23721October 1999

UNDP- ILWorld BankWater and Water for India's PoorSanitationProgram Who Pays the Price for Broken Promises?

South Asia Region

I W -bZ jf b-, -

X__ v_- Ahz 5s1sl i

-;" hs+i'' :4& L IL

+.9t >~~~~~~~~~~~~

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

WATER AND SANITATION PROGRAM

Introduction

Ever since independence, poorIndians have been promised free, I -

safe household water.It is hard to be against such an 4?i

appealing idea. Poor people would - ,

obviously favor it, and nobody else, -

however rich or selfish, would deny A -

them access to such a basic humanneed. For both reasons, politicians rhave been quick to embrace 'freewater' as an election slogan. Andeven economists with sharp pencilswill concede that subsidizing waterfor the poor has benefits for society r9 eS!'X;J

as a whole (see Box 1).How have the promises been

translated into reality over the post 50 years? How and most are forced to live in highly insanitarydoes this appealing idea look at the turn of the conditions. They still have to pay for their water,century? sometimes 1 0 and 20 times more than their richer

In a word, the promises have been broken. The neighbors. Government subsidies now run atpoor still have inadequate access to safe water, Rs 40 billion a year', but most of the benefits go to

the better-off. Even the huge current subsidies comeTief Pnowhere near the investment levels needed.

The Promise of Free Water for the Poor ............................... And, in a cruel twist, the very policies that were

... rests on the assertion that the government's duty is to designed to help suppliers provide water cheaplyprovide safe water for the poor free of charge, since: to the poor have, in fact, resulted in such bad* no one should be denied access to safe household water financial and operational management bybecause he (or more likely she) cannot afford it; suppliers that they provide very little service to the* the poor cannot afford safe water; poor, and embarrassingly inadequate service toa the use of unsafe water by some leads to diseases that harm everyone else.the users and can spread to everyone else; and The purpose of this Report is to examine the* only the government has the organizational and financial gap between promise and reality in water serviceresources to supply safe water. 'Throughout the paper, the Indian Rupee (Rs) is used as the unit of -rrency.

At the time of writing (summer 1999) $1 exchanged for obout Rs 43.

2

WATER FOR N)D A' POOR

The Promises of Water for the Poor verssus the Reality.. 8

Promises Reality* The poor cannot afford safe water, and * The poor actually do pay for water, often farshould not have to pay for it. more than their fair share.* Public subsidies are provided to help * But the subsidies for water benefit mainlythe poor. those who are not poor.a The government can solve the problem * But public provision of water is inefficientby running water programs itself. and ineffective.* The government will raise the financial * But the investment requirements for waterresources needed for water supply. are far too great for the government to afford.

provision to the poor in India. It examines four ways. First, there are 'user charges', the paymentsenduring fallacies (summarized in Box 2) in the in cash and kind that people make willingly, inconventional wisdom of Indian water and sanitation exchange for a reliable supply of water. And thenservice administration. Each of these is discussed there are 'coping costs'- payments that are outsidein turn, drawing on real current experiences. In the system and that ought not to be required, butthis, the paper seeks to summarize current that they have to pay to gain access to water evenknowledge and better inform policy when it is supposedly free.makers in the search for a new politically User Charges: Few poor people in India haveacceptable framework where public policies access to safe, convenient and reliable suppliesachieve real results. and few are charged. Yet there is a growing body

of evidence that poor people in India will pay forThe . 1)Do . ,. reliable, convenient and locally valued water

The poor do not consume as much water as services. Firstly, it is reasonable to assume that poorthe rest of the population, but despite the promises, people - just like everyone else - can be counteddespite the bland assertions of politicians and policy on as reliable paying consumers of dependable,makers, they can and frequently do pay for what convenient and safe water services (as long as thelittle they consume. And they usually pay much billing is fair and reasonable and all other usersmore per liter consumed than those who are are also required to pay). Secondly, in practice,better-off. there are many examples in India where the

Poor people pay for water in two rather different poorly served offer to pay for a fair service (see

n f '!s . : ':: {or Better Service

Vijayawada Municipal Corporation in Andhra Pradesh has a population of about 1,00,000with water provided by about 900 private hand-bored wells, and a municipal system serving36,000 house connections and 6,500 public standposts. Most of the standposts have lost anytaps they once had and run continuously, and there has been persistent clamor for more (free)standposts, justified by 'public demand'. Collection of water consumption charges has beenminimal, frequently not even sufficient to cover the costs of collection.

