Journal Issue

Making Public Infrastructure Entities Commercial

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

AS DEVELOPING countries examine ways of improving the delivery of infrastructure services, reforming public enterprises along commercial lines looks quite promising. The World Development Report 1994 calls for a 3-pronged approach that includes corporatization, explicit contracts, and improved cost recovery.

In spite of impressive progress by developing countries over the last two decades, many people, especially the urban and rural poor, do not yet have access to even minimal infrastructure services. Increasingly, the private sector may be called upon to play a role in improving coverage, but the onus will continue to be on the public sector for several reasons.

First, for the foreseeable future, the public sector is likely to remain responsible for over 80 percent of expenditures on infrastructure. Second, private alternatives are not yet as common in the poorest countries as in middle-income countries. Third, even in countries with a dynamic private sector, some infrastructure services such as road networks and major public works will remain predominantly in the public domain. And fourth, strategic, regulatory, or political reasons pressure governments in many developed and developing countries to retain the responsibility for building and operating infrastructure.

However, as the World Development Report 1994 brings out, governments in developing countries have tended to do poorly in managing the flow of infrastructure services. A survey of World Bank-financed projects designed to improve infrastructure performance reveals the most common problems (see table).

• The goals of public sector infrastructure providers are often ambiguous and changing because governments are forced to balance many, often inconsistent, economic, social, and political objectives.

• Managers do not control day-to-day operations—or decisions on prices, wages, employment, or budgets. Utilities’ prices, for instance, are often imposed on managers regardless of resulting constraints on their ability to pay for inputs, including salaries.

Common management problems in public sector infrastructure entities(percentage of total loans to a sector in which a specific problem was identified, 1980–92)
Source of problem
SectorNumber of


Lack of management

autonomy and accountability

Wages and labor

Source: World Development Report 1994, World Bank.
Source: World Development Report 1994, World Bank.

• Because the financial status of public providers often depends on budgetary decisions unrelated to performance and on pricing decisions driven by politics, financial difficulties are common, in particular in power and water utilities (see chart).

• Governments impose overstaffing on infrastructure agencies to create public sector jobs, passing the additional cost on to taxpayers or consumers, most obviously in transport.

But while problems among public providers appear to be the norm in developing countries (see “Infrastructure for Development,” in this issue), there are also examples of success: Korea and Singapore in most or all sectors, Mexico and Thailand in power, and Botswana and Togo in water supply (see box). Why do they have an enviable performance record? Because they are run on commercial principles like any successful private organization. More specifically, they enjoy clear and coherent goals focused on delivering services, autonomous management and accountable managers and employees, and financial independence—in an environment that includes competition with appropriate regulation.

These commercial principles come naturally to a private business. But for public sector organizations, they often need to be introduced along with three core instruments: corporatization, explicit contracts, and improved cost recovery.


The first step for many countries in explicitly separating infrastructure providers (e.g., power, ports, railways, telecoms, and water entities) from the government is organizing them as public enterprises rather than government departments. Corporatization is the next step, giving a public enterprise a quasi-independent status and subjecting it to the same legal requirements as private firms—standard commercial and tax law, accounting criteria, competition rules, and labor law.

In practice, however, this transformation is often incomplete, because public organizations do not face adequate competition or do not have solely commercial objectives. For example, corporatization should transfer employees from civil service status to contracts governed by ordinary labor law. Yet even under corporate structures, public entities are often reluctant to reduce employment. Also, it takes time and much effort to convert a government department into a public corporation: it took ten years to corporatize fully Indonesia’s major ports. But there are also very quick gains. In Ghana, within two years of the improvement in accounting practices, real operating costs in the state transport corporation were down by 67 percent.

How to run a public utility: Botswana

Since 1970, Botswana has been able to provide potable water to the country’s principal urban areas and operate a financially self-supporting service—one of the few African countries with a safe urban water supply. Indeed, a World Bank review in 1993 of Botswana’s Water Utility Corporation (BWUC) showed that the utility maintains a 24-hour supply of water to all its service areas, with high-quality treatment. Water losses are acceptable, at about 15 percent in the distribution system and 10 percent in the raw-water transmission and treatment processes, reflecting the good quality of BWUC’s engineers, who are attracted by competitive salaries.

BWUC charges commercially oriented tariffs appropriate for the urban conditions in Botswana, not hesitating to adjust prices as needed in order to manage demand. Meters are read and consumers are billed monthly, with 30 days to pay. Supply is cut off immediately if payment is not made, and there are charges for reconnection. Accounts receivable are usually less than 2 percent of all the amounts collectible, attesting to the success of strict billing and collection procedures.

