57742 W O R K IN G PAP E R N O .43 Private Infrastructure in Developing Countries: Lessons from Recent Experience José A. Gómez-Ibáñez WORKING PAPER NO. 43 Private Infrastructure in Developing Countries: Lessons from Recent Experience José A. GómezIbáñez © 2008 The International Bank for Reconstruction and Development / The World Bank On behalf of the Commission on Growth and Development 1818 H Street NW Washington, DC 20433 Telephone: 2024731000 Internet: www.worldbank.org www.growthcommission.org Email: info@worldbank.org contactinfo@growthcommission.org All rights reserved 1 2 3 4 5 11 10 09 08 This working paper is a product of the Commission on Growth and Development, which is sponsored by the following organizations: Australian Agency for International Development (AusAID) Dutch Ministry of Foreign Affairs Swedish International Development Cooperation Agency (SIDA) U.K. Department of International Development (DFID) The William and Flora Hewlett Foundation The World Bank Group The findings, interpretations, and conclusions expressed herein do not necessarily reflect the views of the sponsoring organizations or the governments they represent. The sponsoring organizations do not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the sponsoring organizations concerning the legal status of any territory or the endorsement or acceptance of such boundaries. All queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 2025222422; email: pubrights@worldbank.org. Cover design: Naylor Design About the Series The Commission on Growth and Development led by Nobel Laureate Mike Spence was established in April 2006 as a response to two insights. First, poverty cannot be reduced in isolation from economic growth--an observation that has been overlooked in the thinking and strategies of many practitioners. Second, there is growing awareness that knowledge about economic growth is much less definitive than commonly thought. Consequently, the Commission's mandate is to "take stock of the state of theoretical and empirical knowledge on economic growth with a view to drawing implications for policy for the current and next generation of policy makers." To help explore the state of knowledge, the Commission invited leading academics and policy makers from developing and industrialized countries to explore and discuss economic issues it thought relevant for growth and development, including controversial ideas. Thematic papers assessed knowledge and highlighted ongoing debates in areas such as monetary and fiscal policies, climate change, and equity and growth. Additionally, 25 country case studies were commissioned to explore the dynamics of growth and change in the context of specific countries. Working papers in this series were presented and reviewed at Commission workshops, which were held in 2007­08 in Washington, D.C., New York City, and New Haven, Connecticut. Each paper benefited from comments by workshop participants, including academics, policy makers, development practitioners, representatives of bilateral and multilateral institutions, and Commission members. The working papers, and all thematic papers and case studies written as contributions to the work of the Commission, were made possible by support from the Australian Agency for International Development (AusAID), the Dutch Ministry of Foreign Affairs, the Swedish International Development Cooperation Agency (SIDA), the U.K. Department of International Development (DFID), the William and Flora Hewlett Foundation, and the World Bank Group. The working paper series was produced under the general guidance of Mike Spence and Danny Leipziger, Chair and Vice Chair of the Commission, and the Commission's Secretariat, which is based in the Poverty Reduction and Economic Management Network of the World Bank. Papers in this series represent the independent view of the authors. Private Infrastructure in Developing Countries: Lessons from Recent Experience iii Acknowledgments Parts of this paper originally appeared in José A. GómezIbáñez, Dominique Lorrain, and Meg Osius, "The Future of Public Infrastructure," working paper, Taubman Center for State and Local Government, April 2004; and in José A. GómezIbáñez, "Regulation and the Future of Private Utilities," presented at the Conference on Regulation, Deregulation, and Reregulation in Globalizing Asia, National University of Singapore, Public Policy Program, Singapore, March 22­ 24, 2004. iv José A. Gómez-Ibáñez Abstract Beginning in the late 1980s many developing countries turned to the private sector to provide basic infrastructure and utility services, such as highways, railroads, water, wastewater, electricity, gas, and telecommunications. Recent studies suggest that private involvement often benefited customers and reduced government fiscal problems without harming employees or enriching private providers excessively. There were enough highprofile failures, however, to discredit this reform in many quarters. Private involvement is likely to be more successful if it generates real efficiency gains rather than simply transferring costs among parties, if the systems of regulating the private companies are politically sensitive as well as technically competent, if the costs and constraints