"-ra: FOR ( -- TI5I WoBE n994 Eng ,1. World Development Report 1994 Infrastructure for Development Executive Summary The World Bank Washington, D.C. iii 6 Setting priorities and implementing reform lr% Choosing among alternatives: institutional options and country conditions Sectoral agendas for reform Payoffs from reform -E . * Bibliographical note Appendix: Infrastructure data World Development Indicators iv A Overview Developing countries invest $200 billion a year in performing power sectors in low-, middle-, and new infrastructure-4 percent of their national out- high-income countries. Half the labor in African put and a fifth of their total investment. The result and Latin American railways is estimated to be re- has been a dramatic increase in infrastructure ser- dundant. And in Africa and elsewhere, costly in- vices-for transport, power, water, sanitation, tele- vestments in road construction have been wasted communications, and irrigation. During the past fif- for lack of maintenance. teen years, the share of households with access to This poor performance provides strong reasons clean water has increased by half, and power pro- for doing things differently-in more effective, less duction and telephone lines per capita have dou- wasteful wa.ys. In short, the concern needs to bled. Such increases do much to raise productivity broaden from increasing the qluantity of infrastruc- and improve living standards. ture stocks to improving the quiality of infrastructure These accomplishments are no reason for com- services. Fortunately, the time is ripe for change. In placency, however. One billion people in the devel- recent years there has been a revolution in thinking oping world still lack access to clean water-and about who should be responsible for providing in- nearly 2 billion lack adequate sanitation. In rural frastructure stocks and services, and how these ser- areas especially, women and children often spend vices should be delivered to the user. long hours fetching water. Already-inadequate Against this background, World Development Re- transport networks are deteriorating rapidly in port 1994 considers new ways of meeting public many countries. Electric power has yet to reach 2 needs for services from infrastructure (as defined in billion people, and in many countries unreliable Box 1)-ways that are more efficient, more user- power constrains output. The demands for telecom- responsive, more environment-friendly, and more munications to modernize production and enhance resourceful in using both the public and private sec- international competitiveness far outstrip existing tors. The report reaches two broad conclusions: capacity. On top of all this, population growth and * Because past investments in infrastructure urbanization are increasing the demand for infra- have not had the development impact expected, it is structure. essential to improve the effectiveness of investments Coping with infrastructure's future challenges and the efficiency of service provision. involves much more than a simple numbers game * Innovations in the means of delivering infra- of drawing up inventories of infrastructure stocks structure services-along with new technolo- and plotting needed investments on the basis of gies-point to solutions that can improve perfor- past patterns. It involves tackling inefficiency and mance. waste-both in investment and in delivering ser- This Report marshals evidence in support of vices-and responding more effectively to user these conclusions-identifying causes of failure demand. On average, 40 percent of the power-gen- and examining alternative approaches. The main erating capacity in developing countries is unavail- messages and policy options are summarized in able for production, twice the rate in the best- Box 2. Infrastructure's role and record Box 1 What is infrastructure? This Report focuses on economic infrastriuctiure and The adequacy of infrastructure helps determine one includes services from: country's success and another's failure-in diversi- * Public utilities-power, telecommunications, fying production, expanding trade, coping with piped water supply, sanitation and sewerage, solid population growth, reducing poverty, or improving waste collection and disposal, and piped gas. environmental conditions. Good infrastructure * Public works-roads and major dam and raises productivity and lowers production costs, but canal works for irrigation and drainage. * Other transport sectors-urban and interur- it has to expand fast enough to accommodate ban railways, urban transport, ports and water- growth. The precise linkages between infrastructure ways, and airports. and development are still open to debate. However, Infrastructure is an umbrella term for many ac- infrastructure capacity grows step for step with eco- tivities referred to as "social overhead capital" by nomic output-a 1 percent increase in the stock of such development economists as Paul Rosenstein- infrastructure is associated with a 1 percent increase Rodan, Ragnar Nurkse, and Albert Hirschman. in gross domestic product (GDP) across all countries Neither term is precisely defined, but both encom- Nete ter iprcsldend, btohenm-(Figure 1). And as countries develop, infrastructure pass activities that share technical features (such as economies of scale) and economic features (such as must adapt to support changing patterns of de- spillovers from users to nonusers). mand, as the shares of power, roads, and telecom- munications in the total stock of infrastructure in- Box 2 Main messages of World Development Report 1994 Infrastructure can deliver major benefits in economic . ., ti meeting their demands and puts pressure on suppliers poverty alleviation, and environimenital suistainability-but to be efficient and accountable to users. Competition can only when it provides services that respond . . t' ,. demanid be introduced directly, by liberalizing entry into activi- and does so, r,. ,, til,. Service is the goal and the measure ties that have no technological barriers, and indirectly, of development in infrastructure. Major investments through competitive bidding for the right to provide ex- have been made in infrastructure stocks, but in too many clusive service where natural monopoly conditions exist developing countries these assets are not generating the and by liberalizing the supply of service substitutes. quantity or the quality of services demanded. The costs * Give users and other stakeholders a strong voice and of this waste-in forgone economic growth and lost op- real responsibility. Where infrastructure activities involve portunities for poverty reduction and environmental im- important external effects, for good or bad, or where provement-are high and unacceptable. market discipline is insufficient to ensure accountability The cauises of past poor performance, and the source of im- to users and other affected groups, governments need to proved performance, lie in the incentives facing providers. To address their concerns through other means. Users and ensure efficient, responsive delivery of infrastructure other stakeholders should be represented in the plan- services, incentives need to be changed through the ap- ning and regulation of infrastructure services, and in plication of three instruments-commercial manage- some cases they should take major initiatives in design, ment, competition, and stakeholder involvement. The operation, and financing. roles of government and the private sector must be Public-private partnerships in finalicing have promnise. transformed as well. Technological innovation and ex- Private sector involvement in the financing of new ca- periments with alternative ways of providing infrastruc- pacity is growing. The lessons of this experience are that ture indicate the following principles for reform: governments should start with simpler projects and gain * Manage infrastriucture like a business, not a buireau- experience, investors' returns should be linked to project cracy. The provision of infrastructure needs to be con- performance, and any government guarantees needed ceived and run as a service industry that responds to should be carefully scrutinized. customer demand. Poor performers typically have a con- Governments will have a continuing, if changed, role in fusion of objectives, little financial autonomy or financial infrastructure. In addition to taking steps to improve the discipline, and no "bottom line" measured by customer performance of infrastructure provision under their di- satisfaction. The high willingness to pay for most infra- rect control, governments are responsible for creating structure services, even by the poor, provides greater op- policy and regulatory frameworks that safeguard the in- portunity for user charges. Private sector involvement in terests of the poor, improve environmental conditions, management, financing, or ownership will in most cases and coordinate cross-sectoral interactions-whether ser- be needed to ensure a commercial orientation in infra- vices are produced by public or private providers. Gov- structure. ernments also are responsible for developing legal and * Introdiuce coinpetition-directly if feasible, indirectly if regulatory frameworks to support private involvement not. Competition gives consumers choices for better in the provision of infrastructure services. 