A family of six consuming about 100 liters a day per capita pays about $8.85 a month—about 8 percent of its income. Reducing consumption to 80 liters lowers the water bill to about 5 percent of income. A wealthier family consuming twice that much would pay about $32.25 a month. These charges are high compared with those levied by similar African utilities, but they constrain consumption and ensure that the utility does not have to rely on subsidies from the government or other sectors in the economy.

An innovative feature is the “one-check” system for government users. The Ministry of Finance meets all monthly charges for the government and deducts them from the cash allocations of each ministry or department. This procedure avoids the accumulation of arrears by government users of infrastructure services that is common elsewhere.

Explicit contracts

Organizational changes alone, however, neither provide clear goals nor create incentives for managers to meet these goals. The introduction of market principles can help, although there are limits to the use of markets and to the value of profit maximization as a goal. Providers have problems in recovering costs directly from beneficiaries of services (as with roads), governments have objectives other than profit, or prices have to be regulated outside the supplying entity (as when providers have monopoly powers). When this happens, three types of explicit contracts—which increase autonomy and accountability—can be called upon.

Performance agreements. These agreements, which retain all decisions in the public sector, are negotiated between the government and public managers. They are aimed at increasing the accountability of employees and managers and improving the focus of operations by clarifying performance expectations and the roles, responsibilities, and rewards of all those involved. They force the development of an information and evaluation system for performance monitoring, including standard financial and cost accounting procedures, and detailed quantitative and qualitative indicators (e.g., for roads, this would include measures of the condition of the network, its use, its management, its productivity, and its financing). The information collected can then be used to build in incentives, whether they be pecuniary bonuses (up to 35 percent of total wages in India, Korea, and Mexico), nonpecuniary rewards (award ceremonies or press coverage), or even sanctions for nonperforming staff.

Performance agreements have been often successful in East Asia, thanks to explicit efforts to build incentives for managers and workers into the contracts and to monitor these incentives. In Africa and in many South American countries, however, they have not been as successful, especially with respect to financial targets because incentives were not built in properly. These difficulties explain why many Latin American governments are increasingly privatizing and why most experts suggest that governments who want to maintain ownership of assets should rely more on the alternatives discussed next.

Management contracts. These contracts—which transfer to private providers the responsibility for managing a utility, usually for three to five years—increase the autonomy of management and reduce the risks of political interference in the day-to-day operations of the public entity. They also serve as interim arrangements allowing private firms and public agencies to gain experience with partnerships before engaging in more comprehensive contracts.

Costs are seldom fully recovered in infrastructure

Source: World Development Report 1994, World Bank.

A management contract signed for the power company in Guinea-Bissau is showing that these contracts may work where performance agreements have failed: the new management doubled electricity sales in just three years. The key is to grant the contractor significant autonomy in decisionmaking and relate compensation to performance, such as reduced leakages and increased connections. The contract in Guinea-Bissau specified that 75 percent of the remuneration was guaranteed, but that the remaining 25 percent was based on performance.

Service contracts. These contracts transfer to private suppliers the responsibility for either delivering a specific service at lower costs, or providing expertise that is lacking in the public sector (e.g., design engineering) or too expensive to maintain permanently on public payrolls. The infrastructure supplier sets the performance criteria, evaluates bids from competitive tendering, supervises performance, and pays agreed fees.

Contracting out is most common for maintenance services. Major overhauls of power stations, for example, are routinely contracted out to plant suppliers or specialists in most developing countries. Standard professional services—such as auditing, data processing, and recruitment—are also often contracted out. Catering, cleaning, and ticketing are contracted out in Pakistan’s railway. Meter reading and fee collections in water and sanitation have been handled through service contracts in Chile since the 1970s. Santiago’s public water company even encouraged staff to leave and compete for service contracts.

Contracting out tends to be more cost-effective than using public employees to handle maintenance (known as “force account”). Brazil’s switch to road maintenance by contract reduced costs by some 25 percent for equivalent service quality. Remaining account work groups of the government have also become more efficient since required to compete with private contractors.

Picking the contract. Which of the three types of contracts is the right one depends on the infrastructure activity and the possibilities of contracting services or management out to the private sector. But more important, it depends on the specific cause of poor performance—is the problem with the government, the managers, or the workers?

