Journal Issue

Infrastructure for Development

International Monetary Fund. External Relations Dept.
Published Date:
January 1994
  • ShareShare
Show Summary Details

GOOD infrastructure services are essential to achieve economic growth and to improve the quality of life. But for most developing countries—despite improvements in access—the quantity and quality of services are well below what is demanded. Solutions lie in better institutional incentives, including involving the private sector in service delivery.

Developing countries invest $200 billion a year in new infrastructure—4 percent of their national output and a fifth of their total investment goes to transport, power, telecommunications, water supply, sanitation, and irrigation. The result has been a dramatic increase in infrastructure services (Chart 1). During the past 15 years, the share of households with access to clean water has increased by half, and power production and telephone lines per capita have doubled. All regions have enjoyed per capita gains.

Chart 1Infrastructure has expanded tremendously In recent decades

Note Based on telecom, sanitation, and water data for 1975–90, and road and power data for 1960–90

Source: World Development Report 1994, World Bank.

These accomplishments are no reason for complacency, however. One billion people in the developing world still lack access to clean water, and nearly 2 billion lack adequate sanitation. In rural areas especially, women and children often spend long hours fetching water. The already inadequate transport network is deteriorating rapidly in many countries. Electric power has yet to reach 2 billion people, and in many countries unreliable power constrains output. The demands for telecommunications to modernize production and enhance international competitiveness far outstrip existing capacity. On top of all this, population growth and urbanization are increasing the demand for infrastructure.

Responding to these shortcomings is critical because the adequacy of a country’s infrastructure helps determine its success or failure in diversifying production, expanding trade, coping with population growth, reducing poverty, or protecting the environment. Good infrastructure also raises productivity, lowers production costs, and improves living standards. Indeed, infrastructure capacity grows in step with economic output—a 1 percent increase in the stock of infrastructure is associated with a 1 percent increase in GDP across all countries.

But coping with infrastructure’s future challenges involves much more than a simple numbers game of drawing up inventories of infrastructure stocks and plotting needed investments on the basis of past patterns.

Inefficiency and waste claim a large share of resources that could be used for delivering infrastructure services. A review of power utilities in 51 developing countries shows that technical efficiency has actually declined over the past 20 years. Older power plants consume between 18 and 44 percent more fuel per kilowatt hour than do plants with best “practice” technology. Port facilities in developing countries on average move cargo from ship to shore at only 40 percent the speed of the best-performing ports. And half the labor in African and Latin American railways is estimated to be redundant.

These failings in investment and operating efficiency are not compensated by success in addressing poverty or environmental concerns. Badly designed and managed infrastructure is a source of environmental degradation in both urban and rural areas. The poor not only consume fewer infrastructure services (public transport, sanitation, clean water and power)—they also have less access to services that meet their demands for quality and affordability, and so are often forced to pay more than the rich for inferior service. In most countries, rural areas receive fewer infrastructure services than urban areas (with the obvious exception of irrigation), even in such essential services as drinking water.

Inadequate maintenance has been an almost universal (and costly) failure of infrastructure providers in developing countries. For example, a well-maintained paved road should last for 10–15 years before needing resurfacing, but lack of maintenance can lead to severe deterioration in half that time. An examination of completed Bank highway projects shows that, on average, the estimated economic returns to projects involving primarily maintenance are almost twice as high as returns to projects for new construction. Inadequate maintenance means that power systems in developing countries have, on average, only 60 percent of their generating capacity available while best practice systems in developing countries achieve levels over 80 percent. And water supply systems deliver an average of 70 percent of their output to users—compared with design targets of 85 percent.

Failings in maintenance are often compounded by ill-advised spending cuts during periods of budgetary austerity. Shortfalls in maintenance expenditure have to be compensated for later by much larger expenditures on rehabilitation or replacement. Donor objectives (such as seeking contracts for capital-goods supply or consultancy services) may also play a part in the preference for new investment over maintenance. In many low-income countries, donor financing underwrites nearly half of all public investment in infrastructure.