2 crease relative to those of such basic services as labor force) and produce more than a third of na- water and irrigation (Figure 2). tional output. The kind of infrastructure put in place also deter- Infrastructure services that help the poor also mines whether growth does all that it can to reduce contribute to environmental sustainability. Clean poverty. Most of the poor are in rural areas, and the water and sanitation, nonpolluting sources of growth of farm productivity and nonfarm rural em- power, safe disposal of solid waste, and better man- ployment is linked closely to infrastructure provi- agement of traffic in urban areas provide environ- sion. An important ingredient in China's success mental benefits for all income groups. The urban with rural enterprise has been a minimum package poor often benefit most directly from good infra- of transport, telecommunications, and power at the structure services because the poor are concentrated village level. Rural enterprises in China now em- in settlements subject to unsanitary conditions, haz- ploy more than 100 million people (18 percent of the ardous emissions, and accident risks. And in many Figure 1 As a country's income grows, the amount of infrastructure increases. Infrastructure stocks per capita, 1990 (1985 prices) 9,000 Norway' 0 Spain 1,200 Paraguay Japan ; g g WMexico Zamnbia * - MAustralia 160 - a I ador Niala 11* - Gtiattcniala Bangladtesh Chad l * Rwanda 20 - 150 400 1,000 3,000 8,000 22,000 GDP per capita, 19901 iPPP dollars)a * Middle East and North Africa 0 Sub-Saharan Africa * Latin America and the Caribbean * South Asia East Asia and Pacific 0 E u rope and Central Asia Note: Axes are I, X, a r thh mic. in ir.i. stIctur; n;!udt- rold . rail, power, irrgatrion. and telephones. a. Purcha-ni pn o%eer parith (PPP) dollars are valued in Summers and Heston 1985 intemational prices. Source: Ingram and Fay, background paper; Summers and Heston 1991. 3 vestments have often been misallocated- too much Figure 2 The composition of infrastructure to new investment, not enough to maintenance; too changes with country income level. much to low-priority projects, not enough to essen- tial services. The delivery of services has been ham- pered by technical inefficiency and outright waste. Percent And too few investment and delivery decisions 100 - have been attentive to meeting the varied demands of different user groups, or to the consequences for the environment. 75 - W i;l Inadequate mainitenianice has been an almost uni- versal (and costly) failure of infrastructure pro- viders in developing countries. For example, a well- maintained paved road surface should last for ten to 50 - /fifteen years before needing resurfacing, but lack of maintenance can lead to severe deterioration in half that time. The rates of return from World Bank- 25 - assisted road maintenance projects are nearly twice those of road construction projects. Timely mainte- nance expenditures of $12 billion would have saved road reconstruction costs of $45 billion in Africa in 0 Low- Middle- High- the past decade. On average, inadequate mainte- income income income nance means that power systems in developing countries have only 60 percent of their generating capacity available at a given time, whereas best S Sanitation OI Telecom practice would achieve levels over 80 percent. And * Water * Roads it means that water supply systems deliver an aver- * Railways * Power age of 70 percent of their output to users, compared I Irrigation with best-practice delivery rates of 85 percent. Poor maintenance can also reduce service quality and in- Source: Ingram and Fay, background paper. crease the costs for users, some of whom install backup generators or water storage tanks and pri- vate wells. Failings in maintenance are often compounded rapidly growing cities, infrastructure expansion is by ill-advised spending cuts. Curbing capital spend- lagging behind population growth, causing local ing is justified during periods of budgetary auster- environments to deteriorate. ity, but reducing maintenance spending is a false In developing countries, governments own, op- economy. Such cuts have to be compensated for erate, and finance nearly all infrastructure, primar- later by much larger expenditures on rehabilitation ily because its production characteristics and the or replacement. Because inadequate maintenance public interest involved were thought to require shortens the useful life of infrastructure facilities monopoly-and hence government-provision. and reduces the capacity available to provide ser- The record of success and failure in infrastructure is vices, more has to be invested to produce those ser- largely a story of government's performance. vices. Donor objectives (such as seeking contracts Infrastructure's past growth has in some respects for capital-goods supply or consultancy services) been spectacular. The percentage of households and may also play a part in the preference for new in- businesses served has increased dramatically, espe- vestment over maintenance. In many low-income cially in telephones and power (Figure 3). The per countries, donor financing underwrites nearly half capita provision of infrastructure services has in- of all public investment in infrastructure. creased in all regions; the greatest improvements Project investmenits utisallocated by many countries have been in East Asia and the smallest in Sub- have created inappropriate infrastructure or pro- Saharan Africa, reflecting the strong association vided services at the wrong standard. Demands of between economic growth and infrastructure. users for services of varying quality and affordabil- In other important respects, however, the perfor- ity go unmet even when users are willing and able mance has been disappointing. Infrastructure in- to pay for them. Low-income communities are not 4 Figure 3 Infrastructure has expanded tremendousiv in recent decades. A,v erage decade rate of gro.% th I percent I 0 10 20 30 40 50 60 Pot% er (production per capita) Telecom !marn lines per capita) Saniitation 'share of population w% ith acce,i-) Pav,ed roads ikilometers per capita - ~Va ter iShare ot population %u ith ace,ssl M Nliddle-incorne countries U Low-income countries Noetl Ba,ed on telcom ,anitaton and water data br l'C5-'k'. an road and pno er data r.r 19mi*-'! 5,.'Urtr .Appendi' tables A I and A_ 2 offered suitable transport and sanitation options levels-and have transmission and distribution that provide services they value and can afford. Pre- losses two to four times greater. Port facilities in de- mature investments in capacity-especially in veloping countries, on average, move cargo from water supply, railways, power, ports, and irriga- ship to shore at only 40 percent the speed of the tion-have often absorbed resources that could oth- most efficient ports. Labor misallocations present erwise have been devoted to maintenance, modern- another source of inefficiency. Overstaffing is far too ization, or improvements in service quality. Because common in many activities, especially railways, many infrastructure investments are immobile and while others, such as road maintenance, warrant serve local markets, excess capacity cannot serve greater use of labor-based methods. other markets-and it remains underused. In some These failings in investment and operating effi- cases, large public projects have been overambi- ciency are not compensated for by success in ad- tious, placing a costly burden on the economy. dressing poverty or environmental concerns-for Waste niod ,,, .. . ,,-. claim a large share of re- here, too, the infrastructure record is weak. Badly sources that could be used for delivering infrastruc- designed and managed infrastructure is a major ture services. A review of power utilities in fifty-one source of environmental degradation in both urban developing countries showed