of private capital are not excessive, and if we are willing to adopt more modest and gradual schemes in difficult circumstances. Private Infrastructure in Developing Countries: Lessons from Recent Experience v Contents About the Series ............................................................................................................. iii Acknowledgments ..........................................................................................................iv Abstract .............................................................................................................................v Performance and Perception ........................................................................................10 The Initial Terms of Privatization................................................................................20 Systems of Regulation ...................................................................................................24 The High Cost of Private Capital.................................................................................29 Adapting to Difficult Circumstances ..........................................................................32 References ....................................................................................................................... 40 Private Infrastructure in Developing Countries: Lessons from Recent Experience vii Private Infrastructure in Developing Countries: Lessons from Recent Experience José A. GómezIbáñez 1 Beginning in the late 1980s many developing countries turned to the private sector to provide basic infrastructure or utility services, such as highways, railways, water, sanitation, electricity, gas, and telecommunications. One hope was that private providers would be much more efficient than the stateowned enterprises that they replaced so that they could offer better service at reasonable tariffs and without the need for large government subsidies. A related goal was that the providers could tap private capital markets to finance badly needed modernizations and expansions of capacity. The desire for private participation was complicated, however, because infrastructure was often thought to have elements of natural monopoly so that governments could not be sure that the private operators would pass on some or most of the savings to customers in lower prices or better service. Infrastructure was also believed to generate wider public benefits, such as improved public health or national political integration, so that the government was often interested in a more extensive network than users alone would normally support. As a result, privatization was usually accompanied by some form of government regulation of tariffs, service quality, and coverage, intended to protect consumers from potential monopoly abuse and to advance common social interests. After nearly two decades, however, many people and governments in developing countries are much more skeptical of private participation in infrastructure. Critics often argue that consumers and workers suffer from higher prices and massive layoffs, while any resulting savings are either squandered by politicians or captured by private infrastructure companies, many of them foreign owned. The popular perception of failure is troubling because it conflicts with an emerging empirical literature that suggests that private provision of infrastructure often delivers better service at reasonable prices. But while private participation is often beneficial, there have been enough highly publicized failures to give credence to the critics. Moreover, in some cases the shortcomings in performance were magnified by naïveté about the extent to which reform is a political as well as a technical task. In particular, we underestimated the José A. GómezIbáñez is the Derek C. Bok Professor of Public Policy and Urban Planning at 1 Harvard's Graduate School of Design and Kennedy School of Government. Private Infrastructure in Developing Countries: Lessons from Recent Experience 9 importance of insuring that the benefits in the initial privatization are equitably distributed, that the systems of regulating the private companies are politically sensitive as well as technically competent, that the costs of private capital are not higher than they need be, and that more modest forms of private provision be considered in difficult environments. Performance and Perception Trends in Activity and Popular Support Private participation in infrastructure takes many forms, the principal three being divestitures, concession contracts, and management contracts. In a divestiture the government sells an existing stateowned enterprise (such as a telephone or railroad company) through a stock offering to the general public or an auction to strategic investors. In a concession the government offers a contract to invest in and operate an enterprise or facility for a limited duration, often for 20 or 30 years, with the assets reverting to the state at the end. Some concessions are to build and operate new or "greenfield" infrastructure facilities (such as a new power plant or gas pipeline) while others are to improve and operate existing or "brownfield" facilities or enterprises (such as a