The municipality's standard charge for a house connection is Rs 4,000, plus a monthlyconsumption fee of Rs 40. Recently, the state government released funds under the NationalSlum Development Project to provide a 50 percent subsidy toward these costs. When theVijayawada Municipality announced this program, there was an overwhelming response fromslum dwellers, and more than 5,000 came forward offering to pay the Rs 2,000 as deposit.Using the cash inflow, the municipality has extended its distribution pipelines in unserved areas,and expects to achieve 100 percent coverage without drawing on additional sources of finance.It also expects a gradual reduction in the demand for public standposts.

Source: Personal communication from Arvind Kumar (IAS, Municipal Commissioner, VijayawadaMunicipal Corporation), April 9, 1999.

3

WATER AND SANITATION PROGRAM

Villagers Make Cash and Labor ContributionsIn addition to regular savings efforts promoted by SEWA, water user committees in about 25

villages in rural Gujarat have started separate Village Water Funds to be used for operationsand maintenance expenses. Individuals contribute up to Rs 5 per month to the Water Fund. Thesum collected in these Funds amounts to well over Rs 1,00,000 in some villoges.

Source: Personal communication from Reema Nanavaty, Coordinator, Rural Development,SEWA, Ahmedobad, April 22, 1999.

The Swajal Project (World Bank Uttar Pradesh Rural Water Supply and EnvironmentalSanitation Project, India) has proved that there is a willingness to pay for good quality servicedelivery. This is the first time in India that project communities are contributing towards thecapital cost of water supply schemes. The community contribution in the Swajal Project isestimated at $7.64 million, which is about 10 percent of the total project cost.* For latrines andother individual assets such as compost pits, each beneficiary contributes about 40 percent ofthe capital cost. In addition to contributing towards the capital cost, the communities undertakefull responsibility for operation and maintenance of the water supply scheme, including payingall its costs. In order to do this, they levy user charges at different tariff rates from bothhousehold connection holders and public tapstand users.

*Draft Report for Mid-term Review (MTR) Report Vol 1, Uttar Pradesh Rural Water Supply andEnvironmental Sanitation Project, the Swajal Project, July 1999.

Box 3 for an example of slum dwellers willing to percent; a powerful statistic to reinforce theirpay in Andhra Pradesh). sustainability. Conservative estimates indicate that

Thirdly, there is evidence from numerous there are currently about 50 MFIs disbursingexternally-funded projects that the poor, who have infrastructure/housing loans to the economicallyoften grown weary of free-but-inadequate water weaker sections throughout India. Of these,supplies, can and will contribute to the cost of better approximately 15 are based in urban areas. Thereservices. There is a great variety of projects are also over 1 50 additional MFIs, currentlythroughout India where cost recovery supplying income generation and a smallarrangements appear to be working well. There proportion of consumption loans to the poor. Aare village schemes sponsored by large and small large number of these institutions are poised toIndian Non-Governmental Organizations (NGOs) begin on-lending infrastructure/housing loans tosuch as Self Employed Women's Association their clients.(SEWA) (see, for example, Box 4). There are larger These examples do not yet represent normalschemes involving NGOs and external donors (see, practice in India, but payment for service to thefor instance, Box 5) and there are a few examples poor is increasingly widespread. They serve towhere poor consumers contribute to state underline a very important truth about water andgovernment schemes (see, for instance, Box 6). poor people everywhere, not just in India. While

An alternative form of local financing is simply they obviously prefer to pay less rather thon more,borrowing money to repair or instal a pump or they are willing to payfair user charges for neededother equipment, and repaying the debt over a water if it is actually delivered. And in the case ofperiod of time. In recent years, there has been a new schemes, they are willing to contribute theirmarked growth in the number and outreach of time, labor and cash iftheyare confident that watermicrofinance institutions (MFIs) disbursing will be supplied, and their views are taken intoinfrastructure/housing loans to the urban poor in consideration when decisions. about the schemeIndia. This is largely due to an institutional shift to are being made.recognize, support and facilitate MFIs working at Coping Costs: While user charges are paidthe field level to enable poor communities access more or less voluntarily, in fair exchange forto improved infrastructure. Most successful MFIs, something of value, coping costs include threesuch as SEWA Bank, Ahmedabad, have average kinds of payment which properly ought not to beloan repayment rates ranging from 90-100 required at all. First, there are 'informal' payments,

