The Water and Sanilation Program (www.wsp.org) Oecember 2010 is a mulli-dOflOr par!nership adminislered by !he World Bank lo supporl poClf people in oblaining affordable. safe. and sustainable access lO water and sanilalion services. Fleld Note 73028 Learnings from Bangladesh Experience in Private Provision of Rural Piped Water Bangladesh's experíence ín engagíng the prívate sector to construct and manage domestíc water supplíes, partícularly píped schemes ín rural areas, over the pas! few years, have global relevance. , • Executive Summary tubewells. Thus, lhe unusual problem 01 widespread arsenlc conlaminalion in Bangladesh depended on shallow tubewell technology tar provlslon 01 shallow aquilers pushed Bangladesh lo test whether private operators could ¡rrigalion and rural domestlc water supply for years. Following the detection quickly increase Ihe number 01 piped 01 widespread arsenie contamlnatlon in shallow aquifers, the country was schemes supplying sale drinking water in compelled lo test whether private operators could quickly ¡ncrease the rural areas. While the impetus lo number 01 piped schemes supplying safe drinking water in tu.ral areas. 1I experimenl wilh expanding rural piped water supply Ihrough the private sector has experimented with several approaches : ene such approach has been was uniquely Bangladeshi, the ensuing trled In the World Bank-assisted proJects whlle the Bangladesh government, experiences are globally relevan!. on lis own, has funded olher projects uslng dlfferent approaches. Global Context Thls Field Note analyzes and compares the two models 01 publlc-private Governmenls have Irequently used partnership: the build-operate-transfer model and the turnkey model , their public-private partnerships in urban areas lo manage, or conslruct and financing as well as Implementatlon strategies. manage, water infrastructure. Similar models have been adapted lor small The Field Note then arrives at the conclusion that these firsl efforts at towns. However, such partnerships have been lar less common in rural areas, due puulng in place private sector models to develop rural piped schemes have 10 a general rural prelerence lor stumbled , but not entltely failed ; the two promising models have been communily management. tested, revealing strengths and flaws in the respective deslgns. The lessons learned and the questlons ralsed trom this experience contribute to the Community management 01 rural water supplies has enjoyed widespread lessons that have emerged globally so tar in regard to the prlvate provision success. Nonelheless, il has certain of rural water supplles. limits, as pointed out (Box 1). These limils have led water sector professionals to search lor ways either lo palch Ihe cracks in community management, or lo Introduction Prior to these initialives, Bangladesh had successfully used the private sector to find a private sector-based alternative lar A number 01 countries, including situations in which communily expand irrigalion and rural domestic water management does nol work as intended. Bangladesh, have experimented in supply, in bolh cases based on shallow engaging the private seclor lo conslruct tubewell lechnology. Bangladesh had Currenlly, there are numerous examples and manage domeslic water supplies, almoS! universal water supply coverage 01 using the private seClor lo strengthen particularly piped schemes, in rural wi th this technology. However the deleclion Ihe managemenl 01 rural water supplies, areas over Ihe pasl few years. One 01 arsenic in the shallow aquilers reduced and fewer examples 01 using ¡he seclor approach has been Iried in the World the coverage lo nearly 75 percent which to construcl as well as manage rural Bank-assisled projects, whi le the has become a serious challenge lar sale schemes (Kleemeier 2007). Many 01 Bangladesh government, on its own , drinking water supply in the affecled areas. these examples involve piped schemes, has funded other projects using a One available oplion was lo draw water lor example, in Rwanda, Burkina Faso, differenl approach. from deeper aquilers using deep Senegal, and Maurilania. 