Nearly 1 billion people in the developing world are still without access to clean water for drinking and bathing, and 1.7 billion must contend with inadequate sanitation facilities. World Bank experience shows that new approaches are urgently needed—including changed roles for government, the private sector, and users’ groups; and, paradoxically, levying higher charges.
A daily occurrence in the pueblos jovenes of Lima: A poor family, living in an illegal squatter settlement, waits for the water truck to arrive. When it does, the family members fill a few buckets with water of dubious quality, paying $3 per cubic meter, which often amounts to 10 percent of household income. Because the price is so high, they use little water, cutting down, particularly, on “discretionary uses,” such as washing and bathing. They defecate in a fly- and insect-infested open toilet. The economic, health, and human consequences of these miserable sanitary conditions are tremendous. As a result of diarrhea and other hygiene-related diseases, for example, almost one in ten children dies before its first birthday.
Five miles away in the upper-class district of Miraflores, the situation could not be more different. Elegant houses have dishwashers, washing machines, and, in a city where it never rains, luscious gardens that are watered several times a day. The publicly run water supply company is supposed to charge its customers about 30 cents for a cubic meter of treated water available on tap in the house, but the actual amounts paid are much lower. Many households pay a low flat rate because the meters do not work or are not read, and many simply ignore their bills.
With variations of degree and detail, the Lima story—and the challenge implicit in it—is one that faces most developing countries:
• In many cities of the developing world, large numbers of poor people depend on water vendors, paying at least ten times what a middle-class person pays for a liter of water;
• At the end of the 1980s—the International Water Supply and Sanitation Decade—the number of urban residents without an adequate water supply was about the same as at the beginning of the decade; and
• Most poor people, in both towns and villages, do not have an adequate excreta disposal system.
This article takes a look at why past policies have failed and, drawing on World Bank studies, which new policies appear promising.
Shortcomings of past approaches
For decades, governments have tried to upgrade water quality and sanitation services by using public funds to invest in water and sanitation services. On the surface, this approach seems both sensible and humane. To the poor, it holds the promise of better services at lower cost, and the middle class are satisfied, since their water bills are small. To water company officials, it means funds for expansion and investment. To water company workers, it means secure jobs. To contractors and equipment suppliers, it suggests contracts will be awarded. And for politicians, it means largesse to distribute to constituents.
For a more detailed discussion, see “The demand for water in rural areas: determinants and policy implications,” by the Bank’s Water Demand Research Team, World Bank Research Observer, Vol. 8, No. 1, January 1993, and the Bank’s World Development Report 1992, Chapter 5, by the author.
In a few developing countries (including Botswana, the Republic of Korea, and Singapore), the level of public administration is sufficiently high that the government can ensure that these competing interests are harmonized and efficient services provided to most people. For the most part, however, such large public investment schemes become a vehicle for the interests of powerful groups—the upper- and middle-class-consumers, the contractors, the workers in the water utility, and the politicians. The results are predictable:
• Public resources are used on unnecessarily expensive construction projects (the combination of corruption and overdesign has been shown to quadruple the costs of sewerage infrastructure in some cities);
• Operating efficiency is very low (in Manila, 58 percent of water produced by the utility is not accounted for);
• Employment in the public companies is maintained at very high levels (over 30 workers per thousand connections in most developing country utilities versus about five in an efficient developing country operation); and
• Subsidies for water and sanitation are highly inequitable, especially for sewerage and especially in poorer countries. This is underscored by a recent study in Latin America that compares the subsidies accruing to the poorest 20 percent of the population with those accruing to the richest 20 percent. For water, in a poor country (the Dominican Republic), the poorest segment receives a subsidy that is only 29 percent of that received by the richest segment. In a middle-income country (Costa Rica), however, the figure is 75 percent. For sewerage, the figures are 13 percent for the Dominican Republic and 59 percent for Costa Rica.
In most cases, these inefficiencies are crippling, leaving no money to extend services to the unserved—almost always the poor. Thus investments designed in the name of the poor often end up providing them with nothing, but since they cannot do without water, they pay an exorbitant price, both in money and health.
The poor are willing to pay
What alternative approaches are there? Many governments and donors have responded to this predicament over the years by deciding that the poor should be provided with a lower level of services, as the costs of conventional services in industrialized countries (piped water in the house, and water-borne sewerage) are too high. The standard “appropriate technology” package has included water from a shared water point (either a handpump or a public standpipe) and an improved pit latrine.
