AID"s rural water program in Latin America : what to do about high demand

These investments are needed to meet the needs of the more than 1 billion rural dwellers who were left unserved at the end of the International Drinking Water Supply and Sanitation Decade in 1990. In Ecuador, Guatemala, and Honduras, where this study was conducted, A.I.D. provided grants and loans from 1980 to 1988 for rural WS&S systems totaling roughly $50 million. These were matched more than one-to-one by host government and local communities. In Central America alone, approximately $320 million more will be spent on WS&S by 1995, of which $85 million has been promised by A.I.D. Despite the magnitude of these investments, the standards used for deciding the capacities of A.I.D.-assisted and other projects (which in turn affect their costs) are based on assumptions and judgments almost totally lacking an empirical basis. Among other reasons, such a hard database is unavailable because rural water systems are typically constructed without meters for measuring water consumption. Starting in fall 1989, meters were installed in 16 rural communities in Ecuador, Guatemala, and Honduras with populations of 100 to 1,200 each. During a period of two months, meter readers in each town collected data on actual consumption for 30 days. The authors made determinations of such parameters as average per capita demand, maximum daily demand, and required storage volumes needed to meet actual demands. They also performed regression annually to develop equations for predicting the key design parameters for towns of any size in any of the three countries. Results are described in a 1992 companion report by this study"s authors, WASH Technical Report No.78, "Deriving Design Standards for Rural Water Systems: Case Studies Using Water Demand Data from Ecuador, Guatemala, and Honduras" (PN-ABN-013). The data in the current study showed that average per capita consumption in Ecuador and Honduras was 50-60 gallons per day (gpd), whereas design standards assume 30 gpd. Average consumption in Guatemala was a little below the design standard. The authors concluded that the Guatemala systems have adequate capacity except in their networks and storage tanks, but the systems in Ecuador and Honduras are grossly underdesigned in most components and, in many cases, suffer capacity deficits almost as soon as they are placed in service. All three countries run the risk of low and negative pressures in their distribution systems due to high demand, which poses a potential threat to health. When system capacity is inadequate, there are two ways to rectify the situation. One is to provide more capacity, which treats present demand as a requirement; the other is to ration capacity through conservation, which treats demand as a variable that can be reduced. The authors chose a typical town with an initial population of 600 to analyze and compare the options of increasing capacity versus rationing. They found that if expansion and higher design standards were chosen for dealing with capacity deficits, per capita construction costs would increase by 50 percent, from $80 to $120. For a given construction budget, this would mean that one-third fewer people could be served with improved water supplies. However, expansion cannot be easily justified on economic grounds because the costs would far outweigh the benefits. This is a consequence of the existing tariff structure, which charges users a nominal monthly flat rate for water service. The preferable option is to ration capacity through conservation, for which five alternatives were considered. The most expensive is metering, which would cost nearly as much as increasing capacity. However, most of the cost would be for personnel, and only $10 more per capita would be required for hardware initially. This option would double the fees that households now pay for water and would work only if communities found it acceptable. The increased cost to donors such as A.I.D. would be modest. A second alternative, rationing with flow restrictors or special faucets that limit flow, would be less expensive, but success stories using these devices are few. They are subject to tampering and bypass and require in-house water storage, which can pose a health risk. Enforcement of rules about nonessential use of water and reduction of waste offers a third alternative for conservation. This option is potentially effective and inexpensive. However, if the current high rates of water use are due to watering gardens and livestock upon which one"s livelihood depends, and for which sources other than piped supplies are not readily available, enforcement of rules may not achieve the desired conservation. The fourth and fifth choices are rationing by planned intermittent supply, and rationing by unplanned shortages. These two options are expensive in nonpecuniary costs and operate contrary to the goal of sustainable development. Therefore, they are unworthy of serious consideration. This report recommends that rationing rather than increasing design flows and expanding capacity be adopted. In communities without meters, rule enforcement should be attempted. If enforcement of rules fails, metering would be the next best option. In communities with meters, redesigning the tariff to be more economical should be done. Existing and new systems should be equipped with master meters in order to develop a database for improving the design of new systems and the operating efficiency of existing systems. Studies to determine community preferences and willingness to pay should be conducted as a basis for future planning. These should examine selecting an appropriate rationing scheme, increasing cost recovery, and promoting greater self-sufficiency. (Author abstract)