concession to expand and manage an existing port, highway, or water company). Finally, in a management contract or lease the private company is responsible only for management and not investment; as a result, management contracts are shorter than concessions: usually 3 to 5 years, with options for renewal. Privatization attracted over one trillion dollars of new private investment in developingcountry infrastructure between 1990 and 2006 according to World Bank estimates.2 Investment grew from negligible levels in the mid1980s to a peak of approximately US$114 billion in 1997, the year the Asian financial crisis began, fell for five of the next six years, and only began to recover in 2004. Preliminary figures suggest that by 2006 investments levels had reached their 1997 peak measured in current dollars but were still roughly 25 percent below the peak measured in constant dollars (see Figure 1). By sector, the bulk of the funds flowed into telecommunications, which also led the recovery of 2004, with electricity second, transport third, and much smaller investments in gas, water, and wastewater (Figure 1 again). 2 These figures are from the Public Private Infrastructure (PPI) database created by the World Bank. This database defines privatization as occurring if a private firm takes operational control of an existing enterprise or facility or a new concession. The investment is recorded in the year of the financial closing of the transaction and includes the sales price or concession fee paid to the government plus any additional investment commitments during the life of the concession. No effort is made to track actual investments after the closing, which may be higher or lower than the minimum commitments made. 10 José A. Gómez-Ibáñez Figure 1: Private Investment in Infrastructure in Developing Countries by Sector, 1990­2006 in Constant 2006 Dollars (2006 figures preliminary) 140 120 Total 100 2006 US$ billion 80 Telecom Energy 60 40 Transport 20 Water 0 1990 1995 2000 01 02 03 04 05 06* Source: Izaguirre (2007). * Preliminary. Figure 2: Private Investment in Infrastructure in Developing Countries by Region, 1990­2006 in Constant 2006 Dollars (2006 figures preliminary) 90 80 70 LAC 2006 US$ billion 60 50 40 EAP ECA 30 20 SSA 10 SA MENA 0 1990 1995 2000 2005 2006 Source: Izaguirre (2007). * Preliminary. By region, Latin America and East Asia led both the buildup and the falloff of the 1990s, Eastern Europe and South Asia led the recovery that began in 2004, while neither subSaharan Africa nor the Middle East and North Africa have seen much private investment so far (Figure 2) By type of privatization, divestitures generated the most funds while concessions for greenfield projects were a close second, concessions for existing facilities were a distant third, and management contracts usually did not register on this metric (Figure 3). Private Infrastructure in Developing Countries: Lessons from Recent Experience 11 Figure 3: Private Investment in Infrastructure in Developing Countries by Type of Project, 1990­2006 in Constant 2006 Dollars (2006 figures preliminary) 160 140 Total 2006 US$ billion 120 100 Greenfield 80 projects 60 40 Divestitures 20 Concessions 0 1990 1995 2000 05 06* Source: Izaguirre (2007). * Preliminary. While many privatizations went smoothly, there was a steady stream of wellpublicized controversies that fueled unease with the reforms (Table 1). The earliest highprofile controversies occurred in Asia when the Thai government seized an elevated private expressway in Bangkok in 1993 and the State of Maharashtra in India cancelled Enron's concession to build a power plant at Dabhol in 1994. Following the Asian financial crisis of 1997­1998, Indonesia, Pakistan, and the Philippines forced many independent private power producers to renegotiate the takeorpay contracts they had with government power companies, a process that exposed charges of corruption in the award of the original contracts. In Latin America, the first dramatic failure was the bankruptcy of roughly two dozen private toll roads in Mexico after the unexpected devaluation of the peso in late 1994. This failure was followed by the seizure of a water concession in Tucumán, Argentina (1996); violent rioting that stopped planned water and electricity privatizations in Cochabamba, Bolivia (2000) and Arequipa, Peru (2002); Argentina's unilateral reduction in utility tariffs in the wake of the government's default and devaluation in 2002; the cancellation of the La Paz water concession in 2005; Bolivia's nationalization of its natural gas industry in 2006; and Venezuela's nationalization of its electricity and telephone companies in 2007. Eastern Europe, which was much less active in privatizing infrastructure, had only one highprofile infrastructure failure when the M1/M15 toll road concession in Hungary went bankrupt for lack of traffic in 1998 and was nationalized in 1999.3 3 There were a number of serious disputes between governments and private operators involved in a water concession in Sofia and ports projects in Moldova and Latvia, but these did not receive the attention of the M1/M15 motorway (EBRD, 2007). 