How reliability of supply affects willingness to pay for piped water: Punjab, Pakistan
Source: World Bank Water Demand Research Team (forthcoming.).
But while reducing costs is, indeed, an indispensable element if the poor are to be served, there are two serious flaws in this response: It ignores the inefficiencies and inequities in the existing system, and it is paternalistic. Little, if any, account is taken of what services (poor) people want, and how much they are willing to pay.
As it turns out, both research and practical experience increasingly show that in many circumstances—far more than had previously been thought—poor people want and are willing to pay for services of a relatively high quality, and they will pay substantially more if the service is reliable (see chart). This is underscored by a recent Bank study of Brazil, Haiti, India, Pakistan, Tanzania, and Zimbabwe, which revealed four types of rural communities:
• Type I Communities: willingness to pay for private connections is high and willingness to pay for public water points is low. This surprisingly large group probably includes many communities in Southeast Asia, South Asia, Latin America, the Middle East, and North Africa. The appropriate strategy is to offer and even encourage (specifically by amortizing connection costs into monthly water bills) private connections, recover all costs via the tariff, and deliver a reliable service. In these communities, public taps can provide free water without affecting the demand for private taps.
• Type II Communities: only a minority of households are willing to pay the full costs of private connections, but the majority are willing to pay the full costs of public water points. With a heavy reliance on public water points, there must be some charge for water from these to finance the system. Kiosks appear to be an attractive and flexible option in many settings. Those who wish to have house connections should be able to do so, but must have metered connections and must pay the full cost. Many of the better-off communities in Africa and poorer communities in Asia and Latin America probably fall into this category.
• Type III Communities: households’ willingness to pay for improved service is high, but not enough to pay the full costs of an improved service. These are typically arid areas in South Asia and Africa. Willingness to pay for improved water service is high (as a proportion of income), but the costs of supply are so steep that improved systems cannot be built and operated without subsidies. If transfers are available from the government or from foreign donors, the community chooses to spend some of the funds on an improved water supply. The primary services offered are public taps, wells, or boreholes, although metered yard taps should be allowed where possible, with tariffs set to recover full costs.
• Type IV Communities: willingness to pay for any kind of improved service is low. These are typically poor communities in which (a) traditional water supplies are considered more or less satisfactory by the population, or (b) water supply is seen as the financial responsibility of the government. For the time being, the appropriate rural water supply policy is simply to do nothing, although once government paternalism ceases, those who fall in (b) might express a willingness to pay, thereby becoming a Type II community.
Indeed, this type of movement from a “low-level equilibrium trap” (in which a low level of services is provided, willingness to pay and, thus, revenues are low, and the operation consequently deteriorates) to a “high-level equilibrium” (in which users get a high level of service, pay for it, and maintain the desired system) is exactly what happened in rural northeast Thailand over a 20-year period. Early project failures with handpumps, and then motorpumps, were initially attributed to technologies too complex to maintain and the inability of the villagers to pay for improved supplies. But gradually it became clear that the main problem was a failure to offer the villagers the service they wanted. Subsequently, yard-taps were allowed, with the users paying the full costs of connection—treating the poor as “Type I” instead of “Type IV.” The project was a success and the low-level cycle broken.
The story is much the same in most urban areas, both for water and, to the surprise of many, sanitation. Experience shows that people living in dense settlements are willing to pay a substantial amount for the removal of excreta and sanitation from the vicinity of the household. (They are not, however, typically prepared to pay for treatment and safe disposal. Public funding should, accordingly, be applied primarily to treatment and disposal, services that have wider communal benefits.)
In Orangi, a squatter settlement of Karachi with severe sanitation problems, over half a million people have self-constructed, self-financed, and self-maintained water-borne sewers (see box). In low-income urban areas in northeast Brazil, households were given the option of low-quality services at low cost, and modified sewerage systems at moderate cost. Virtually all families chose the latter option and have continued to pay the agreed-upon charges and meet their obligations for maintaining the network of feeder sewers.
One caveat on this general theme, however, is that it is not always simple to judge what poor households might consider relatively high levels of service. Detailed assessments of the demand for sanitation in Kumasi, Ghana, and Ouagadougou, Burkina Faso, for example, have shown that poor families consider a ventilated improved pit latrine to be a comparable level of service to a water-borne sewerage system, probably because they perceive the simpler, more autonomous technology to be more reliable.
