydro Works (SOHIDRA); the State Secretariat of Agrarian Development (SDA); and the State Secretariat of Cities. National Policy for Urban WSS Aggregation Leaving Aside Rural Areas The process that gave birth to SISAR did not originate from a sector reform policy enacted nationally or at In the early 1970s, Brazil implemented a major ser- the state level, but from a state government invest- vice aggregation reform through PLANASA—the ment program supported by international financial National Sanitation Plan. Municipal governments— institutions. What prompted the state government to which, until then, had been playing the role of create the SISAR model were the conclusions that service  providers—began to delegate these services ­ water and sanitation investments in rural areas were to state governments, encouraged by the criterion no longer effective and that they would need substan- to  access federal investments predominantly tial improvements and replacements in the very short through these companies. Between 1970 and 1983, term. In this context, SISAR was originally set up investments in the sector reached an average of in  1996 in the municipality of Sobral, aggregating 0.46  ­percent1 of GDP. But PLANASA was affected 18  water users associations. Its creation took four by the global financial crisis starting in 1983. Between years. SISAR required an intense participatory pro- 1980 and 1990, investments in the sector dropped cess, as every family in a community had to agree to 0.24 percent of GDP, then to 0.16 percent during before its association would join. Many debates also the 1990s; in 2012 they reached 0.11 percent—the took place with municipalities that initially resisted lowest level yet. In Brazil, for the purposes of ­ the idea but later came to support it. SISAR/Sobral WSS service provision, communities located outside was so successful that, in 2001, the state government city borders are considered rural. This represents decided to expand the model—through CAGECE—to a significant portion of the population that remains seven additional regions, one for each river basin in unserved by state companies, which focused on the state. urban areas, serving municipalities’ headquar- ters  and their surrounding areas. This institu- The eight SISARs in Ceará are autonomous legal tional  vacuum of water and sewerage services in ­entities—non-profit associative organizations—­making rural areas gave rise to aggregation models such their own decisions, having their own governing ­ bodies, as SISAR. and managing their own operations and finances. 2 Case Study—SISAR Ceará, Brazil They  are based on community associations that join of SISAR did not cause conflicts over asset ownership, together to share service delivery—some of the activi- since the  communities that join usually did not have ties being carried out by local associations and some by any service at all. SISAR—to ensure the proper operation and mainte- nance of water supply systems. SISAR is presided over The number of locations gathered under SISAR by a general a ­ ssembly, in which each affiliated associa- Sobral (which was renamed SISAR/BAC for the name of tion has one vote. It also has administrative and fiscal the Acaraú and Coreaú river basins) reached 74 in 2006 ­ councils. Its executive structure includes (a) a board and 153 in 2016. It presently provides water to 25,548 made up of the presidents and chairpersons of the affil- connections (family households), equivalent to 89,500 ­ iated associations; (b) a team hired and paid to provide communities inhabitants. SISAR usually operates in ­ billing and collection services, as well as systems main- aggregation ­ with 25 to 800 families. The scope of ­ covers tenance and social support to management; (c) a resi- the entire water supply process—abstraction, treatment, dent hired by each local association to carry out daily network and c and distribution through ­ ­ onnections, and systems operation. Operating assets are owned by the administrative, financial, and all associated functions—​ state, whereas non-operating assets (office, workshop, operational activities, with the latter being shared with vehicles, and maintenance tools) belong to SISAR. local associations. Experiences in Ceará with sewerage CAGECE provides  technical, training, and business solutions have so far been incipient and have focused management support to SISAR. The aggregation model on only on individual households. MAP 1. Service Area of SISAR Ceará Case Study—SISAR Ceará, Brazil 3 TABLE 1. Evolution of the Number of Locations and maintenance capacities. In order to monitor the Served by SISAR Ceará achievements of SISAR/BAC, some performance indi- Year Number of aggregated locations cators were selected and targets were set. They encom- 1996 18 pass quality, technical and financial elements: 2001 30 • A water quality target of 95 percent was set in refer- 2006 74 ence to a water quality index; the recorded perfor- 2011 101 mance fluctuates monthly, with values ranging from 2016 153 65 percent to 91 percent. • An asset recovery target of 63 percent of the existing The process of aggregation was voluntary and largely systems has been defined; the level achieved so far incentivized, as communities and their associations reaches 33 percent. joined SISAR in order to obtain funds to invest in WSS • The continuity of service is monitored, with a equipment. Generally, investment comprises the state- target of 24 hours per day. The values currently ­ level rural development program financed by the World range from 6 to 15 hours. Bank (PDRS, the São José Project), which dates back to 1995. Other investment sources have also contributed • Leakage is not monitored at present as macrometers to SISAR aggregation, such as funds from the federal have not yet been fully installed. However, it should government made available through FUNASA—the be noted that the communities are 100 percent cov- National Health Foundation–as well as the Ministry ered by micrometers. of  National Integration and Kreditanstalt für • A billing ratio target has been set at 20 percent and Wiederaufbau (KfW). Entry into SISAR aggregation was currently reaches 18 percent. done on a voluntary basis from 1996 to 2010. Since • A collection ratio