12 José A. Gómez-Ibáñez Table 1: High-Profile Controversies Involving Private Infrastructure Asia Bangkok (1993): takeover of Second Stage Expressway Maharashtra (1994): cancellation of Enron's Dabhol power plant Indonesia, Pakistan, and the Philippines (1998): renegotiation of independent power producer (IPP) contracts after Asian financial crisis Manila (2001): withdrawal of water concessionaire Manila (2004): withdrawal of airport concessionaire Latin America Mexico (1995): toll road bankruptcies Tucuman, Argentina (1996): cancellation of water concession Cochabamba, Bolivia (2000): cancellation of water concession Arequipa, Peru (2002): cancellation of electricity concession Argentina (2002): economic crisis and devaluation leads many concessionaires to initiate proceedings before the International Center for the Settlement of Investment Disputes (ICSID) and others to withdraw La Paz (2005): cancellation of water concession Bolivia (2006): nationalization of gas industry Venezuela (2007): nationalization of electricity and telephone companies Eastern Europe Hungary (1998): bankruptcy of M1/M15 toll road concession Africa Senegal (2000): cancellation of electricity concession Tanzania (2005): cancellation of water concession Mali (2005): withdrawal of electricity and water concessionaire Industrialized Countries California (2000): electricity crisis Britain (2001): bankruptcy of Railtrack, the privatized railroad infrastructure company Britain (2007): bankruptcy of Metronet, one of two private firms with 30-year contracts to renew and maintain London underground's infrastructure and rolling stock In Africa, management or concession contracts have been cancelled in a number of countries, but the bestknown disputes involve Senegal (2000), Tanzania (2005), and Mali (2005). The developed countries were not immune to bungled infrastructure privatizations either, the best known including California's electricity crisis in the summer 2000; the bankruptcy in 2001 of Railtrack, the private company responsible for all of Britain's railway infrastructure; and the bankruptcy in 2007 of Metronet, one of two companies with 30year contracts to renew and maintain the London Underground's infrastructure and subway trains. It is important to recognize that these severe disputes represent a very small percentage of the infrastructure privatizations. According to the World Bank database only 48 projects became so contentious that they were cancelled or renationalized between 1990 and 2001. The record was worse in some sectors than others as cancellations accounted for only a few percentage points of the private investment in energy and telecommunications but 9.3 percent of the Private Infrastructure in Developing Countries: Lessons from Recent Experience 13 private investment in transportation and 11.3 percent of the private investment in water and sanitation. Overall, however, only 1.9 percent of the projects and 3.2 percent of the investment value were terminated (Harris, 2003; Harris et al. 2003). Surges and slowdowns in new private investment are to be expected, moreover, since different regions and sectors vary in their stages of infrastructure and economic development. Much of the decline at the end of the 1990s was due to the reduction in new investment in Latin America, for example, and particularly in telecommunications and electricity. However, that falloff occurred in part because most of the large Latin American countries had already privatized their telephone and electricity industries. Many fewer Latin American telephone or electricity companies remained in public hands as candidates for divestiture and private investment.4 It is only natural that the recent rebound is being lead by South Asia and Eastern Europe, both regions that are experiencing rapid economic growth but have seen relatively little infrastructure privatization to date. Finally, all of the pressures that contributed to the spread of private infrastructure in the 1990s still remain. Infrastructure is still thought to be a key bottleneck to human and economic development, and the governments of the developing countries still lack the financial resources to meet perceived infrastructure needs. In Latin America, for example, the rate of investment in infrastructure actually dropped during the 1990s as governments cut back public investments sharply while private investment surged less than they had hoped. Governments now have more realistic expectations of the potential of the private sector, but they also can not afford to ignore it. Thus while the actions of Presidents Chavez of Venezuela and Morales of Bolivia receive the most attention in the press, countries like Brazil, Mexico, and Chile keep offering enough new projects that private investment in the region has been growing slowly in real terms since 2003 (Figure 2). Nevertheless, it is clear that these controversies have helped erode support for private involvement in infrastructure, and made implementing or sustaining the reforms much more challenging. The shift in popular opinion is most obvious in Latin America, the region with the most experience with private infrastructure. Latinobarómetero (2005, p.76), which surveys some 19,000 Latin Americans in 18 countries every year, reports that the percentage agreeing strongly or somewhat with the statement that "the privatization of state enterprises has been beneficial for the country" fell from 46 percent in 1998, when the question was first asked, to 21 percent in 2002, at the height of the Argentine financial crisis, and had recovered to only 31 percent by 2005. Moreover, analysis of the responses shows that support is lowest among the poor and in countries with the most extensive reforms (Carrera et al., 2005). 