2 learnings Irom Bangladesh Experience in Private Provision 01 Rural Piped Water Box 1 l lmlts to Communl ty Management The management 01 rural water supplies by cOOlmunity committees or associations has many advatages, but certain limitations as welL The community management model is embedded in a larger and longer process tha! relies on insufficient publiC linancing. The process starts with a publicly-funded rural water supply projecl. However, government project budgets, even supplemented by donar assistance, are i1evitably too limited lo mee! demando Many communilies must be excluded as a resull. Bangladesh itself illustrales this situation in a way. The government wilh donor support, parlicularly Irom UNICEF, mounled a huge program lo promote shallow wells. Despite that enormous investment, only around 20 percent 01 the shallow tubewells have been constructed with public lunds . The real impact 01 that government project came Irom ils eflectiveness in mobil izing the private sector to construct 5.3 million shallow lubewells Ihrough consumer funding. However, very few rural water supply projects have this effect 01 leveraging alternative investment lunds and linancing. Over lime, community management Irequently declines in quality. Brocklehursl enumerates lamiliar problems: Irained and experienced committee members lose inleresl or move out 01 Ihe communily; committee members lorge! skills or comm ittees become inactive during long periods when repai rs and other interventions are nol necessary; far Ihe same reason, commiltees have Irouble gelling consumers lo conlinue making conlribulions lo the maintenance fund: and internal squabbles and corruplion tear commillees apar\. Yacoob and Walker (1991) add Ihal rural piped schemes, in particular, are loo expensive lor communities lo maintain on a nonprolit basis using volunteers, il one considers Ihe true economic cost 01 Ihe time and labor that communily members are expected to donate. Public funding comes in "feasl or lamine cycles." Ihat typically leaves loo lew lunds lor the O&M phase. Rural waler supply projects las! lor a specilic period 01 time, al which poinl the lunding disappears. However, communily managers need continuous support lo deal wilh Ihe kinds 01 problems listed aboye. Furthermore, as the water supplies age, they require repairs beyond ¡he financial means, technical capabililies, or equ ipment 01 communilY waler committees. The Malawi rural piped scheme program illustrales this on a grand scale: the communily-managed schemes lelt into disrepair as Ihe government withdrew lunding Irom the water department Ihal provided backup lo Ihe village committees (Kleemeier 2001). Bangladesh's efforts al using the private the Bangladesh case is Ihe World Bank- private sector lor bolh the conslruction sector to expand sale water supply assisled projecl in Paraguay. The and subsequent management 01 rural through rural piped schemes lits into Ihis Paraguay project is somewhat dilferenl piped schemes. and the private inlernalional wave 01 experimentation. Irom Ihe Alrican projecls ciled here in contractor-operator had lo provide a Probably Ihe inilialive most relevant lo thal the government contracted with the significanl portion 01 the financing 3 - Prior to the discovery 01 naturally occu rring arsenic, rural consumers in Bangladesh took on the responsibility to install and maintain the vast majority 01 shallow tubewells by buying services and supplies lrom the private market Bangladesh Context Al firsl glance, Bangladesh seems an unlikely place to test private sector models lor rural piped schemes , simply because most 01 the rural population already had sale, inexpensive, and convenient water supplies by the early 1990s. This achievement largely resulted Irom widespread coverage by shallow tubewells, which tapped the shallow aquifers extend ing through most of the country. Had the public seclor played Ihe main role in fínancing, constructing, and managing shallow lubewells, coverage would have been limiled. Instead, rural consumers Ihemselves took on the responsibility lo install and maintain Ihe vasl majority 01 shallow lubewells by buying services and supplies Irom the prívate market. 