that technical effi- and rural areas. The poor often consume fewer in- ciency has actually declined over the past twenty frastructure services and pay higher prices than do years. Older power plants consume between 18 and the nonpoor. For example, households obtaining 44 percent more fuel per kilowatt-hour than do water from vendors pay much more than those plants in power systems operating at best-practice households connected to water systems. In most 5 Diagnosing the causes of poor performance Figure 4 Urban populations have better access to safe drinking water than The problems of insufficient maintenance, misallo- rural populations. cated investment, unresponsiveness to users, and technical inefficiencies present daunting challenges for future reforms-challenges compounded by Percentage of rural new demands and constrained resources. The solu- population served tions lie in the successes and failures of policy and lull1 in the lessons from recent policy experiments. There is great variation both within and across S() -countries in the efficiency of providing infrastruc- hO - @ture services. Moreover, good performance by a . . - r country in one infrastructure sector is not necessar- 41t - * *S d ily associated with good performance in other sec- * . . * :. ,* , tors. Some developing countries-and not always 20 . the richer ones-perform at high levels. C6te I I *i d'Ivoire meets the 85 percent best-practice standard 0 in water supply, while in Manila only about 50 per- 0 20 40 6') SO 100, cent of treated water is delivered to customers. In Prcnentage or urban I . population served railways, the availability of locomotives is high o se where maintenance is good: at any given time, India has 90 percent of its locomotives available. Avail- * Nliddle East and North Africa ability is low where maintenance is neglected: 50 * Latin America and the Caribbean percent in Romania and 35 percent in Colombia, East Asia and Pacific compared with a developing country average of * Sub-Saharan Africa about 70 percent. For telephones, call completion * South A.ia rates are 99 percent in the best-performing coun- * Europe and Central Asia tries, 70 percent in the average developing country, and far lower in some. These findings indicate that the performance of infrastructure derives not from Source: Appendid. table A.2. general conditions of economic growth and devel- opment but from the institutional environment, which often varies across sectors within individual countries. countries, rural areas receive fewer infrastructure Therefore, to understand what accounts for good services than do urban areas (with the obvious ex- performance-and bad-requires understanding ception of irrigation), even in such essential services the institutional arrangements for providing infra- as drinking water (Figure 4). But countries that structure services and the incentives governing their have made concerted efforts to provide infrastruc- delivery. This Report identifies three reasons for ture in rural areas-for example, Indonesia and poor performance. Malaysia-have succeeded in reducing poverty First, the delivery of infrastructure services usu- dramatically. ally takes place in a market structure with one dom- Given this mixed performance, improvements in inating characteristic: the absence of competition. investment and operation are required as a matter Most infrastructure services in the developing of urgency. In addition, the demands on infrastruc- world are provided by centrally managed monopo- ture are growing. More competitive global trade re- listic public enterprises or government depart- quires more reliable and sophisticated transport, ments. Almost all irrigation, water supply, sanita- power, and telecommunications. Governments fac- tion, and transport infrastructure is provided in this ing increased fiscal stringency can no longer sustain manner. Until a few years ago, telephone services in open-ended financing of infrastructure. And soci- most countries were the responsibility of a state- eties today hold infrastructure to higher environ- owned post, telephone, and telegraph enterprise. mental standards, as evidenced by sections of The bulk of power has also been provided by a pub- Agenda 21, the primary policy document agreed to lic monopoly. As a result, the pressure that competi- by countries at the 1992 United Nations Conference tion can exert on all parties to perform at maximum on Environment and Development. efficiency has been lacking. 6 Second, those charged with responsibility for de- to be more efficient and more responsive to the livering infrastructure services are rarely given the needs of users is clearly the challenge. But is it pos- managerial and financial autonomy they need to do sible? Three converging forces are opening a win- their work properly. Managers are often expected to dow of opportunity for fundamental changes in the meet objectives at variance with what should be way business is done. First, important innovations their primary function-the efficient delivery of have occurred in technology and in the regulatory high-quality services. Public entities are required to management of markets. Second, a consensus is serve as employer of last resort or to provide pa- emerging on a larger role for the private sector in in- tronage. They are compelled to deliver services frastructure provision, based in part on recent expe- below cost-often by not being allowed to adjust rience with new initiatives. Third, greater concern prices for inflation. The other side of the coin is that now exists for environmental sustainability and for public providers are rarely held accountable for poverty reduction. their actions. Few countries set well-specified per- New technology and changes in the regulatory formance measures for public providers of infra- management of markets create new scope for intro- structure services, and inefficiency is all too often ducing competition into many infrastructure sec- compensated for by budgetary transfers rather than tors. In telecommunications, satellite and micro- met with disapproval. wave systems are replacing long distance cable Third, the users of infrastructure-both actual networks, and cellular systems are an emerging and potential-are not well positioned to make alternative to local distribution networks. These their demands felt. When prices reflect costs, the changes erode the network-based monopoly in strength of consumer demand is a clear signal of telecommunications and make competition possi- what should be supplied. Through the price mecha- ble. In power generation, too, combined-cycle gas nism, consumers can influence investment and pro- turbine generators operate efficiently at lower out- duction decisions in line with their preferences. But put levels, while other innovations are reducing prices of infrastructure services typically do not re- costs. New technology makes competition among flect costs, and this valuable source of information suppliers technically feasible, and changes in regu- about consumer needs is lost. For example, power lations are making competition a reality by allowing prices in developing countries have generally fallen, competitive entry in activities such as cellular while costs have not. As a result, prices now cover phone service or power generation. Technical and only half the supply costs, on average. Water regulatory change in other infrastructure sectors- charges and rail passenger fares typically cover only ranging from transportation to water supply and a third of costs. Excess consumer demand based on drainage and irrigation-also make them more below-cost prices is not a reliable indicator that ser- open to new forms of ownership and provision. vices should be expanded, although often it is taken Alongside such changes are new perceptions of as such. the role of government in infrastructure. An aware- Users can express preferences in other ways, ness is growing in many countries that government such as local participation in planning and imple- provision has been inadequate. Brownouts and menting new infrastructure investments. But they blackouts in power systems, intermittent water sup- seldom are asked, and investment decisions are all plies from municipal systems, long waiting periods too often based on extrapolations of past consump- for telephone service connection, and increasing tion rather than on true assessments of effective de- traffic congestion provoke strong reactions. Reforms mand and affordability. in some industrial countries have increased the Individually, each of these three points is impor- competition in