objective has been set at 100 per- then, it became mandatory for associations that cent and currently reaches 96 percent. wanted to join to have access to investment funds. This performance monitoring, which is carried out by TABLE 2. Sources of Funding for Investment Projects CAGECE in its role as management supporter, proved to within the SISAR Ceará Service Area be one of the key factors contributing to the model’s effi- ciency. Indeed it allows a sequenced strategy, spreading Number of Share of Investment source communities total (%) efforts and changes over time, thus not burdening the State program with World Bank 98 64.1 service provider with having to do too much too quickly. loan It also proved very useful to report to member associa- State program with KfW loan 27 17.6 tions on progress being made, thus facilitating account- Federal program through FUNASA 24 15.7 ability and engagement with local Communities. Federal program through Ministry of 4 2.6 Integration Operational Sustainability Achieved after 15 Years, Based on Staffing Arrangements A Gradual Improvement Strategy Although the primary aim of the aggregation was pro- Monitored through Performance Indicators vision of water access, aggregation also allowed econ- Prior to joining SISAR, most localities had no water sup- omies of scale when purchasing water treatment ply. As such, SISAR aggregation was primarily aiming at products and performing equipment maintenance. water provision, thereby improving population health, It  also allowed cross-subsidies, which benefit and a higher service quality with guaranteed operation households in smaller localities. However, the SISAR ­ 4 Case Study—SISAR Ceará, Brazil ­aggregation model was not always financially sustain- aggregated utility, SISAR. The target for the water qual- able. In addition to the investment subsidies received, ity index was set at 95 percent; best performance cur- SISAR/BAC also received funds from CAGECE to cover rently fluctuates between 65 percent and 91 percent. operating costs. These were used to pay for part of the The continuity of service provision target aims at 24 technical staff and vehicles. These operational subsi- values range from 6 to 15 hours. hours; present ­ dies were gradually phased out and were completely stopped after 15 years. Data from 2016 show the finan- Strong Citizens Engagement and Clear Accountability cial sustainability achieved by SISAR/BAC, which gen- Improve Support for Aggregations erates an operating surplus of 9.6 percent (US$1.78 Accountability mechanisms, embedded in the aggrega- million in revenue and US$1.63 million in OPEX). This tion and utility routine processes, help minimize cus- result partly draws on the staffing arrangement cho- tomers’ sense of distance from the utility and ­overcome sen by SISAR. Its technical staff consists of 20 employ- political resistance. The creation of SISAR Ceará, an ees, with an average wage of US$1,346 per month. It is organization based on community associations, complemented with local staff belonging to each required an intense participatory process, as every member association at an average wage of US$148.3 family in a community had to agree before the associa- per month per operator. These workers only work tion would join SISAR. The São José investment project part-time and usually carry out other productive routinely involves rural workers’ unions and producer activities. Hence, for local operations, SISAR draws on cooperatives. Along with the SISAR aggregation pro- community labor hired by the association, which cess, the social participation of communities and their reduces average labor costs, compared with hiring associations has intensified, as several m ­ eetings and full-time itinerant employees. This staffing arrange- training sessions take place yearly. ment allows SISAR to control labor costs and is key to achieving and maintaining operational sustainability. Financial Support and/or Incentives (a “Big Push”) Are Important to Help Services Get Out of the Low-Level Equilibrium Trap Aggregation Case Study at a Glance To boost the success of aggregation reforms, national Key Lessons Learned from Aggregation Case Study and external stakeholders can provide financial support Aggregation Takes Time to Show Results; Gradual to aggregating utilities to help them achieve the aggre- Improvement Strategies Are Particularly Successful gation purpose. In most cases, these subsidies are used Both the aggregation design and implementation take to fund investment programs thus acting as the Big time; in particular, implementation is a continuous pro- Push, which helps WSS services get out of the low-level cess that can spread over decades. As a result, aggrega- equilibrium trap. In Ceará, the investments imple- tion benefits also take time to materialize. A  gradual mented by the aggregated utility SISAR were funded for improvement strategy with regard to the main purpose 18 percent by federal credits and for 82 percent by the of the aggregation proved successful in many case stud- central government through loans from international ies, as it spread the efforts and changes to be made over financial institutions. SISAR investments were targeted time, thus not burdening utilities with having to do too toward initial set-up of WSS infrastructure, as most much too quickly. Sequenced strategies often use per- localities covered by the aggregated utility had no water formance-based targets that are monitored and regu- supply. larly reported on. In the Brazilian state of Ceará, some performance indicators were selected and targets were Note set to monitor the gradual achievements of the 1. According to the IPEA, Institute for Applied Economic Research. Case Study—SISAR Ceará, Brazil 5 © 2017 International Bank for Reconstruction and Development / The World Bank. Some rights reserved. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. This work is subject to a CC BY 3.0 IGO license (https://creativecommons.org/licenses​ by/3.0/igo). The World Bank does not necessarily own each component of the content. It is your responsibility to /­ determine whether permission is needed for reuse and to obtain permission from the copyright owner. If you have questions, email pubrights@worldbank.org.  SKU W17044 6