Misallocated project investments by many countries have created inappropriate infrastructure or provided services at the wrong standard. Demands of users for services of varying quality and affordability go unmet even when users are willing and able to pay for them. Low-income communities are not offered suitable transport and sanitation options that provide services they value and can afford. Premature investments in capacity—especially in water supply, railways, power, ports, and irrigation—have often absorbed resources that could otherwise have been devoted to maintenance, modernization, or improvements in service quality.

This poor performance provides strong reasons for doing things differently—in ways that make better use of new and existing resources to meet the demands of users. In short, the concern needs to broaden from increasing the quantity of infrastructure facilities to improving the quality of infrastructure services. Fortunately, the time is ripe for change. In recent years, there has been a revolution in thinking about who should be responsible for providing infrastructure facilities and services, and how these should be delivered to the user.

Against this background, the World Development Report 1994 (WDR 1994) considers new ways of meeting public needs for services from infrastructure—ways that are more efficient, more user-responsive, more environment-friendly, and more resourceful in using both the public and private sectors. The report reaches two broad conclusions:

• Because past investments in infrastructure have not had the development impact expected, it is essential to improve the effectiveness of investments and the efficiency of service provision.

• Innovations in delivering infrastructure services—along with new technologies—point to solutions that can improve performance.

Why the problems?

As developing countries look for ways to tackle insufficient maintenance, misallocated investment, unresponsiveness to users, and technical inefficiencies, it is critical to understand why these problems have become so pervasive. The WDR 1994 found, perhaps surprisingly, that although the availability of infrastructure (per capita coverage) tends to be correlated with GDP per capita, efficiency and effectiveness of infrastructure provision are not. Consequently, plots of coverage against performance in water, power, telecommunications, roads, and railways show little relationship across a wide sample of low- and middle-income countries. Moreover, there is no close correlation between a country’s efficiency of provision in one sector and its performance in another.

These findings indicate that efficiency and effectiveness of infrastructure provision derive not from general conditions of economic growth and development, but from the institutional environment, which often varies across sectors in individual countries. This suggests that changes in the institutional environment can lead to improved performance, even when incomes are low, because in each sector some low-income countries perform well. As a corollary, even many high-income countries give evidence of waste and inefficiency in infrastructure when the policy and institutional conditions are poor.

What are the main reasons for poor performance? First, the delivery of infrastructure services in the developing world usually takes place in a market structure with one dominating characteristic: the absence of competition. These services often are provided by centrally managed, monopolistic public enterprises or government departments. This holds for almost all irrigation, water supply, sanitation, and transport infrastructure, and it held almost universally for telecommunications until a few years ago.

Second, those charged with delivering infrastructure services are rarely given the needed managerial and financial autonomy. Managers are often expected to meet objectives at variance with what should be their primary function—the efficient delivery of high-quality services to satisfy customer demand. Public entities are required to serve as employer of last resort or to provide patronage, besides being compelled to deliver services below cost. At the same time, public providers are rarely held accountable for their actions. Few countries set well-specified performance measures, and inefficiency is often compensated for by budgetary transfers.

Third, the users of infrastructure—both actual and potential—are not well positioned to make their demands felt, since prices typically do not reflect costs. For example, power prices in developing countries cover only half the supply costs on average. Water charges and rail passenger fares typically cover only a third of costs. Excess consumer demand based on below-cost prices is not a reliable indicator that services should be expanded, though often it is taken as such. Users can express preferences in other ways (e.g., local participation in planning and implementing new infrastructure investments), but they seldom are asked.

Creating the institutional and organizational conditions that oblige suppliers of infrastructure services to be more efficient and more responsive to the needs of users is clearly the challenge. But is it possible? Several converging forces are opening a window of opportunity.

• Technology is changing the way infrastructure can be provided in most sectors. In telecommunications and power generation, technological innovation has eroded economies of scale; electronic devices permit remote-controlled metering of electricity, water, and gas use (and eventually may permit more direct road pricing); a variety of new technologies, such as digitalization in telecommunications, reduce the cost and complexity of maintenance; and computerized systems facilitate network operations.