telecommunications, in road freight tant. Together, they go a long way toward explain- and airline transport, and in power generation- ing the disappointing past performance of much proving that alternative approaches are possible. infrastructure. Rival suppliers and infrastructure The poor performance of planned economies has users might have exerted pressure for better ser- also provoked a reassessment of the state's role in vices, but they were prevented from doing so. Gov- economic activity. ernments-by confusing their roles as owner, regu- These developments have led governments to lator, and operator-have failed to improve service search for new ways to act in partnership with the delivery. private sector in providing infrastructure services. New opportunities and initiatives Most dramatic have been the privatizations of such enterprises as the telephone system in Mexico and Creating the institutional and organizational condi- the power system in Chile. Elsewhere, various tions that oblige suppliers of infrastructure services forms of partnership between government and the 7 private sector have evolved. Port facilities have bidding for the exclusive right to operate a port for been leased to private operators-the Kelang con- ten years), for customers within a market (tele- tainer facility in Malaysia being among the first. phone companies competing to serve users), and Concessions have been granted to private firms, for contracts to provide inputs to a service provid- particularly in water supply; C6te d'Ivoire is one of er (firms bidding to provide power to an electric the earliest examples. Contracting out services, as utility). Kenya has done with road maintenance, is well Involving users more in project design and opera- under way in many countries. Private financing of tion of infrastructure activities where commercial new investment has grown rapidly through build- and competitive behavior is constrained provides operate-transfer (BOT) arrangements under which the information needed to make suppliers more ac- private firms construct an infrastructure facility countable to their customers. Users and other stake- and then operate it under franchise for a period of holders can be involved in consultation during pro- years on behalf of a public sector client. This ap- ject planning, direct participation in operation or proach has been used to finance the construction of maintenance, and monitoring. Development pro- toll roads in Mexico and power-generating plants grams are more successful when service users or the in China and the Philippines. affected community has been involved in project An increasing regard for the environmental sus- formulation. User participation creates the appro- tainability of development strategies and a deepen- priate incentives to ensure that maintenance is car- ing concern for poverty reduction after a decade of ried out in community-based projects. stagnation in many regions of the world also give These elements apply whether infrastructure ser- impetus to infrastructure reform. Creating pressures vices are provided by the public sector, the private for change, environmental issues are coming to the sector, or a public-private partnership. To this ex- fore in transport (traffic congestion and pollution), tent, they are indifferent to ownership. However, irrigation (increased waterlogging and salinity of numerous examples of past failures in public provi- agricultural land), water supply (depleted re- sion, combined with growing evidence of more effi- sources), sanitation (insufficient treatment), and cient and user-responsive private provision, argue power (growing emissions). At the same time, a for a significant increase in private involvement in decade of reduced economic growth-especially in financing, operation, and-in many cases-owner- Latin America and Sub-Saharan Africa-shows ship. that poverty reduction is not automatic and that All countries will not be able to increase private care must be taken to ensure that infrastructure both involvement at the same rate. Much depends on the accommodates growth and protects the interests of strength of the private sector, the administrative ca- the poor. pacity of the government to regulate private suppli- ers, the performance of public sector providers, and Options for the future the political consensus for private provision. With this in mind, the Report sets out a menu of four To reform the provision of infrastructure services, main options for ownership and provision: this Report advocates three measures: the wider ap- plication of commercial principles to service pro- Option A. Public ownership and operation by viders, the broader use of competition, and the in- enterprise or department creased involvement of users where commercial Option B. Public ownership with operation con- and competitive behavior is constrained. tracted to the private sector Applying commercial principles of operation in- Option C. Private ownership and operation, volves giving service providers focused and explicit often with regulation performance objectives, well-defined budgets based Option D. Community and user provision. on revenues from users, and managerial and finan- cial autonomy-while also holding them account- Far from exhaustive, these four options merely illus- able for their performance. It implies that govern- trate possible points in a broader array of alterna- ments should refrain from ad hoc interventions in tives. management but should provide explicit transfers, Option A: Public ozwnershzip and public operation. where needed, to meet social objectives such as pub- Public provision by a government department, pub- lic service obligations. lic enterprise, or parastatal authority is the most Broadening competition means arranging for sup- common form of infrastructure ownership and op- pliers to compete for an entire market (e.g., firms eration. Successful public entities run on commer- 8 cial principles and give managers control over oper- vided, and many others allow private firms to con- ations and freedom from political interference, but struct electricity-generating plants and sell power to they also hold managers accountable, often through the national power grid. Where competition among performance agreements or management contracts. suppliers is possible, private ownership and opera- And they follow sound business practices and are tion require little or no economic regulation beyond subject to the same regulatory, labor law, account- that applied to all private firms. The necessary com- ing, and compensation standards and practices as petition can also occur across sectors-between private firms. Tariffs are set to cover costs, and any road and rail, or between electricity and gas. For ex- subsidies to the enterprise are given for specific ser- ample, because it competes with suppliers of other vices and in fixed amounts. Water authorities in energy sources, the private gas company in Hong Botswana and Togo and national power companies Kong has no special economic regulation. in Barbados and Thailand perform well. The high- Where systems are being fully or partly priva- way authorities in Ghana and Sierra Leone and the tized and there is no cross-sectoral competition, reg- restructured road agency in Tanzania are promising ulation of both private and public providers may be examples of this approach. But few successful ex- required to prevent the abuse of monopoly power. amples of Option A persist because they are vulner- Experience with regulation and with systemwide able to changes in governmental support. Many privatization in developing countries is still very public entities perform well for a time and then fall new. The Chilean form of regulation, which involves victim to political interference. regular, automatic price adjustments and a well- Option B: Plublic ozwnership zwithi private operation. specified arbitration system, appears to be working This option is typically implemented through lease well. And systems that have been privatized have contracts for full operation and maintenance of pub- been very successful at expanding service. Venezue- licly owned infrastructure facilities, or through con- la's telephone company expanded its network by 35 cessions, which include responsibility for construc- percent in the first two years after its privatization; tion and financing of new capacity. Arrangements Chile's by 25 percent a year, Argentina's by 13 per- between the owner (government) and the operator cent a year, and Mexico's by 12 percent a year. (firm) are set out in a contract that includes any reg- Option D: Communiiity