1 As 01 2004, 5.3 million shallow tubewells had been privalely constructed, accounling lor over 75 percent 01 all tubewells and 83 percent 01 shallow tubewells in Ihe country (UPI 2005a, p. 66: UPI 2005b, p.22, 35). This success was deall a blow with ¡he discovery that naturally occurring arsenic contaminates groundwater down 10 depths 01 200 lo 250 lee! ac ross large parts 01 the country. By ¡he late 1990s, water from aboul 20 percenl 01 Ihe lor the construction . The World Bank- prívate provision 01 rural water supplies shallow lubewells could no longer be assisted projects in Bangladesh were (Box 2), bul none 01 these assessments considered sale lor domestic use . Falling designed along similar lines, in an eHorl has considered Ihe Bangladesh case. water lables had also begun lo affecl lO mobilize sufficient financing lo While much is unique about ¡he rural shallow wells adversely, parlicularly in construct rural piped schemes an a water supply situation in Bangladesh. large seale. efforts lo use the prívate sector lo , Tho privale sector carne lo play 1M pivoIaI rolo in s/l811ow ~I ~ I . parUy <1"" to 1M _y Iow 0051 at>d ¡inance, construct, and manage ru ral smpticoly oIlho l&CmoIogy. any Sorne observers have begun lo draw Ih& QOY$rvnent. """"'s. and NGOs 10 "'&ale l ho cond~ ..... tor piped schemes have confronted the private mor"OJlaclUr... s. supph..... BntnsM ano:! Paul 2000) , Learnings lrom Bangladesh Expefience in Private Provlsion 01 Rural Piped Water !he dry season. Bangladesh, therelore, Any alternative proposed lar Bangladesh many households had !heir own private needed an alternative water supply had lo compele with the convenience 01 handpumps located wilhin their lechnology to serve a growing segmenl shallow lubewells. People had grown compounds, even in their kitchens or 01 Ihe rural populalion Ihal had once accustomed lO sharing a drinking water bathrooms. People were simply not been covered by ostensibly safe and source among an average 01 two lo Ihree prepared to walk any dístance lo relíable shallow lubewells. households. In practice this meanl Ihat letch water, Box 2 Global Lessons from the Pnvate ProvlSlon of Rural Water Supplles Some observers have already begun lo evaluale the lessons learned lo date !rom Ihe prívate provision 01 rural water supplies: • Demand : Demand exists lar good qualily services, allhough il tends lo be undereslimaled. However, prívate providers musl adapl lo the characterislics 01 the local customers, lar example, clearly wrilten bilis, Ilexible payment schedules, and inlormation campaigns. Oisconnection has lo be carried out fair ly and as a lasl resort. bul is probably a necessary policy, • Subsidies and financing: Soft linancing and subsidies will almos! always be required. It is more effective to provide subsidies for access (Ior example, connections) than lor consumption (Ior example, tariffs). • Operators: Attracting good operators is not easy, and implementing agencies will have lo look in unexpected places lar them, The operator is nol usually a lormal small business wilh lechnical skills. The profile 01 a successful independenl operator is a person with a sharp commercial sense, investmen! ability, and skill al managing relations with local authorities. II the government opens Ihe door to such individuals, or lo lirms outside the water sector, linding operators is easier, Over time. ¡he operators will adapl and scale up. However, given Iheir inexperience, the government will have lo provide operators with training in bid preparation, allow ample lime lor bid preparation, and provide other types 01 support such as encouraging joint ventures. • Regulalion: The agency regulation 01 small companies in rural areas ís less necessary, more difficult, and more burdensome on ¡he companies involved than the regulation 01 large companies in urban areas. Furthermore, national regulalors lind il dillicult to devote suflicient attention lo small private operators, Local authorities and community organizations can play this role much belter, but they have to be trained and supported to provide effective oversight. The communities also have to be introduced to the idea 01 local private operators through consultative. educational, and promotional activities, Perlormance standards in contracts, enforced through close monitoring by local entilies, are a promising means 01 regulation, In this context, periodic audits, consullations, and incentives are more elfective in producing results than detailed provisions lor linancial penaliies, In addition lO conlracting, Iranchising, licensing, and sell-regulalion Ihrough member associations have been successlul approaches. Seuce. Ver"'. 2002, Econ Ooe Re ...a'ch 2003, A<>QI.OIIlII and T'che 2000. T,;che AeQ ........ ar>d II,.,...