Innovative sewerage in a Karachi squatter settlement
In the early 1980s, when Akhter Hameed Khan, a world-renowned community organizer, began working in the Karachi slums, the streets were filled with excreta and waste water, making movement difficult and creating enormous health hazards. The local residents asked him to persuade the Karachi Development Authority (KDA) to provide a traditional sewerage system for free, as it did (or so they perceived) to the richer areas. But after months of petitioning the KDA, he realized that this would never happen.
Dr. Khan then turned to the community to help him find alternatives, a step he would later describe as the most important thing he did in Orangi—liberating the people from the demobilizing myths of government promises. With a small amount of core external funding, the Orangi Pilot Project (OPP) was started. The goal: to make the desired sewerage system affordable and to develop organizations that could provide and operate the systems.
On the technical side, the achievements of the OPP architects and engineers were remarkable and innovative. Thanks partly to the elimination of corruption and the provision of labor by community members, the costs (for an in-house sanitary latrine and house sewer on the plot, and underground sewers in the lanes and streets) were cut to less than $100/household.
The organizational achievements were equally impressive. The OPP staff played a catalytic role—explaining the benefits and technical possibilities to residents, conducting research, and providing technical assistance. The households financed their share of the costs, participated in construction, and elected lane managers (representing about 15 households), who, in turn, elected neighborhood committees (typically around 600 houses), which manage the secondary sewers.
As the power of the OPP-related organizations increased, they were able to convince the municipality to provide funds for the construction of primary and secondary sewers. In the end, the OPP has led to the provision of sewerage to over 600,000 poor people in Karachi, besides inspiring at least one other municipal agency in Pakistan to try the same approach.
An innovative approach
Clearly, the way forward is to give people choices. This means switching from a supply-driven mentality to one in which households—rich and poor alike—are offered a menu of levels of service, with associated price tags. To do so, suppliers will have to take on the challenges of minimizing costs and expanding the range of services available. And the government can help by creating the right enabling environment.
Minimizing costs. Poor people want and are willing to pay for good services. What they cannot and should not pay is a premium for overdesigned and inefficiently delivered services. The key to minimizing costs is incentives and accountability. Most public utilities in developing countries have few incentives to reduce costs and are high-cost providers of services. In Bogota, Colombia, for example, the unit costs of those parts of the billing system contracted out to private companies are one fifth of those handled by the public water supply company.
Offering a greater range of services. This is a vital element of a demand-driven strategy. In some instances, it will mean offering a quite different type of service—say, an improved pit latrine as an alternative to a sewer. In others, it will mean developing options in which small reductions in the level of service result in large cost reductions. These effects can be very dramatic. For example, the modified sewers in Orangi cost about one tenth of the conventional sewers provided by the Karachi Development Authority.
Rethinking institutional arrangements. But it will be difficult to reduce costs and expand the menu if the various actors lack the incentives for providing these services. A starting point for the assessment of any arrangement, then, is to assess the interests of particular stakeholders, and the opportunities they have for pursuing these interests.
• Users. Rich and poor users alike have an interest in getting the services they want at the minimum possible cost. Accordingly, where the institutional environment is such that subsidies can be obtained through lobbying, powerful interest groups will inevitably lobby for such subsidies.
• Providers of finance. Where this is public finance, the requirements for returns on capital are often weak, with, consequently, relatively little attention to using such capital efficiently. Where this is private finance—whether it be for tanker trucks in the pueblos jovenes of Lima, or for the water supply in Abidjan, Côte d’Ivoire—the investors correctly expect a reasonable return on their investment.
• Utility staff. In the private sector, the performance of managers is largely in terms of the returns they secure on investments. They thus have a strong interest in efficiency (and innovation that will result in efficiency). Public utility managers, however, are often political appointees, who inevitably give a high weight to serving the interests of the politicians that appointed them. Professional utility workers are often strongly influenced by conservative “professional standards.” And nonprofessional utility workers, who often receive their jobs as a result of political patronage, tend to become an effective political bloc. The effects are almost universally detrimental to performance. A widespread problem is that large investments generate opportunities for utility staff to supplement their formal incomes through kickbacks.
• Politicians. In an effective political system, politicians have a central role in articulating priorities and directing resources to programs addressing these priorities. In São Paulo, for instance, progressive politicians played a major role in ensuring that a technocratic water supply agency provided services to the poor (see box). But there are enormous dangers in giving politicians the ability to directly influence the operations of service providers, for instance, by setting personnel policies and appointing staff.