and tuser provision. Commu- ulatory provisions. The private operator typically nity and user provision is most common for local, assumes all commercial risk of operation and shares small-scale infrastructure-such as rural feeder in investment risk under concessions. Leases and roads, community water supply and sanitation, dis- concessions are working well for railways in Ar- tribution canals for irrigation, and maintenance of gentina; for water supply in Buenos Aires and local drainage systems-and it often complements Guinea; and for port facilities in Colombia, Ghana, central or provincial services. Successful community and the Philippines. Concessions also include con- provision requires user involvement in decision- tracts to build and operate new facilities under the making, especially to set priorities for expenditures BOT arrangement and its variants. Proliferating in and to ensure an equitable and agreed sharing of the recent years, concessions to build and operate facili- benefits and costs of service provision. Technical as- ties include toll roads in China, Malaysia, and South sistance, training, and compensation of service oper- Africa; power plants in Colombia, Guatemala, and ators are also very important. When these elements Sri Lanka; water and sanitation facilities in Malaysia are present, community self-help programs can suc- and Mexico; and telephone facilities in Indonesia, ceed over long periods. A community organization Sri Lanka, and Thailand. Each has brought private in Ethiopia devoted mainly to maintaining roads financing to support new investments. (the Gurage Roads Construction Organization) has Option C: Private ozwnership and private operation. worked well since 1962 because it sets its own prior- The private ownership and operation of infrastruc- ities and allocates its own financial and in-kind re- ture facilities is increasing-both through new entry sources. by private firms in infrastructure markets and Financing: essential for all options. Implementing through divestiture of public ownership of entire the foregoing institutional options and mobilizing systems. Private ownership is straightforward when funds to expand and improve services require care- services can be provided competitively, and, in fully designed financing strategies. Foreign and do- many infrastructure sectors, it is possible to identify mestic sources of finance will need to be tapped, but such activities and to allow private provision. For there are limits to the capacity of any economy to example, twenty-seven developing countries allow obtain funds from abroad, especially debt finance. cellular telephone service to be competitively pro- Balance of payments constraints, and the limited 9 tradability of infrastructure services, mean that for Activities that can be competitively provided most countries an ongoing infrastructure program should be separated and opened to private suppli- has to be sustained by a strategy for mobilizing do- ers and contractors. Where possible, entire sectors- mestic funds. telecommunications, railways, power generation- Private financing in one form or another at pres- can be privatized, but with regulatory oversight. ent accounts for about 7 percent of total infrastruc- Sectors that are unlikely to be privatized (such as ture financing in developing countries (the share roads) can be operated on commercial principles, may double by the year 2000), while bilateral and using contracting for construction and periodic multilateral foreign aid accounts for around another maintenance. Leasing or concessions can be used to 12 percent. Although an increasing share of the do- operate facilities that may be difficult to privatize mestic savings needed to finance infrastructure pro- for strategic reasons, such as ports or airports. More- vision can come from private sources, governments over, technical and managerial capacity at the will continue to be a major source of funds for infra- provincial and local level is likely to be sufficient to structure, as well as a conduit for resources from the realize the benefits of decentralization. Responsibil- donor community. As transitional measures to pro- ity for local services-such as urban transport, vide long-term financing where sufficient private water supply, sanitation, and local roads-can be support is not likely to be forthcoming, govern- turned over to local governments. ments are revitalizing existing lending institutions Low-income coun1tries with modest capacity. In these for infrastructure and creating specialized infra- countries, commercial principles of operation can structure funds. form the basis for reform in several sectors. Com- In the future, governments will often need to be mercial approaches can be supplemented with re- partners with private entrepreneurs. The task for forms in procurement and contracting practices that both the public and private sectors is to find ways to foster competition and develop the domestic con- route private savings directly to those private struction industry. Many activities (such as road riskbearers that are making long-term investments maintenance and the collection of solid waste) can in infrastructure projects-projects that have vary- be contracted out to the private sector. Contracting ing characteristics and for which no single financing can have a salutary effect on all infrastructure be- vehicle is appropriate. Official sources of finance, cause, as experience shows, public providers be- such as multilateral lending institutions, can facili- come more efficient when they are exposed to com- tate the process by supporting the policy and insti- petition from private contractors. tutional reforms needed to mobilize private financ- Concessions or leasing arrangements are proven ing and use it more efficiently. ways for a low-income country to draw on foreign expertise, as are the various BOT options that can be Implementing reform used to increase the capacity of systems. Conces- sions and leases have been widely used in water Just as the differences across infrastructure sectors supply, ports, and transport sectors. BOT schemes imply that no single option can be applied to all sec- have been extensively used in middle-income coun- tors, infrastructure provision must be tailored to tries, and their application is now spreading to low- country needs and circumstances, which vary income countries. These arrangements help develop widely. To see how, consider a middle-income coun- local expertise and foster the transfer of new tech- try with a thriving private sector and well-devel- nology, but they do not require the establishment of oped institutional capability, and a low-income independent regulatory bodies because regulatory country with a small private sector and relatively procedures are specified in the underlying contract. undeveloped institutional capacity. Community approaches, with technical and fi- Middle-incomie countries with good capacity. The nancial support, can be efficient and sustainable in four major options can all work well in these coun- supplying services using intermediate technologies tries. The broad reform instruments for such coun- in rural areas and in the low-income settlements tries are clear: apply commercial principles, increase that often develop outside existing urban service competition, and involve users. These actions lead areas. Competition is possible in many activities but to an increase in private involvement and finance, may be impeded by unnecessary regulations. Truck- and to a reduction (or decentralization) of activities ing and many types of urban passenger transport remaining with government. Some countries are fol- can be provided privately, under regulations that lowing this path for a wide range of sectors, and deal only with safety and service standards. many more for only a few sectors, especially tele- Some countries may benefit from arrangements communications, power, and roads. that increase the effectiveness of aid by coordinating 10 the efforts of donors to focus on common objectives. For example, the Sub-Saharan Africa Transport Pol- Figure 5 Annual gains from eliminating icy Program coordinates donor assistance for road mispricing and inefficiency are large maintenance and in several countries has supported relative to investment. the establishment of road boards that oversee execu- tion of road maintenance. More generally, external assistance should aim to build institutional capacity Billons of U.S. dollars in those countries where it poses a serious constraint. 