4

WATER FOR IHDIAS POOR

which can vary from burdensome hospitality (for anecdotal. In Bangalore, for example, 1 2 percentinstance, gifts, food and accommodation for repair of respondents to a survey had paid 'speed money'technicians to induce them to visit a rural settlement) to employees of the water authority to get betterto outright bribes (for instance, to a water agency service, with an average payment of Rs 275. Evenofficial, so that he will ignore an illegal connedion). among lower income groups, nearly half theSecond, coping costs include payments which are respondents were willing to pay higher official feesnot contemplated in the original design of the water for services rather than under the table2 . Everyonescheme, but which pay for real services that are knowledgeable about the sector seems to have amade necessary by the scheme's inadequacies. fund of such anecdotes, and overoll there seemsExamples would be payments to water vendors, or little doubt that bribes and other informal paymentsinvestments in privately owned pumps or storage constitute a significant burden on water users,facilities. There is a third kind of coping cost, which including the poor.does not involve cash payments, but which is There are no national estimates of how muchnevertheless a very real cost and places a major people pay to intermediaries such as water vendorsburden on the poor - the time lost in collecting or tankers, or in private investments to supplementwater, and in illnesses caused by unsafe water or public supplies. But there are many reports aboutinadequate supplies. Even though these costs are what has happened in particular places, whichnot paid in cash, they have exactly the same impact seem to be fairly representative. In Baroda, forof reducing poor peoples' incomes, since time example (see Box 7), half the households werespent collecting water, or lying ill in bed, or looking reported to have made private investments in waterafter other ill family members, cannot be spent supply totaling Rs 280 million, and they spent fourearning money elsewhere. times as much on these measures as they did on

It is of course difficult to find out how much the public system. In Ahmedabad, slum dwellerspeople pay in bribes, which by their nature are were found to be paying vendors up to Rs 4 a dayconcealed, and the evidence is somewhat for water to avoid standing in line2 . In Dehra Dun,

Villagers Willingly Pay for Reliable ServiceIn the village of Dakhin Durgapur in West Bengal, the only safe sources of drinking water were three

handpumps installed 22 years ago. One was completely out of order, and the other two functioned poorly becauseof broken handles and damaged platforms. A regional NGO, Rama Krishna Mission (RKM), worked with thevillage youth club (Khudiram Smiriti Sangha) to organize meetings with women in the village, and they raised theproblem of handpumps because they had to walk long distances to find alternative water sources. The womenagreed to find out more about the problem and after training by RKM, conducted a household survey andanalyzed the results. The survey results were shared with panchoyat members and through RKM, the panchayat.was able to obtain funds to purchase material for pump repairs and for reconstructing the platforms. Thecommunity also decided that a committee should be established for each pump, and that each householdcollecting water from a pump should contribute towards the cost of operations and maintenance. This was donesuccessfully for two of the handpumps, the third needs to be raised above ground level and a committee has notyet been established. For the two rehabilitated pumps the fee was set at 25 paise per household per month, andhas been collected regularly. In the past two years, the community has spent Rs 120 on maintenance, and one ofthe committees has a bank balance of Rs 200.

The village of Asurali, like many others in rural Bengal, has a community fund called the Solo Anna Fund whichis replenished from time to time through community work at a common place. A handpump wos installed in thevillage 20 years ago, but ceased to function after two or three years, and people continued to rely on the old 70feet-deep village well. However, the well has recently become unreliable in summer, and the community decided touse the Solo Anna Fund to instal a new handpump. Because the balance in the fund was insufficient, additionalamounts were collected from each household.

Source: Rama Krishna Mission Lokshiksha Parishad, Quarterly Report on Community-based Sustainable PotableDrinking Water Supply in West Bengal, 1999.

ISee Som.el Paul, A Report card on Public Se,y,ces in Indi.n Ccties, Public Affirs Center, Bangalore, 1995.'Per-onol . o.. ninotimn from Mihir R. Bhatt, Honorary Drector of the Dki,ster Mitigation Institte, Ahn,edobad, G.jarat, Apnl 10, 1999.

5

WATER AND SANITAT ON PROGRAM

Water Harvesting Initiatives in GujaratTatana is a small and visibly poor village in Gujarat where the authorities twice dug a deep

well and installed a pump to provide drinking-water. But on both occasions the wells failedwithin four or five years. A voluntary organization in a neighboring village, Bhagirathi UttarBuniyadi Vidyalay, helped the villogers of Tatana to consider the option of roof rainwaterharvesting tanks. This is an old technology in the region but had fallen into disuse in recentyears. Under a program run by the Gujarat Water Supply and Sewerage Board (GWSB), asubsidy of 70 percent of the capital cost of such tanks is available (with larger subsidies forthose in fluoride areas or below the poverty line).

