Setting It Up: Financing
How Will Costs be Recovered?

Designing Subsidies

Excerpt from:
World Bank Experience with the Provision of Infrastructure Services for the Urban Poor: Preliminary Identification and Review of Best Practices.
Christine Kessides. Informal Publication by the World Bank. Washington, D.C. January 1997.

In the few past cases where localized [World Bank] projects were extended into national programs, direct cost recovery from households was either negligible (Botswana, Indonesia's KIP) or partial (Tunisia). The key to financial sustainability was the central government’s commitment to subsidies that were manageable - a condition that requires keeping investment costs low. The financial impact on government of such a commitment need not be unreasonable. Estimates based on project experience suggest that an upgrading program covering minimal water, sanitation, access roads and drainage improvements for the entire unserved urban population in LAC, East and South Asia, and Sub-Saharan Africa, based on policies permitting high settlement densities, may require 0.2 - 0.5 percent of GDP annually on average over a fifteen year period. (1) These levels would be feasible as reallocations of existing urban investment in some countries without necessarily claiming additional public resources for the urban sector.

The specific design of a subsidy and its impact on incentives is critical for a program's effectiveness. Several of the recent urban and water/sanitation projects [of the World Bank] have specified subsidies as a fixed amount on a per capita basis, calibrated to the estimated costs of a very basic service level. This design keeps the total subsidy budget within a predictable limit while ensuring that the poorest households and communities will be able to obtain at least minimal services.

Probably the best example of effective subsidy design for water and sanitation is found in Chile, which already has virtually universal coverage for these services but provides a targeted subsidy (a voucher scheme) to ensure that all households can afford the basic consumption level. The program, introduced to replace tariff cross-subsidies in 1990, features an efficient partnership among levels of government and the utility to direct the subsidy to the target population of urban poor.

• The central government provides the subsidy funds and transfers them to the local government, which also administers the means testing.

• The utility charges the municipality for the subsidized portion of the water/sewerage bill of its eligible customers.

• Between 25-80 percent of the costs of a minimum monthly consumption level is covered by the subsidy, and households are held strictly accountable for paying the remainder of the bill, up to 5 percent of their monthly income. (2)

This scheme requires a high quality of administration and has not been replicated to date in other developing countries.

1 Based on the basic level of service (standpipe water, latrine, gravel access road, unlined storm drainage channels) for densities of 360-500 persons/ha. Adding the estimated requirements for incremental primary infrastructure investment would bring this range to about 0.3-0.7 percent of GDP per year. This does not include rehabilitation of deteriorated installations, O&M, or treatment of wastewater. See Christopher Banes, John Kalbermatten, and Piet Nankman, “Infrastructure Provision for the Urban Poor: Assessing the Needs and Identifying the Alternatives”, TWUDR draft, May 1996.

2 Raquel Alfaro Fernandois, “The Introduction of Competition into a Natural Monopoly and of Social Considerations into Entrepreneurial Management: The Case of EMOS”, Paper presented at Regional Seminar on Public and Private Cooperation Alternatives for Urban Development in Latin America and the Caribbean: The Privatization of Basic Urban Services”, Quito, Ecuador, February 15-17, 1995.