250 Well-designed programs of training and technical Investment cooperation, as well as efforts to collect and dissem- 200 200 inate information on policy options and perfor- mance across countries, can supplement donors' F ical advice and financial assistance in creating an appro- l 1! burden - priate enabling environment for successful reform 123 and development of infrastructure. - Re,ource Potential payoffs from reform Because of the great variation in performance, the payoffs from increasing the efficiency of infrastruc- 0 ture provision will differ from country to country Subsidies Costs Annual and from sector to sector. But the rewards are poten- incurred from incurred from infrastructure tially large across the spectrum, making the commit- mispricing technical investment ment to reform imperative and worthwhile. inefficiencya Reform will produce three types of gains: reduc- tion in subsidies, technical gains to suppliers, and U Water U Development finance gains to users. It is possible to make rough estimates of the first two types of gains. The first source of D Railways I1 All other sources gains is the reduction in the fiscal burden of service Roads provision-costs not recovered from users. Al- though a conservative estimate can be made for * Power only three sectors (power, water, and railways), the total savings are nearly $123 billion annually- nearly 10 percent of total government revenues in a. Costs for the water sector are due to leakages; for railways-f.Liel inctriciency overstaffing, and locomotive developing countries, 60 percent of annual infra- un. i L3, lb [Iii for roads-added investment caused by structure investment, and approximately five times poor maintenance; for power-transmission, annual development finance for infrastructure (Fig- distribution, and generation losses. ure 5). Eliminating underpricing would not produce Source: Ingram and Fay, background paper; Appendix a net resource savings to the economy (as the costs would be covered by users), but the fiscal relief would be substantial. The payoffs from better infrastructure services go The second source of gains is the annual savings beyond reducing technical inefficiency and financial to service providers from improved technical effi- losses. Improvements in productivity and pricing ciency. The savings possible from raising operating would permit more effective delivery of service in efficiency from today's levels to best-practice levels response to demand. They would also enhance the are estimated at around $55 billion a year-pure growth and competitiveness of the economy. And savings equivalent to 1 percent of all developing they would allow vastly greater mobilization of re- countries' GDP, a quarter of annual infrastructure sources for needed new investments-by generat- investment, and twice annual development finance ing higher revenues and by creating a policy envi- for infrastructure. Looked at another way, if the an- ronment conducive to the inflow of new investment nual technical losses of $55 billion could be re- resources. directed for three years-at current costs of roughly This Report's agenda for reforming the incen- $150 per person for water systems-the 1 billion tives and institutional frameworks in infrastructure people without safe water could be served. poses major challenges-but promises major bene- 11 fits. The way ahead is one of continuing innovation emphasis needs to be on improving environmental and experimentation, and both industrial and de- conditions. Increasingly, infrastructure needs to veloping countries will learn from each other. In match new demands as developing countries be- some countries, the challenge is to keep pace with come more closely integrated into the global econ- rapid economic growth and urbanization. In others, omy. Infrastructure is no longer the gray backdrop it is to restore growth in ways that also provide of economic life-underground and out of mind. It greater opportunities to the poor. Everywhere, the is front and center in development. 12 Highlights of World Development Report 1994 Investment in infrastructure has a high potential payoff in terms of economic growth. But, as discussed in the box, making infrastructure available is not enough to guarantee high returns; services must be delivered effi- ciently and in response to demand. Good performance in infrastructure sectors comes from having good sectoral policies, not necessarily high per capita incomes (see the figure on the facing page). Box 1.1 Returns on infrastructure investment-too good to be true? Recent studies in the United States suggest that the im- frastructure on production costs. Studies (summarized pact of infrastructure investments on economic growth in Aschauer 1993) found that infrastructure significantly represents startlingly high rates of return (up to 60 per- reduces production costs in manufacturing in Germany, cent). Too good to be true? Possibly. The results presented Japan, Mexico, Sweden, the United Kingdom, and the in Box table 1.1 may overestimate the productivity of in- United States. One estimate suggests that three-quarters frastructure for two reasons. First, there may be a com- of U.S. federal investment in highways in the 1950s and mon factor that causes growth in both output and infra- 1960s can be justified on the basis of reductions in truck- structure that is not included in the study Second, it may ing costs alone. be that growth leads to infrastructure investment, and While there is still no consensus on the magnitude or not that investment produces growth. A number of stud- on the exact nature of the impact of infrastructure on ies have found that causation runs in both directions. Yet growth, many studies on the topic have concluded that more sophisticated estimates that address these issues ei- the role of infrastructure in growth is substantial, signifi- ther have concluded that the positive results were not cant, and frequently greater than that of investment in much affected by different econometric methods or have other forms of capital. Although the indications to date found no noticeable impact of infrastructure on growth. are suggestive, there is still a need to explain why the Neither finding-of an extremely high impact or of a findings vary so much from study to study. Until this negligible impact-is entirely credible, and research ef- problem is resolved, results are neither specific nor solid forts continue in an attempt to refine the methodology. enough to serve as the basis for designing policies for in- An alternative approach estimates the impact of in- frastructure investment. Box table 1.1 Results from studies of infrastructure productivity Iniplied rate Sample Elasticity" of returulb Autilor/year Inifrastruictutre measuire United States 0.39 60 Aschauer 1989 Nonmilitary public capital United States 0.34 60 Munnell 1990 Nonmilitary public capital 48 states, United States 0 0 Holtz-Eakin 1992 Public capital 5 metro areas, United States 0.08 - Duffy-Deno and Eberts 1991 Public capital Regions, Japan 0.20 96 Mera 1973 Industrial infrastructure Regions, France 0.08 12 Prud'homme 1993 Public capital Taiwan, China 0.24 77 Uchimura and Gao 1993 Transportation, water, and communication Korea 0.19 51 Uchimura and Gao 1993 Transportation, water, and communication Israel 0.31-0.44 54-70 Bregman and Marom 1993 Transportation, power, water, and sanitation Mexico 0.05 5-7 Shah 1988,1992 Power, communication, and transportation Multicountry, OECD 0.07 19 Canning and Fay 1993 Transportation Multicountry, developing 0.07 95 Canning and Fay 1993 Transportation Multicountry, OECD 0.01-0.16 - Baffes and Shah 1993 Infrastructure capital stocks and developing Multicountry, developing 0.16 63 Easterly and Rebelo 1993 Transportation and communication a. Percentage changes in output with respect to a I percent change in the level of infrastructure. b. Ratio of discounted value of increase in dependent variable to discounted value of investment in infrastructure. 14 Figure 1.5 Efficient and effective delivery of infrastructure services does not always accompany increased a va ii ability. Faults per 100 main lines Percentage of power delivered 0 - 100 0 Egypt SaudiArabia Turkey V_ TI..,,,.; I 50 * Syria 80 * e t* Chile * Argentina o 100 Tunisia Jordan Romania 60 nia;, 150 k_Chad 40 Mauritius 200 . 20 Zimbabwe 250 I I 0 0 50 100 150 0 20 40 60 80 100 Telephone main lines per thousand persons Percentage of households *th electricitN Percentage of paved roads Percentage of diesel in good condition locomoti% e a% ailabilitv 100 MauritiusO * 100 * Mexico Botswana a * t -. t r S d 80 KoCe d*voire Trinidad & Tobago 80 - *r, Korea Oman * 0 Poland 60 - .