¡ 2000, and "".I'ey·visser el" 2000 I5 One impetus lor developing rural piped schemes to supply potable water came Irom ROA, which has been working lor decades on how to make deep tubewell irrigation schemes a more prolitable enterprise lor private individual s This was ¡he maio lesson learned Irom 2003). The study found ¡hal nol only did government research and training the World Bank-assisted project tha! rural Bangladeshis prefer pipad institute, had been working lor decades mounted an emergency response lo the schemes, people would be willing lo pay on how to make deep tubewell irrigalion arsenic crisis. The Bangladesh Arsenic lor this service. Specifically, the lindings schemes a more prolitable enterprise lor Mitigation Water Supply Project suggesled Ihallhe government wou ld private individuals (Box 3). A recenl idea (BAMWSP) had originaJly tested several only have 10 conlribule a porlion 01 was to construc! schemes !ha! would technologies, bu! lound ¡ha! people were capi tél cosls, since consumers would be supply water lar both iHigation and nOI ¡nterested in supply tha! was less witling and able lo pay ~U operation and domestic use, so tha! operators would conveníenl than shallow lubewells. Pipad mainlenance (O&M) COSIs, and parl 01 have a second and year-round revenue water schemes were Ihe obvious winner the construClion COSIS. stream. Furthermore, the thinking went, by this criteríon. consumers could use even Ihe domeslic Meanwhile, Ihere was a second ímpetus Research by the Water aod Sanitalion lar developing rural piped schemes 10 water supplies lor income-generating Program (WSP) aod BRAC. a oatianal supply potable water, one tha! came out activilies such as kitchen gardens, load nongovernmental organization (NGO), 01 the ¡rrigation sector. The Rural processing, lish ponds, nurseries, and so reinforced th is conclusion (Ahmed el al Development Academy (ROA), a on . These activities would help Box 3 Rural Development Academy ROA is a training and research institute operat ng under Ihe Rural Oevelopment and Cooperatives Division, Ministry 01 Local Governmenl, Rural Developmenl and Cooperatives. ROA is modeled on Ihe Bangladesh Academy lor Rural Development (BARD), under ¡he same minislry and departmen\. BARD lounder, Dr. Akhtar Hameed Khan, and his stall succeeded in doubling and tripling rice yields 01 small farmers in Comilla districl Ihrough a strategy based on in:egrating services such as credit, inpul supply, and exlension with in!rastructure, particularly mechanized irrigation. ADA was established in 1974 lo provide similar services in norlhern Bangladesh , bul wilh a special facus on water resources. Starting in the early 1980s, ROA began research on making deep tubewell irrigalion schemes more profitable lor their owners. To reduce the costs 01 installing the lubewel1s themselves, ADA developed less expensive drilling and conslruclion methods. To make irrigation more prolitable, the academy designed buried pipelines, which made more officionl use 01 water than Ihe traditionat open :::hannels. To increase revenue, domestic water supply was inCurlJUréit~tI wilh ¡rrigalion. Finally, 10 increase Ihe ability 01 households 10 pay lar water, RDA eslablished microcredit programs and oflered Iraining in income-generating activities in villages where schemes were bu il!. This was integraled rural development in the tradition 01 Dr. Khan and BARD. Since 1999, the Governmenl 01 Bangladesh has lunded live projects, a total 01 BDT 677 millioo, to test the ROA approach lo "illage water supply and irrigalion. As 01 Jaruary 2008, RDA had completed 73 schemes, wilh another 56 planned Q( under construclion. The schemes are lacated Ihroughoul the country, although many are in the dislricts surrounding Ihe ADA campus. ADA has a Centre for Irrigalion and Water Mana.;:¡ement (CIWM) which implemenls the water projecls. CIWM slaff number over 100, 01 which under 20 work on the technical aspects 01 (he water schemes, and around 70 work on Ihe microcredit program. , • Learnings Irom Bangladesh Experience in Private Provisk>n 01 Rural Pipe8 materiat in I~is put-,¡icatoon is copY'"ighled. Reqvesl s tOl permission 10 re¡)o!o C,ed,K EO,a!Jelh KI""""',e, ar.::l WSP wspOworlstar' Phone 1880-21 8159001 '90 1..j·'901 5/9028 o Phone (91·111 24690488-89 Phone 192-51) 2279641-..j6 i F;¡x 1880-218159029 FJ~ 191-11) 2..j628250 FilX (92-5112826162,22796.18,2822196 E-mail wspsa~'worldbilnk_org Web slte www_wsp .org