• Economic adjustment in many countries through the 1980s has led to a new pragmatism regarding the roles of government and of the market. The worldwide liberalization of markets, and insights about the potential for institutional reforms have changed attitudes about the relative risks from market and government failures.

• The shift toward more pluralistic and decentralized governance in many developing countries has strengthened public pressures for more affordable and environmentally friendly solutions in infrastructure.

These trends have been reinforced by a surge of initiatives to provide and finance infrastructure in new ways.

Options for the future

For a more effective and efficient delivery of infrastructure services, the WDR 1994 advocates three elements of institutional and policy reform:

Apply commercial principles of operation by giving service providers focused and explicit performance objectives, well-defined budgets based on revenues from users, and managerial and financial autonomy, while also holding them accountable for their performance. Governments should therefore refrain from ad hoc interventions in management, but provide explicit transfers, where needed, to meet social objectives, such as public service obligations. Private sector involvement in management, financing, or ownership will in most cases be needed to ensure a lasting commercial orientation in infrastructure.

Broaden competition by arranging for suppliers to compete for an entire market (e.g., firms bidding for the exclusive right to operate a port for ten years), for customers within a market (telephone companies competing to serve users), and for contracts to provide inputs to a service provider (firms bidding to provide power to an electric utility).

Involve users more in project design and operation of infrastructure, especially where commercial and competitive practices are insufficient to provide the information needed to make suppliers more accountable to their customers. Involvement of users and other stakeholders can include consultation during project planning, direct participation in operation or maintenance, and monitoring.

These elements apply whether infrastructure services are provided by the public sector, the private sector, or public-private partnership. However, numerous examples of past failures in public provision, combined with growing evidence of more efficient and user-responsive private provision, argue for a significant increase in private involvement in financing, operation, and—in many cases—ownership.

How rapidly and extensively private involvement can increase in any country depends on the strength of the private sector, the administrative capacity of the government to regulate private suppliers, the performance of public sector providers, and the political consensus for private provision. With this in mind, the WDR 1994 sets out a menu of four main options for ownership and provision, which must be tailored to fit each country’s needs:

Option A. Public ownership and public operation. Public provision by a government department, public enterprise, or parastatal authority is the most common form of infrastructure ownership and operation. Successful public entities run on commercial principles and give managers control over operations and freedom from political interference, but they also hold managers accountable by using performance agreements or management contracts. And they follow sound business practices and are subject to the same regulatory, labor law, accounting, and compensation standards and practices as private firms. Tariffs are set to cover costs, and any subsidies to the enterprise are given for specific services and in fixed amounts. Water authorities in Botswana and Togo and national power companies in Barbados and Thailand perform well, as do the highway authorities in Ghana and Sierra Leone and the restructured road agency in Tanzania. But few successful cases of Option A persist, because of their vulnerability to changes in governmental support.

Option B. Public ownership with private operation. This option is typically implemented through lease contracts for full operation and maintenance of publicly owned infrastructure facilities—or through concessions, which include responsibility for the construction and financing of new capacity. The private operator typically assumes all commercial operation risk and shares in investment risk under concessions.

Leases and concessions are underway for railways in Argentina, water supply in Buenos Aires and Guinea, and port facilities in Colombia, Ghana, and the Philippines. Concessions to build and operate facilities include toll roads in China, Malaysia, and South Africa; power plants in Colombia, Guatemala, and Sri Lanka; water and sanitation facilities in Malaysia and Mexico; and telephone facilities in Indonesia, Sri Lanka, and Thailand.

Option C. Private ownership and private operation. The use of this option is increasing, both through new entry by private firms in infrastructure markets and through divestiture of public ownership of entire systems. Private ownership is straightforward when services can be provided competitively. For example, 27 developing countries allow cellular telephone service to be competitively provided, and many others allow private firms to construct electric generating plants and sell power to the national power grid. The necessary competition can also occur across sectors—between road and rail, and electricity and gas.