* . Tunisia 60 * .Z.a. Colombia Chile Portugal* * * r 40 40 /5. * 40t * Argentina *Panama Uruguay Colombia * o *-* -Ronmania S Sudam 20 - * Peru 0 CostaRica 20 N,.:.,: * * Sri Lanka 0 Jamaica 0 1 0 _ 0 500 1,000 1,500 2,000 2,500 0 200 400) 600 800 1,000 Kilometers of paved roads Rail passenger-kilometers + per million persons ton-kilometers per million US$ of GDP Ptrcentage of accounted-for water 100 N.-, i.. ~ 80 Indonesia 4 Gb Togo * Somalia Jamaica * * Middle East and North Africa 60 * * Ecuador Jord0 0 Latin America and theCaribbean Ethiopia Nepal 40 - i h-hr.'r..,. East .Xsia and Pacific 0 SLlb-Saharan Africa 20 0 South Asia * Europe and Central Asia 0 0 20 40 W:t 80 100 Percentage of population with access to safe water Source: WDI table 32. 15 Inefficient operation and poor pricing policies often lead to unreliable services and greater inequities. As illustrated on the facing page, public-private sector partnerships can improve infrastructure perfor- mance as well as contribute to higher flows-and better use-of financial resources for these sectors. Figure 2.2 Costs are seldom fully recovered Figure 1.6 There is very high unmet in infrastructure. demand for telephone connection. Ratio of revenues to costs Number ot countnes 2.0 25 20 - 151 _ _ _ _ _ _ _ _ Fnancial autonomo , I 5111 Less 1.0- 3.0- 6.0- 9.01- 12.0 or than 1 2.9 t.9 8.9 11.9 more - Waiting time for telephone main lines. 1992 (y'ea rsj Telecom Gas Pow%er Water Saiinpl 95 developing countrien qL ,r tL Telecom. ITI 1194; gas. World Bank data. po%, er. S. Plii lI 19Q4 Besant Jones 1993; %vater. Bhaha and Falkenmark 1993 Box 1.5 Households' responses to unreliability of water supply In 1991, micro-level research on household responses to In Jamshedpur, the connection charges for piped deficient water supply by public utilities was under- water vary between $1.66 and $16.66. The residents of taken in Faisalabad (Pakistan), Istanbul (Turkey), and the periurban areas, served by the local municipal au- Jamshedpur (India). These surveys revealed that nearly thorities, incur capital costs of $50 to $65 in installing all households in the three cities are dependent on multi- tubewells and $150 to $300 in digging wells to avoid de- ple sources of water, including house taps, wells, tube- pendence on the (unreliable) public water supply. De- wells, public taps, rivers, and street vendors. Not all al- spite the existence of a piped water system, at least 17 ternatives are available to all households. Because access percent of the population meets 90 percent of its water to a source increases with income, poorer households needs from wells and handpumps. Over and above the bear a disproportionate share of the burden of deficient monetary costs that consumers bear, households in infrastructure. The private expenditures incurred for Jamshedpur spend, on average, two hours a day fetch- water supply indicate consumers' willingness to pay for ing and storing water. The burden of these activities falls reliable water. in nearly all cases on women. In Istanbul, the poorest households surveyed spend a The pattern of private augmentation of the public larger share of their income (about 5 percent) to supple- water supply at substantial private costs to consumers is ment inadequate water supply than do wealthier ones observed also in Faisalabad, Pakistan. Less than 20 per- (which spend about 1 percent). These expenditures on cent of the households with piped water use this source informal sources of water, including self-provision from exclusively; 70 percent have motor pumps and 14 per- wells or storage facilities, are in addition to the user cent have handpumps. charges for publicly supplied water, which amount to I to 2 percent of annual income. 16 Figure 3.2 Leases and concessions in Figure 5.1 Large shares of official infrastructure sectors are common, even development finance for infrastructure in low-income countries. go to energy and transport. Official development finance Number of countrien with leases (billions of U.S. dollars) and conce-sions 30 40 35 25 30 15 10 Low- Mfiddlv- Total 0 X 1 income imL me 1984 1985 1986 1987 1988 1989 1990 1991 * WaterandsewEerage * Other infrastructure F] Power * Communications * TransFport LI \\ater and sanitation * Trans.port Source: Berg 14 Tn *Energ% Source: Appt.ndi\ table ..4 Box 3.4 Success of a lease contract-Guinea's water supply When the Republic of Guinea's water supply sector was To make sure the necessary tariff increases would be restructured in 1989, it was one of the least developed in affordable, the Guinean lease contract included an inno- West Africa. At that time a new autonomous water au- vative cost-sharing arrangement. Under the agreement thority, SONEG, took over ownership of the urban water negotiated by the government, the two sector entities, supply infrastructure and assumed responsibility for and the external financier (the World Bank), the con- sector planning and investment. SEEG, 49 percent gov- sumer tariff was to be adjusted gradually from the first ernment-owned and 51 percent owned by a foreign con- to the tenth year of the contract. During this period the sortium, was created to operate and maintain the sys- World Bank agreed to assume a declining share of the tem's facilities. foreign exchange expenditures of operation, and the cen- Under the ten-year lease contract signed with tral government covered a declining share of the debt SONEG, SEEG operates and maintains the system at its service. By the tenth year tariffs were expected to cover own commercial risk. Its remuneration is based on user the full cost of water. Tariff increases have to date ex- charges actually collected and fees for new connections. ceeded the planned schedule, rising from $0.12 per cubic SEEG also benefits from improvements it achieves in the meter in 1989 to about $0.75 in 1993. Despite higher tar- collection ratio, from reduced operating costs, and from iffs, the collection ratio for private customers has in- reductions in unaccounted-for water. Since SONEG has creased dramatically-from less than 20 percent to more ultimate responsibility for capital financing, it has strong than 75 percent in 1993-and technical efficiency and incentives to seek adequate tariffs and to make prudent service coverage have improved. investments based on realistic demand forecasts. 17 Natural monopoly only exists for certain activities within any infrastructure sector. Separating out ("un- bundling") these activities opens an array of alternatives for competition and private sector involvement to achieve greater efficiency and new investment. An existing enterprise can be converted to more market-based provision through one or more of the arrangements shown in the figure. Figure 3.1 Unbundling activities increases the options for competition and private sector involvement. Conpehtion Competitive Competition activities in the market with entry of Business new firms practices, r w venvironment, safety; and antitrust Right of acc;ev Monopoly rt ni' onpo ____ ___ ____ ___ ___ Pr ice, qLIJiIit%. Competition for and et% ice Monopoly the market obligation activities via concession i 1contr.1cr > . 9 or lease 111 ^ ,i r. Pric- qL1.1 t% 3nd 'er% ite NlonopolN Itate-oaned Integrated Integrated ObIigMiL0on m ystateow monopoly monopoly i.a 1tarute - monopoly.,,--- Initial status Industry structure Options for competiLton Object of regulation EM+ UnLtirndIings h-IFoi No unbundling 18 How to order World Development Report 1994 CUSTOMERS IN THE UNITED STATES: CUSTOMERS OUTSIDE THE UNITED STATES: Complete this coupon and return to Contact your local World Bank publications distributor The World Bank for information on prices in local currency and payment Box 7247-8619 terms. (For a complete list of distributors, see back.) If Philadelphia, PA 19170-8619 no distributor is listed for your country, use this order U.S.A. form and return it to the U.S. address. Orders that are To have your order shipped faster, charge sent to the U.S. address from countries with distributors by credit card by calling (202) 473-1155 will be returned to the customer. or send this completed order coupon by facsimile by dialing (202) 676-0581. Quantity Title Stock No. Price Total World Development Report 1994 (paperback, English) 60992 $19.95 World Development Report 1994 (paperback, French) 12534 $19.95 World Develoment Report 1994 (paperback, Spanish) 12535 $19.95 World Development Report 1994 (hardcover, English) 60991 $37.95 * SHIPPING AND HANDLING charges are $5.00 per order. Subtotal US$ If a purchase order is used, actual shipping will be charged. Shipping and Handling* US$ For air