Ten families started digging pits for the tanks. They ran into rock and had to hire drillingmachines, for which the families had to pay Rs 7,000 each extra (the GWSB scheme limitsexpenditure to Rs 15,000 for a 10,000 liter tonk). The families received the first one-thirdpayment from GWSB on schedule, but release of subsequent payments was delayed. Thevillagers decided to go ahead with the construction, borrowing money on the local market atrates up to 3 percent a month. Because of the drilling and other features to improve strengthand durability, each family spent about Rs 24,000 in total for its tank, of which the subsidyprogram will reimburse - when payments are eventually made - about half. Once the tankswere completed, the villagers combined their resources to raise another Rs 60,000, with whichthey drilled a borewell and installed a plastic pipe to supply additional water to their newunderground tanks.

The idea of 'trading water' is not new in the region, and tankers have long been used tobring extra water into a village for festive occasions or for irrigotion purposes. Nowadays, it isnot uncommon for poor families to combine their funds and pay Rs 200 to 350 for a tanker tofill one of the new household rainwater harvesting tanks.

Source: Personal communication from Trupti Soni, PRAVAH- Ahmedabad, May 12, 1999.

a slum resident paid a private contractor Rs 3,000 More significantly, the poor paid far more for waterto instal a house connection, saying that was much than the city-wide average would suggest. In thefaster and cheaper than having the municipality dry season, the combination of low pressure andprovide the connection4. In rural Gujarat, water long queuing times drove the cost as high astrading and supply by tanker is a well established Rs 50 per cubic meter for those using public taps.practice, for drinking as well as irrigation water, The time costs associated with inadequate waterand people pay between Rs 1 75 and Rs 350 per systems are not just those related to collecting water.tanker.5 All of these examples suggest that peoples' The country suffers huge costs because of illnesseswillingness to pay for water is a good deal higher caused by unsafe and insufficient drinking water.than the authorities' willingness to charge for it. By one estimate, India loses 90 million days a year

The third kind of coping cost involves the time due to water-borne diseases, costing Rs 6 billionlost in collecting water. One estimate put the in production losses and medical treatment7 . Anational cost of fetching water at 150 million mon- more detailed study in 1995 concluded that eachdays (in reality, woman-days) a year, costing the year India lost 30.5 million 'disability-adjusted life-Indian economy Rs 10 billion in lost production6 . years' because of poor water quality, sanitationMore detailed estimates have been made in and hygiene. Itestimated thatimproved water andparticular cities. The Dehra Dun study, referred to sanitation services could halve this loss, with savingsearlier,calculatedthatforthewholecitythesecosts to the country of between Rs 1 10 billion andaveraged Rs 1 0 per cubic meter consumed, while Rs 290 billion, depending on the value assignedthe water authority was billing an average of only to a 'life-year"'.Rs 2 per cubic meter to those with connections. These rather technical-sounding statistics are

'See Choe, Vorley and Mjlini, Coping with Intermittent Water S,pply: Problems and Prospets. Dehro DOn, USAIO, 1996.'Person-l co- -unic-tion Irom Trupti Soni, PRAVAH, Ahrnedobod, Gujonut, Moy 12, 1999.

'See Raijt Chaudhuri, Water: What are Our Rights to It?, Safety Watch, calcutta, 1998.'Ibid

'See Corter Brondon and Kirsten Hommen, Voiuing En-vironentul Costs in India, World Bonk, 1995-

6

WATER FOR lNDIA S POOR

not just about losses to India's national productivity. systems, and Rs 25-30 billion to rural schemes (halfThey translate into very real human suffering, from the central government, and half from theespecially for India's poor. About 80 percent of states). The actual beneficiaries of this subsidy areIndian children suffer from water-borne diseases people who receive some sort of water service -every year9 . And nearly 7,00,000 children die of clearly people who do not get services do notdiarrhoeal diseases directly as a result of drinking benefit from the subsidy, regardless of how muchunsafe water"0 . Poor women, and their daughters, is spent. In rural areas, it is generally the better-offsuffer doubly. It is they who take care of whoever is who live close to good water sources, while thesick in the family. And it is they who, in sickness poor have marginal land and less reliable sources,and health, spend long weary hours collecting and suffer mostfrom falling watertables, seasonalwater for the family. The problems faced by women variations, etc. In urban areas, most of the subsidiesparticularly in dealing with government programs also benefit the non-poor, who typically pay farare illustrated further in Box 8. less than the real cost of the water they consume

(in both Baroda and Dehra Dun, the proportionS5. . I - ,, , ,< ,. , was only about 20 percent).the FA4.vtev-off To justify these huge subsidies, it is usually

Total subsidies for domesticwater supply in India claimed that people, especiilly poor people,are currently about Rs 40 billion a year. Of that cannot afford the real cost of water. But this is aamount, roughly Rs 1 2 billion goes to urban water highly dubious argument. First, it is clear from the

Baroda Municipal Corporation --- Who Gets the Subsidies, Who Pays?