4 Between 1990 and 2003 the share of electricity connections in Latin America provided by private companies increased from 3 to 60 percent while the share of telephone connections increased from 3 to 86 percent (Andres, Foster and Guasch, 2005). 14 José A. Gómez-Ibáñez Similar disenchantment is found in the postCommunist countries of Europe and Central Asia which have also seen many privatizations, although more in services and manufacturing than infrastructure. A 2006 survey of 28,000 individuals in 28 postcommunist countries found that only 19 percent of those polled wanted to leave companies that had been privatized in the hands of their current owners with no change while 81 percent thought that the terms of the privatization should be revisited (Denisova et al., 2007). Interestingly, the dissatisfaction stems less from faith in state ownership than from concerns about the fairness of the privatization process. Only 36 percent of those favoring change wanted to renationalize the companies and leave them in state hands; the remaining 64 percent wanted either to renationalize the companies but then privatize them again using a more transparent process or to leave the companies in the hands of the current owners but make the owners pay the fair value of the assets they had received. Typical Winners and Losers Several observers have noted that the decline in popular support is at odds with recent empirical research that suggests that private provision of infrastructure has been largely, although not entirely, beneficial (Birdsall and Nellis, 2005; Carrera et al., 2005; Fay and Morrison, 2007; Nellis, 2006; and Shirley, 2005). Many of the early studies of private infrastructure examined small numbers of cases in depth, an approach that is useful for identifying the factors that seem to contribute to success or failure but less helpful for determining whether the typical project was successful or not. Some of the more recent studies develop data sets for large numbers of private companies and stateowned enterprises (SOEs) in an effort to estimate the typical effect of private participation on performance. These studies often rely heavily on data from Latin American electricity distribution and water companies, which raises questions as to how much their results translate to other regions and sectors. The studies don't always compare privatized companies with remaining SOEs, so that one can't be sure that all types of enterprises were not improving. The data also often end three to five years after privatization, which leaves doubts about the longterm effects. Despite these flaws, the studies generate reasonably consistent and intriguing results. Most studies agree that the efficiency of the firm tends to increase significantly during or after privatization. The focus is mainly on labor productivity since data on capital stocks are often unavailable, unreliable, or inconsistent across firms. In an analysis of panel data from 1,230 electricity and water enterprises in 71 developing countries around the world,5 Gassner et al. (2007) found that productivity of privatized companies increased over time and 5 The sample includes 302 enterprises with some form of private management and a matched set of 928 stateowned enterprises each with at least three years of data. Approximately 69 percent of the enterprises with private management come from Latin America and 22 percent from East Asia. Private Infrastructure in Developing Countries: Lessons from Recent Experience 15 relative to that of SOEs, although by different means in water than electricity. In water, the number of subscribers connected increased by 62 percent and raw water production increased by 3 percent but without a statistically significant change in employment. In electricity, by contrast, there was no statistically significant change in connections or output but employment was cut on average by 30 percent and collection rates increased by 48 percent. The difference between the two sectors probably reflects a greater tendency for governments to require concessionaires to increase connections in water than in electricity both because water connection rates are typically much lower and because water connections generate important public health benefits. Many other studies find similar gains in efficiency.6 In a careful analysis of panel data from 116 Latin American private and stateowned electricity companies, for example, Andres et al. (2006) found that the trends in output and connections did not change significantly with privatization but the trends in the quality of service (measured by the duration and frequency of interruptions) went up, employment went down, and labor productivity went up. The largest gains were in the three years surrounding the transition from public to private but trends improved after the transition