mail delivery outside the United States, add $8.00 for one item plus $6.00 for each additional item. Total US$ Other languages available in paperback at US$19.95 (Please note: Publication dates vary by language edition. Some editions will not be available until later in 1994.) Arabic (order #12539) Chinese (order #12537) German (order #12536) Japanese (order #12538) Portuguese (order #1254 1) Russian (order #12540) CHECK METHOD OF PAYMENT O Enclosed is my check payable to The World Bank. O Charge my O VISA O MasterCard Q American Express Credit card account number Expiration date Signature OJ Bill me. (Institutional customers only. Purchase order must be included.) PLEASE PRINT CLEARLY Name Address City State Postal Code Country Telephone 1044 Distributors of World Bank Publications ARGENTINA EGYPT, ARAB REPUBLIC OF Korean Stock Book Centre SOUTH AFRICA, BOTSWANA Carlos Hirsch, SRL Al Ahram P.O. Box 34 For single titles: Galeria Guemes Al Galaa Street Yeoeido Oxford University Press Florida 165,4th Floor-Ofc. 453/465 Cairo Seoul Southemn Africa 1333 Buenos Aires P.O. Box 1141 The Mfiddle East Observer MALAYSIA Cape Town 8000 AUSTRALIA PAPUA NEW 41. Sherif Street University of Malaya Cooperative GUINEA, FIJI, SOLOMON Cairo Bookshop, Limited For subscription orders: ISLANDS, VANUATU, AND P.O. Box 1127, jalan Pantai Baru hinternational Subscription Service WESTERN SAMOA FINLAND 59700 Kuala Lumpur P.O. Box 41095 D.A. hinformation Services Akateeminen KuijakauppaCrihl 648 Whitehorse Road P.O. Box 128 MEXICO Craohannsbrl 22 Mfitcham 3132 SF-00101 Helsinki 10 EINFOTEC Victoria Apartado Postal 22-860 SPAIN FRANCE 14060 Tlalpan. Mexico D.F. Mundi-Prensa Libros. S.A. AUSTRIA World Bank Publications Castello 37 Gerold and Co. 66, avenue d'Ifna NETHERLANDS 28001 Madrid Graben 31 75116 Paris De Lindeboom/InOr-Publikaties A-1011 Wien P.O. Box 202 Libreria Internacional AEDOS GERMANY 7480 AE Haaksbergen Consell de Cent, 391 BANGLADESH UNO-Verlag 08009 Barcelona Miaro Industries Development Poppelsdorfer Ailee 55 NEW ZEALAND Assistance Society (MIDAS) 53115 Bonn EBSCO NZ Ltd. SRI LANKA AND THE MALDIVES House 5, Road 16 Private Mail Bag 99914 Lake House Bookshop Dhaunmondi R/Area HONG KONG, MACAO New Market P.O. Box 244 Dhaka 1209 Asia 2000 Ltd. Auckland 100, Sir Chfittampalam A. 46-48 Wyndham Street Gardiner Mawatha BELGIUJM Winning Centre NIGERIA Colombo 2 jean Dr Lannoy 7th Floor University Press nimited Av. du Roi 202 Central Hong Kong Three Crowns Building Jericho SWEDEN 1060 Brussels Private Mail Bag 5095 For single titles: HUNGARY Ibadan Fritzes Fackboksforetaget CANADA Foundation for Market Economy Regeringsgatan 12, Box 16356 La Diffuseur Dombovari Ut 17-19 NORWAY S-103 27 Stockholm 151A Boul. de Mortagne H-1117 Budapest Narvesen Information Center Boucherville, Qu6bec Book Departmnent For subscription orders: J4B 5E6 INDIA P.O. Box 6125 Etterstad Wennergren-Williams AB Allied Publishers Private Ltd. N-0602 Oslo 6 P. 0. Box 1305 Renouf Publishing Co. 751 Mount Road S-171 25 Solna 1294 Algoma Road Madras - 600002 PAKISTAN Ottawa, Ontario Mirza Book Agency SWITZERLAND KlB 3W8 INDONESIA 65, Shahrah-e-Quaid-e-Azam For single titles: Pt. Indira Limited P.O. Box No. 729 Librairie Payot CHILE jalan Borobudur 20 Lahore 54000 Case postale 3212 Invertec IGT S.A. P.O. Box 181 CH 1002 Lausanne Av. Santa Maria 64010 Jakarta 10320 PERU Edifidio INTEC, Of. 201 Editorial Desarrollo SA For subscription orders: Santiago IRAN Apartado 3824 Librairie Payot Kowkab Publishers Lima 1 Service des Abonnements CHINA P.O. Box 19575-511 Case postale 3312 China Financial & Economic Tehran PHILIPPINES CH 1002 Lausanne Publishing House International Book Center 8, Da Fo Si Dong Jie IRELAND Suite 1703, Cityland 10 THAILAND Beijing Goverrnment Supplies Agency Condominium Tower I Central Department Store 4-5 Harcourt Road Ayala Avenue, H.V. dela 306 Silom Road COLOMBIA Dublin 2 Costa Extension Bangkok Intoenilace Ltda. Makati, Metro Manila Apartado Aereo 34270 ISRAEL TRINIDAD & TOBAGO, ANTIGUA Bogota D.E. Yozmot Literature Ltd. POLAND BARBUDA, BARBADOS, P.O. Box 56055 International Publishing Service DOMINICA. GRENADA. GUYANA. COTE D'IVOIRE Tel Aviv 61560 UJ. Piekna 31/37 JAMAICA, MONTSERRAT, ST. Centre d'Edition et de Diffusion 00-677 Warzawa KITTS & NEVIS, ST. LUCIA. Africaines (CEDA) ITALY ST. VINCENT & GRENADINES 04 B.P. 541 Licosa Commissionaria Sansonti SPA For subscription orders: Systematics Studies Unit Abidjan 04 Plateau Via Duca Di Calabria, 1/1 IFS journals #9 Watts Street Casella Postale 552 Ut. Okrezna 3 Curepe CYPRUS 50125 Firenze 02-916 Wargzawa Trinidad. West Indies Center of Applied Research Cyprus College JAPAN PORTUGAL UNITED KINGDOM 6, Diogenes Street, Engomi Eastern Book Service Livraria Portugal Miaroinfo Ltd. P.O. Box 2006 Hongo 3-Chome, Bunkyo-ku 113 Rua Do Csrmo 70-74 P.O. Box 3 Nicosia Tokyo 1200 Lisbon Alton, Hampshire GU34 2PG DENMARK KENYA SAUDI ARABIA, QATAREnld SamifundsLitteratur Africa Book Service (E.A.) Ltd. Jarir Book Store Rosenoerns A1ll6 11 Quaran House, Mtangano Street P.O. Box 3196 DK-1970 Frederiksberg C P.O. Box 45245 Riyadh 11471 Nairobi DOMINICAN REPUBLIC SINGAPORE, TAIWAN, Editors Taller, C. por A. KOREA. REPUBLIC OF MYANMAR,BRUNEI Restauraci6n e Isabel la Cat6lica 309 Pan Korea Book Corporation Gower Asia Pacific Pte Ltd. Apartado de Correos 2190 Z-1 P.O. Box 101, Kwangwhamun Golden Wheel Building Santo Domingo Seoul 41, Kallang Pudding, #04-03 Singapore 1334 El i. T -.+ i)lD E7,; C3n>>-t od ..A UVU3Wdo El f p p l -jC2f DU3 65 ;' t / aM M 1:-;I1 S 6\ 1 2/ZJ T H E W 0 R L D B A N K I his CxCLctive st-mmaryI rep-rodtuces tile orlviW alltld several otliel- ellleelts of IWor/0d )we,(/pl1lfn1t R(por/ 1994. Le on omllc in frt r IStLiCtLt-C-i-IICIcLidinig transport, electr ic powec, telecommullicatln(ls W.lter and(i sanitation, and waste disposaa---produces SerViCeS that are vital to deVe'Oplllmllt. 'I'lhe avail- aIblity of infrasiruicture has increased significantly in developinmg colilntnles oVerI' rFcelit decades. Yct pCer1or-manCeI-S 0tCI [)001is oftenICO; inadleutel mat'linltlen ce la(ls to premature detetroioratioln of ftlcil- ties, and ser-vices flrequently do nOt m1LItch uSCIS' needs anld willingness to pa. 'I'hec potential for illnfra-StrluCture services to conltribtLeC to poVerlty rdltuction andci to elViv lilro llnntal imilp'ovem llent iS often ne,lected. 'I'he tlime is ripe to broaden the focUs of policy beyond incretasilg the quan- tity of inIfrastrILcturIe_ ficilitics to im11proving the qLuality and efficiency of services. I'lhe sevcten tieln annual Worltld )velopmni Repori exam.IIines thei record of' botlh successes adi fa,lures In illnfrastructure provision anid concludes that obtaininig miiore cf'thctiv anld cfficifnt deliverv of s'eIvICeS r-CequilreCS ImpI)Or'tant Chanllges in intcetives. 'I'lce Report identifies thl-ree WayS of refiorming Insuttitionls to imllprove icenltives: * Apply principles of commercial m ianiagemeilniit-olpcratilg intiastru-cstuetlC 1m1ore likel a btisi- necss than a buIrcaLCracy, With cler- objectives, with manaLgrial a1nd fincllaICl aLtito nonily 1y,ad witll clS-toMelr atiSLaCtionl .1 keyl ImlealSuIr of )Cer forln IIce. * 1IltIodI.uCe C lIpetition----dirctlk wIhe- feasible, illdil-rctly where not. By gJiinur more optioms, comlpetitioni makes prooxvi-is imiore efticieilt and miiore accmulnableb to ulsers. * (,iVC LISCIS anitd other stakeholders a stroll voice and r-eal responsi bility ill planning, oper- ati n ( rot ilating, and fillanci lln' serVice's. (,Ovcrin men ts will need to act more often as f cilitator, coordinator, and guardian of thle pub- lic intercst (cspeciall)' to eInsure access for the poor)--and less often as direct opel-atolr pI'he pri- vate sector will need to) provide maagcment expcrtise and a sigiiifican tl greater Share off illaliec tha in the paSt. l-roll a brl-oad me nL.u of sp)citic illstitutioi0al afrangemen is tor ilnfriastrictLnrC pro- vision, thle Report assesses foil- options by sectora II tWrype ot coutry. I Report notes that improved performance of infrst ructure Can e niable COtLiis to achieve substantial eticiecly gains, a highiel lcvel of sel-vice fr(omil existilln infr'StlruCtrllet' i-csotir-ces, and Improved iccess t( essenitial ser'vices for tilch Poo'. I'lhis Report inelidtcies thle WoX-/ld D)evcIpilolp t Inidict;lors, Vwhich offer com pirllensiVe, (cuI'r'tre data Oii Social anid ecoiomilc developIlmlent in llmo than 2t)() t ries and terrimrieS 'I he In_diators are also availahle on diskettc. Ani appendix to the Report provides seletetd sttistics Onl iniflraStruCtIlrCe. COVER DESIGN BY BRIAN NOY ES THE MAGAZINE GROUP ISBN 0-8213-2542-6