Thus, there are no pre-defined, ready-made, universal solutions. But, clearly, suppliers need incentives to reduce costs and be accountable to users. The necessary restructuring will often reduce the direct role of public agencies as a supplier of services, and increase the roles of the private sector, NGOs, community groups, and users themselves.
Private sector involvement in the water sector is increasing rapidly, in both industrial and developing countries. SODECI, the privately run utility in Abidjan, is considered one of the best-run water companies in Africa. Guinea, which recently let a lease contract for supplying water to its principal cities, experienced dramatic improvements in the financial condition of the utility in just the first 18 months as a result of raising the efficiency of bill collection from 15 percent to 70 percent. EMOS, the utility that serves Santiago, has used private sector contracts for functions such as meter reading, pipe maintenance, billing, and vehicle leasing, and is one of the most efficient utilities in Latin America.
Participatory politics and water in São Paulo
For the residents of the Brazilian city of São Paulo, the 1980s meant extraordinary progress in the provision of water supply and sanitation services. At the start of the decade, just 32 percent of favelas (low-income, informal settlements) had a piped water supply, and less than 1 percent had a sewerage system. But by 1990, these figures had jumped to 99 percent and 15 percent, respectively! How was this possible? Interestingly enough, the story is one of how the rules of the game change as political participation by the poor increased.
Up until the early 1980s, the state water utility serving São Paulo—known as SABESP—carried out its duties in a narrow, technocratic manner. That meant not providing services to the favelas—despite community pressures to do so—as the settlements were not “legal” and did not permit the provision of services according to the SABESP’s conventional technical standards.
This left a void for a small municipal agency, called COBES, to fill. And it did, by experimenting with new technical and institutional ways of providing water and sanitation services to the poor. On the technical side, this involved reducing the cost of providing in-house services by using plastic pipe and servicing narrow roads where access was limited—rather than just providing “second-class” service. On the institutional side, this meant the community assuming significant responsibility for facilitating community relations and supervising the contractors’ work.
But with the emergence of democratic politics in the early 1980s, the picture changed, reflecting the growing pressures on SABESP to serve the favelas. Part of the reason was that the community organizations strengthened their impact by channeling their efforts through the municipal agencies and through responsive officials and politicians (including the mayor and governor). In addition, since COBES had shown how it was, in fact, possible to serve the favelas, SABESP had no option but to respond.
The lesson from São Paulo is that democratic politics can play a fundamental role in both allowing the demands of the poor to be expressed and transforming a sophisticated technical agency from being part of the problem to part of the solution.
The government, however, has a vital role to play by devising the rules of the game, with special attention to protecting the public from potential abuse of monopoly power. It must, therefore, concentrate on the things that it, and only it, can do. Its job is to define and enforce an appropriate legal, regulatory, and administrative framework. This includes tasks as fundamental and diverse as: rewriting legislation so that water markets can come into existence; rewriting contract laws so that the private sector can participate with confidence; building a capacity for environmental and, where appropriate, economic regulation; and developing financial mandates for utilities that encourage conservation.
The way forward
Providing water and sanitation services to the poor in developing countries is a daunting challenge. Costs are rising as plentiful sources of water of adequate quality become more difficult to find. Exacerbating matters, over the next 40 years, the urban population in these countries will rise threefold, and the domestic demand for water fivefold.
For the poor, the paternalism of the past has meant a large number of unfulfilled promises and continued misery. The worst of all possible worlds is that many poor people remain unserved. Better, but not yet good, is a situation in which the poor are provided services but at unreasonably high costs, because of the inefficiency and inertia of the supply institutions. The ideal is that they are provided services that they want and are willing to pay for, by institutions that have incentives to be efficient and responsive. In a word, what poor people need is not charity but a choice of services at a fair price.
In summary, as argued in the Bank’s World Development Report 1992, large gains—in environmental quality, equity, and direct economic returns—can be realized by adopting an approach that comprises four key elements: (1) managing water resources better; (2) providing, at full cost, those “private” services that people want and are willing to pay for (including water supply and the collection of excreta and wastewater); (3) using scarce public funds primarily for those services (treatment and disposal of wastes) that provide wide communal benefits; and (4) developing flexible and responsive institutional mechanisms for providing these services, with a larger role for community organizations and the private sector.
The good news is that many developing countries are undertaking reforms along these lines.