The Baroda Municipal Corporation earned revenue of Rs 1.24 per cubic meter of water supplied in the early 1 990s. Atthe time, operations and maintenance expenditures alone amounted to Rs 1.49 per cubic meter, and total costs (beforedepreciation charges) were Rs 2.32. The 50 percent subsidy is paid for out of Octroi income, contributed by everyone inthe city, including those who get very little water such as those who live in slums. Industrial and commercial users payhigher charges to cross-subsidize domestic consumers, and pass on the costs through higher prices to all their customers.The main beneficiaries of all these arrangements are middle and upper income households, who pay less than 20 percentof the actual cost of the water they consume.

One-third of all households reported spending time collecting water, averaging one hour a day (fully one-fourth ofhouseholds with piped connections reported having to spend time collecting water). Because the supply is unreliable,better-off households resort to a variety of coping mechanisms, such as underground storage tanks, lift pumps to reach theupper floor, or pumps to extract larger flows from the municipal pipelines. Over 20 percent of households employed oneor more of these devices, and another 1 9 percent used water from private or shared borewells. Investments in such itemsaveraged nearly Rs 3,000 per household, with annual expenditures of around Rs 440 per household. Municipal watercharges added only another Rs 100. In sum, to obtain water, the people of Baroda spent more than four times as much onprivate measures as they did on the public system.

Poor households typically cannot afford large cash outlays and are forced to rely on public sources - 62 percent ofhouseholds in the income group below Rs 1,500 a month did so. Contrary to popular opinion, however, water from publicsources is not free. The Corporation collects water charges from local residents. For example, one group of 15 houses paidRs 480 for installation of a standpost, plus Rs 180 a year in charges. In addition, they paid Rs 2,000 for construction of awashing place and soak pit at the site. More important than these cash payments are the opportunity costs of time spentcollecting water, and therefore not being able to earn an income. These opportunity costs (conservatively estimated at onlyRe 1 per hour) amounted to about Rs 26 million annually just for collecting water in Baroda.

Poor households, in the income group below Rs 1,500 a month, incurred opportunity costs of Rs 250 per annum onaverage. For households with monthly incomes above Rs 6,000, the corresponding figure was naturally very low (onlyRs 13), since these households solve the problem by paying more in capital and recurrent expenditures. When opportunitycosts and all other expenses are included, the average city-wide cost of water amounted to 1.7 percent of householdincome. But for poor households in the income group below Rs 1,500 a month, it was 2 percent.

Source: Chetan Vaidya, Willingness to Pay for Water and Sanitation Services: A Case Study of Baroda, HSMI, New Delhi, 1998.

'See William Gunyo-, ,d,. Mk-g Go Fun-,din- g Wo,k h-/ade-, WoterAid, 1998'Se- Ashok Nigo, et o1, Fresh Wate, for I aChddren and Nature, UNICEF and WWF, 1 998.

7

WATER AND SANITATION PROGRAM

Women Especially Disadvantaged by Government Programs EThe NGO SEWA works especiolly with women, and in response to their concerns has, in

recent years, helped members undertake a number of water schemes in rural Gujarat. Aparticipatory assessment carried out in 1997 highlighted the special difficulties faced by womenwhen dealing with government programs.* Often, water was turned on only after dark, when it was unsafe for women to walk suchdistances alone. Elsewhere, water was released when the women were out in fields for work.* Many women were not aware that the government service exists to benefit them. It was easyfor landlords to divert pipelines away from villages to irrigate their fields or demand a higherpayment from poor families for access to water sources.* Women who were versed in their rights said they had been restricted by social custom or lawfrom confronting exploitative upper caste males, or reporting technical problems to higherofficials. Borewell motors that had worn down from frequent electricity blackouts or taps thathad run dry were thus left unattended.