If the problem is with the government, the performance agreement may be the preferred instrument because these agreements are reciprocal. For instance, they can be used to put pressure on the government to keep its promises to authorize and enforce tariff increases. If the problem is with management, much depends on whether abilities or incentives are in question. Performance agreements with incumbent public managers assume that their capabilities are adequate. Thus, management contracts based on performance are more effective in the short run for organizations with weak management skills. However, for dealing with incentives—and this applies to workers, too—performance agreements need to make a clear link between performance and rewards. Management and service contracts have the added advantage of signaling to civil servants and public managers that if they fail to deliver, there are alternatives in the private sector.

Cost recovery

The third core instrument to successful provision of infrastructure services on a commercial basis is the establishment of reliable revenue sources to ensure the financial autonomy of public agencies. Reliance on revenues directly related to services delivered increases the productivity of suppliers and often benefits users. With fewer budgetary transfers and the freedom to set prices—within the limits defined by regulation—there is less occasion for the government to interfere. Just as important, better cost recovery allows public managers to rely on the price system to assess users’ willingness to pay.

Among public utilities in developing countries, gross revenues typically cover costs only in telecommunications (see chart). Even so, local services are usually underpriced, with the losses made up from significantly above-cost charges for long distance and international service. This difference between tariffs and costs is a type of tax on users. In all other sectors, the gap between revenues and costs implies a government subsidy to users, varying from 20 percent for gas to 70 percent for water. In many of the countries, these subsidies actually benefit the rich more than the poor. In Costa Rica, for example, the richest 20 percent receive 1.3 times as much subsidy for water and sewer services as the poor.

In many countries, the best way of reducing the gap between costs and revenues is to cut costs and achieve productive efficiency. Costs due to poor debt management are excessive in about a third of Bank-supported infrastructure projects. Maintenance problems that cause water or power losses are even more common and costly. In Mexico City at the end of the 1980s, neglect of maintenance and lags between tariff increases and cost increases in the water sector required a federal subsidy of about 0.6 percent of GDP a year.

Once costs are controlled, well-established pricing principles can help achieve financial autonomy and reduce distortions in the allocation of resources. The infrastructure pricing strategy of countries as different as Botswana, Chile, Korea, and Singapore aims at cost recovery sufficient to guarantee the financial independence of public utilities. Their utilities’ prices recover the three main cost components of most infrastructure utilities: connection, usage, and peak capacity costs. The cost of connecting a customer and maintaining that connection to distribution or collection networks is typically levied as a periodic flat fee, often linked to charges based on usage in a two-part tariff.

Cost recovery and the poor. Many governments fear that fully recovering costs will hurt the poor, yet increasing prices may actually help the poor. This is true because prices that enable cost recovery in the delivery of services provide the utilities with the resources and the incentive to expand networks to service the areas where the poor live. In the mid-1970s to early 1980s, people in 17 cities were paying private water vendors an average of 25 times the prices charged by public water utilities. In Nouakchott, Mauritania, and Port-au-Prince, Haiti, vendors were charging up to a hundred times the public utility price. Expanding the utilities network to give the poor access would mean that they would pay less than they are now actually paying to private providers.

Political commitment

Reforming the public sector, however, will not work without political support. This is why the commercialization process has to be designed in a way that increases the odds of this occurring. Explicit or implicit contracts between policymakers and managers or operators can be used to define clear rules for performance and force their monitoring and implementation. Service contracts seem to be the most reliable in this respect. Moreover, they test the capacity of the private sector to contribute to the provision of infrastructure. Performance agreements have been the least successful because they can too easily accommodate discretionary decisions driven by the many conflicting or evolving government interests. Pricing strategies can also play a key role: how subsidies are designed and how they are financed are critical to the political viability of a commercialization strategy.


Emerging Stock Markets Factbook 1994

The all-new 8th edition of the Factbook is more valuable than ever.


  • Data on more than 50 emerging markets and detailed profiles on 25 of the top emerging markets

  • 6 big chapters give an emerging markets overview, a performance review for the past 10 years, and detailed indexes and/or time series data summaries on each market

  • Vital background information on investment conditions such as investor protection, ease of entry and exit, and withholding taxes on overseas investors

  • A complete directory of addresses, communication facilities, senior officials and hours of operation for each market.

  • New easy-to-read format

International Finance Corporation

June 1994. 276 pp. Stock No. 12820. Price $75.00.

Other Resources Citing This Publication