Where systems are being fully or partly privatized and there is no cross-sectoral competition, regulation of private and public providers may be required to prevent monopoly power abuses. Experience with regulation and systemwide privatization in developing countries is still new. Chile’s form of regulation, which involves regular, automatic price adjustments and a well-specified arbitration system, appears to be working well.

Option D. Community and user provision. This option is most common for local, small-scale infrastructure—such as rural feeder roads, community water supply and sanitation, distribution canals for irrigation, and maintenance of local drainage systems—and it often complements publicly provided services. Successful community provision needs user involvement in decisionmaking, especially to set expenditure priorities so as to ensure an equitable and agreed sharing of the benefits and costs. Technical assistance, training, and compensation to service operators are also important. When these elements are present, community self-help programs can succeed over long periods, as in the Gurage community roads organization in Ethiopia, which has handled local maintenance since 1962.

Financing and payoffs

Where will the money come from to implement these options? Private financing in one form or another now accounts for about 7 percent of total infrastructure financing in developing countries, and its share may double by the year 2000, while bilateral and multilateral foreign aid accounts for around another 12 percent (about a third of the latter comes from the World Bank and the International Development Association). Although an increasing share of the domestic savings needed to finance infrastructure provision can come from private sources, governments will continue to be a major source of funds for infrastructure as well as a conduit for resources from the donor community. As transitional measures to provide long-term financing where sufficient private support is not likely, some governments are revitalizing existing infrastructure lending institutions and creating specialized infrastructure funds.

In future, governments will often need to be partners with private entrepreneurs. The lessons of this experience to date are that these partnerships should start with small, simple projects; investors’ returns should be linked to project performance; and any government guarantees needed should be carefully scrutinized. The task is to find ways to route private savings directly to private risk-bearers making long-term investments in infrastructure projects that each have unique characteristics and for which no single financing vehicle is appropriate. Official sources of finance, such as multilateral lending institutions, can facilitate the process by supporting the policy and institutional reforms needed to mobilize private financing and use it more efficiently.

The payoffs from increasing the efficiency of infrastructure provision will differ from country to country and from sector to sector, given the great variation in performance. But the rewards are potentially large across the spectrum, making the commitment to reform very worthwhile. First, a reduction in subsidies in power, water, and railways alone would reduce the fiscal burden by nearly $123 billion annually—or about 10 percent of total government revenues in developing countries, 60 percent of annual infrastructure investment, and approximately five times annual development finance for infrastructure (Chart 2). Though eliminating underpricing would not produce a net resource saving to the economy (as the costs would be covered by users), the fiscal relief would be substantial.

Chart 2Potential payoffs from reform Annual gains from eliminating mispricing and inefficiency are large relative to investment

Source: World Development Report 1994, World Bank.

1 Costs for the water sector are due to leakages: for railways-fuel inefficiency, overstating, and locomotive unavailability; for roads-added investment caused by poor maintenance; for power-transmission, distribution, and generation losses.

Second, service providers stand to gain from better technical efficiency. The savings possible from raising operating efficiency from today’s levels to best practice levels are estimated at around $55 billion a year—pure savings equivalent to 1 percent of all developing countries’ GDP, a quarter of annual infrastructure investment, and twice annual development finance for infrastructure. Looked at another way, if the annual technical losses of $55 billion could be redirected for three years—at current costs of roughly $150 per person for water systems—the 1 billion people without safe water could be served.

Finally, improvements in productivity and pricing would permit more effective delivery of service in response to demand and enhance the growth and competitiveness of the economy. They would also allow vastly greater mobilization of resources for needed new investments—by generating higher revenues and by creating a policy environment conducive to the inflow of new resources.

The World Development Report 1994 has been prepared by a team led by Gregory K. Ingram and comprising John Besant-Jones, Antonio Estache, Christine Kessides, Peter Lanjouw, Ashoka Mody, and Lant Pritchett. The work was carried out under the general direction of Michael Bruno.

Other Resources Citing This Publication