Source: Reema Nanavoty and Rina Agarwala, Managing Water for the People by the People,SEWA, Ahmedabad, 1997.

previous section that poor people actually do pay subsidy quickly becomes an 'entitlement' for thosefor water, however 'unaffordable' it may be. And who benefit, and the group of those consideringsecond, 'affordability' is a very slippery concept, themselves entitled to the subsidy grows rapidly.which entails judgements about how other people This seems to be true everywhere, and India'sshould spend their money, as well as averages that experience is no different from the rest of the world.can be highly misleading. But for whatever they The unfortunate result is that, in India today, mostare worth, rough benchmarks of affordability (costs people continue to feel that the government shouldof domestic water in relation to household income) subsidize the cost of their water consumption, andhave been worked out in many developing manyfeelwatershouldbefree.Thisisacompletelycountries, and the upper threshold usually impractical and unaffordable delusion and onediscussed is about 5 percent. In India, an upper which lies at the heart of India's shocking servicethreshold of 3 percent seems to be considered record. Concerted political will is required tonormal"1 , and of course wealthier families pay a change this public mindset.much lower proportion. All in all, the affordabilityargument is simply not persuasive. Public Provision is Inefficient

This is not to deny that there are a few extremely and Ineffectivepoor and marginalized people in every community, What has actually been accomplished throughpeople who really cannot afford to pay for water these enormous public subsidies? In rural areas,or food or any other basic necessities of life. But more than two million handpumps have beensuch unfortunate people are much more likely to installed under successive Plans, as well asbe known and supported by their own community 1,16,000 piped schemes, including 1.5 millionthan by any government subsidy system 12 . standposts. As a result, the government can boast

It emerges from the previous paragraphs that that 52 percent of the rural population is 'fullysubsidies for domestic water in India are at present, covered' and almost all the rest is 'partially covered'extremely large, spread very widely, ineffective in by improved water supply (fully covered means ahelping the poor, and mainly an advantage to the source meeting the standard of 40 liters per capitabetter-off. This is not the result of a vast conspiracy per day within a distance of 1.6 km).but a lack of a detailed understanding of the effect These accomplishments are unfortunatelyof high subsidies on individual incentives. Any diminished by the huge gaps still remaining. It is

"See Wilhelm Kedderman, Ghogha Regional Water Supply and Sanitation Pnojed, Netherlands Economic Institute, 1999."'This should not be interpreted to mean that all forms of subsidy are always bad, Indeed, there are many situationswhere cross-subsidization is highlydesirable, Cross-subsidization occurs when revenues from one group of consumers are used to subsidize another group within the same water system. Thishappens, for instance, when everyone pays the some tariff even though some consumers are harder (and more expensive) to serve because they live inoutlying or hilly ares. It also happens when the toaff is set so arge that consumers pay more per titer than small consumers Cross-subsiditian at this type isnormal in water systems throughout the wortd, and is a practical way for water authorities to ensure service to all their clients, especially the poor.

8

WAFER FOR IRDIA'S POOR

obvious that a great many facilities have not been drinking water, such as that which led to theproperly maintained, so that by the government's outbreaks of cholera in Lucknow in 1 98713 and inown 1994 figures, in rural areas 4,60,000 Chennai in 199314.handpumps (22 percent of the total) required repair A recent joint review of water resourcesor rehabilitation, and 2,50,000 (12 percent) were management by the Government of India and thecompletely defunct. The status of rural piped World Bank noted that urban water systems:schemes is somewhat better, but in 1994 there were "deliver on average 50 to 60 percent of theirstill 45,000 schemes (26 percent of the total) and capacity to end users, compared with 80 to 852,80,000 stondposts, requiring repair or percent in other countries. Poor, and sometimesrehabilitation. It is commonly asserted that at any non-existent, management leads to waste andone time at least one-third of all public handpumps inefficiency, with the resultant large claim onare out of order. No doubt a large share of these resources that could be redeployed for servicecasesoffacilities'outof order' isduetoinadequate improvements. The lack of human resourcemaintenance and repair programs, a responsibility development and personnel policies tailored toof the authorities. But the figures also suggest that meet organizational needs has led to boththe communities served have no great stake in the widespread overstaffing and labor misallocations.handpumps or standposts provided, probably Overstaffing is endemic, the ratio of staff per 1,000because they were not much involved in the service connections, an accepted measure ofplanning and decisions leading to their installation. efficiency of water utilities, ranges from 40 to 60It appears that 'coverage', meaning number of staff per 1,000 connections in India." A footnotecommunities served by new instcllations, was a adds that the regional average is around 10more important consideration than 'sustainability', staff per 1,000 connections; international bestmeaning ensuring that installations would keep practice is around two to three staff per 1,000functioning indefinitely. connections15 .

At the same time, there is widespread scepticism As if all these problems were not enough, Indiaabout the accuracy of government figures on faces one more enormous problem with its drinkingcoverage. The definitions in use tend to be static, water -- deteriorating quality. One can argue aboutso that once a community has been 'served' it is how much responsibility the government bears forconsidered covered, regardless of later population regulatory inaction in the post, but it is clear thatgrowth or the current conditions of the facilities or the government will have to play a major role inwater table. And, of course, the use of coverage the future. According to one .. .targets inevitably leads to some overstatement of government estimate, 44 millionaccomplishments. For these reasons, even the sad people suffer from such waterpicture painted by the government survey over- quality problems as excessive C -

estimates the coverage and understates the gaps fluoride, iron, nitrate, arsenic and -

in rural areas. salinity. And in many places whereThe status of municipal systems is no better. quality is still good, irrigation

Virtually all are in a parlous financial condition. In pumping has caused water tablesonly a few cases are revenues sufficient to cover to fall below levels reachable byoperating costs, much less depreciation and low-cost technology. -

borrowing charges. The combination of inadequate The better-off contrive torevenues and poor procurement, construction and overcome all these problems in onemaintenance practices has led to low operating way or another. In rural areas, richefficiencies. In most urban areas, large areas farmers can sometimes use 1 -7,(usually slums) receive no or very little service, and agriculture programs to develop 4

even in the formal residential areas, water supply wells so as to assure their ownis intermittent. This is so common that it is taken as drinking water and sell it to others the normal state of affairs. But intermittent supply by tanker. In urban areas, they ,is not merely an inconvenience for consumers, it juggle with their plumbing and - s F,!'

also provides opportunities for contamination of instal booster pumps to

'See Bi-nie et i, U,ban En-vonenial S.ervices M,ste, Pila for Luckn-, 1996-2021, DFID, New Delhi, 1996.'See Worid Bank, Urban Water Suppiy and Sanitali.n Repert, 1995."Ibd. 1-

WATER AND SANITATION PROGRAM

compensate for low pressure, and put in storage Even if questions about how well financialtanks and boreholes to deal with intermittent supply. resources are being spent currently are set aside,To safeguard quality, they instal filters, boil water, it is clear that huge additional outlays will beand nowadays increasingly buy it in bottles (sales needed to make good the gaps and deficienciesof bottled water have increased 30 percent every listed above. The recent joint review of wateryear for the past several years). No one has resourcesmanagementbytheGovernmentofIndiaestimated the total 'supplementary' investment of and the World Bank orrived at rough estimates ofthis type, but the countrywide amounts must be recurrent expenditure and investment needs forenormous (see Box 7 for evidence from Baroda). water and sanitation separately. For water in rural

The better-off find the money, and these areas, the annual requirement just for adequateexpenses serve to justify their view that water operations and maintenance is estimated at Rs 29charges (if they are paid at all) should be kept at billion. The investment requirement is over Rs 200very low current levels. The poor cannot afford billion to rehabilitate and repair all existing schemeslarge lump sum investments. With a few exceptions and fill in gaps where necessary, and an additionalof the type noted earlier (see Boxes 3-6), the poor Rs 450 billion to bring the whole rural populationbear the main burden of these problems by waiting to the 'full coverage' standard of 40 liters per capitain line for inadequate amounts of poor quality water. per day within a distance of 1,600 meters.

For urban areas, the report assumed heroicallyPublic Provision is too Expensive that operations and maintenance costs would be

It was noted earlier that the government spends covered by appropriately increased tariffs, andhuge amounts on water, to the order of Rs 40 billion included an estimated investment requirement fora year currently - about Rs 25 billion for rural and water supply schemes of Rs 284 billion over fiveabout Rs 16 billion for urban water. During the years. Another estimate, bythe Indian Expert Groupperiod 1951 -97, total Plan outlays for rural water on Commercialization of Infrastructure, was much(including very modest amounts for sanitation) were larger - Rs 845 billion. And a third party observedRs 202 billion, and for urbon water (also including that both estimates were probably on the low side,sanitation) about Rs 136 billion. since neither took into account a characteristic

Not all this money had been well managed or feature of urban systems, namely the rising realwell spent. In rural areas, for example, calculations cost per cubic meter of accessing the 'next' remoteshow that the weighted average per capita water source, typically two to three times theinvestment requirement is Rs 630, whereas the current costs.actual investment cost per capita has been Rs 760 How much public funds are likely to be availablein recent years. The 20 percent difference is at least to cover these needs? Certainly, nothing evenpartially attributable to inefficient procurement approaching the gigantic sums just cited. The somepractices. Huge amounts are spent on staff rather report that estimated rural operations andthan facilities or equipment - for example, in Uttar maintenance needs at Rs 29 billion a year notedPradesh, Jal Nigam spends nearly 40 percent of that current and prospective funding for thisits Rs 4 billion annual budget on its 20,000 strong purpose was to the order of Rs 2.5 billion a year,staff. less than one-tenth of the requirement. And the

The situation of urban utilities is no better: urban report projected that about Rs 30 billion"Most are in financial disarray. A recent survey would be available annually for both water and

reported by MOUAE suggests that out of 17 local sanitation - following past patterns, that wouldbodies, seven were able to meet their O&M costs mean about Rs 15 billion a year for water supply.whereas only two covered also their debt servicing Even the lowest estimate of requirements is fourcosts. Analysis of other recent studies suggests that times this amount.out of the 15 for which information was available, Given the myriad other problems which Indiaonly four cities cover their O&M costs completely, faces, and the keen competition for public funds,and another four cover more than 80 percent of it is simply not realistic to assume any large increasethese costs. Information on the coverage of debt in government funding for the sector. In fact, theservicing (capital costs) was available for only initial promises about the poor and water havesix cities, of which three almost fully covered been turned on their head: it is now the governmentthese costs."'

6 that cannot afford to pay for water.16. World Bank, Urban Water Supply and Sanitaion Report, 1998.

10

WATER FOR NDQNS POOR

Conclusion

In summary, it emerges that: coverage. It requires* Very large numbers of poor people, supposedly the active involvementunable to afford safe water, still do not receive it, of water users inand have to pay if they do receive it. planning facilities in* The huge subsidies allocated to the sector making informedprimarily benefit the better-off. choices about service* The public authorities responsible for services levels, which in turnare generally ineffective. facilitates their* The recurrent financial and investment willingness to pay.requirements of the sector vastly exceed the public 3. Subsidies arefunds likely to be available. always troublesome,

This harsh set of realities stands in sharp contrast and are best avoided,to the poverty of action to embrace significant since they invariablychanges to conventional thinking or to alter the end up benefiting -continuing false promises by politicians and the mainly those outsidegovernment, at various levels, to continue to the original target

provide free w ater services. Anrlysis of the group. Any subsidiesappro a ches by all the leading partis going into should, therefore, besthe September 1999 elections indicates little limited and tightlypolitical mandate by any party other than more focused. On the otherfree services. Political will to change the frame of hand, cross-subsidies . , fpresent approaches is vital. can play a useful role.

Successful experiences by governments, NGOs, 4. Private sector andcivil society and private sector organizations civil society orga-suggest several common themes that would need nizations (including NGOs) are much moreto be embraced if India is to realize its hope of effective at delivering services than governments.serving its poor citizens by significantly increasing Government's resources are best focused insustainable access to safe and reliable services: facilitating and regulating other parties'

1. People, including poor people, can afford contributions than continuing to attempt to provideand are willing to pay for safe water: if they get a services for all.reliable supply; if they get the service levels and 5. There are sources other than governmentother features theywant; and if they are sure other for the funds required for operations andpeople are paying their fair share too. investment, including increased revenue from water

2. Sustainability is more important than charges and the mobilization of private savings.

1 1

UNDP-World Bank Water and

Sanitation Program - South Asia

55 Lodi Estate

New Delhi 110 003, India

Tel: (91)-(0)1 1-4690488/9

Fax: 4628250

E-mail: wspsa@worldbank.org

October 1999

The World Bank does not accept responsibility for the views expressed herein, which are those of the authorsand should not be attributed to the World Bank or its affiliated organizations. The findings, interpretationsand conclusions are the result of research supported by the Bank. The designations employed and thepresentation of the material are solely for the convenience of the reader and do not imply the expression ofany legal opinion whatsoever on the part of the World Bank or its affiliates concerning the legal status ofany country, territory, city, area or of its authorities, or concerning the delimitations of its boundaries ornational affiliation.Designed end produced by Media Workshop Indi. Pvt Ltd. Tel: 91 -585691/92. E-moil: mwipl@vsnl.com