period as well. Estache and Rossi (2005) examined 127 electricity companies in Latin America and found that private firms subject to certain forms of regulation used 20 to 60 percent less labor per unit of output than their stateowned counterparts. And Estache et al. (2002) estimated that productivity increased in Brazilian and Argentine freight railways after privatization, not because outputs increased but rather because inputs, particularly labor, decreased.7 Consumers usually captured enough of the efficiency gains to benefit, as a group, from privatization. Prices often increased with privatization: Gassner et al. (2007) found a statistically significant increase in prices among their water companies but not their electricity companies, for example, while Andres et al. (2006) found a statistically significant increase in prices among their electricity companies. But the price increases were often accompanied by improvements in access, reliability, and quality that provided offsetting benefits. In two of the most frequently cited studies, McKenzie and Mookherjee (2003, 2005) calculate the effects on consumer welfare of electricity, water, and telephone privatization in Argentina, Bolivia, Mexico, and Nicaragua. Prices for electricity and water increased in many cities while the average price for telephone service declined almost everywhere and connections to all types of utilities increased almost everywhere. The connection gains were greatest for poor households in the 6 There are a few studies that find private participation makes little difference in efficiency of utilities in developing countries, but they often have smaller data sets and are from earlier periods. Estache and Rossi (2002) find little difference in public and private water companies in Asia but they rely on a crossection of 50 companies in a single year. 7 A number of other studies show similar results for other types of infrastructure such as Ros (1999) and Ros and Banerjee (2000) for telephones and Estache et al. (2002a) for ports. 16 José A. Gómez-Ibáñez middleincome countries (Argentina and Mexico) where overall access levels were already relatively high, but greatest for middleclass households in the low income countries (Bolivia and Nicaragua) where overall access was initially relatively low. McKenzie and Mookherjee calculated that the total consumer surpluses increased for most countries and sectors, the principal exceptions being two cases where the price increases were exceptionally large: electricity throughout Nicaragua and water in Cochabamba (but not in other Bolivian cities). The gains were on the order of 0.5 to 2 percent of total household expenditure and were typically larger in percentage terms for lowincome or middleincome households than for highincome households. The few other studies that attempt to calculate the combined effect of price, access, and quality changes on consumer welfare often arrive at similar results. For example, Clarke et al. (2000) examined in detail the first eight years of the private management contract for urban water supply in Guinea. The prices charged before privatization were well below costs so they had to be increased approximately seven fold over the first five years. However, connections increased by 50 percent, the water was now treated so that it met World Health Organization (WHO) drinking water standards, and the supply was improved enough to permit 24hour service. Considering only the increases in prices and connections (and not the improvements in purity or reliability), Clarke and his colleagues calculated that consumers gained on net an amount equivalent to 75 percent of their expenditures on water in the year before privatization, roughly similar to the gains McKenzie and Mookherjee had estimated in Latin America. Other recent studies that also calculated positive consumer gains include Boston Institute for Developing Economies (2006) and Plane (1999). The general public seems to have benefited as well through the fiscal impacts of the reforms, although the extent to which this is true is more controversial and unclear (Birdsall and Nellis, 2005, pp. 23­24). The privatizations often generated substantial fiscal resources in the form of both the purchase prices or fees the government received for the enterprise or concession and the reduction in ongoing government budgetary support. Some observers believe these funds were generally spent wisely to reduce debt or taxes or to increase social spending (Mackenzie and Mookherjee, 2005, pp. 72­73). To the extent that these funds allowed governments to cut taxes or expand social services, the net effect was probably progressive if only because the poor were less likely to be connected to the utility network before privatization and thus benefited little from the old SOE subsidies (Estache et al., 2002, p. 13). However, the fiscal windfalls were not used wisely in every country. In Brazil, for example, Macedo (2000) argues that the fiscal gains from the privatizations of the early and mid 1990s were squandered on a futile defense of the Brazilian currency. And in other countries observers disagree about how the fiscal dividend was spent. In Argentina, for example, some argue that the bulk of the proceeds went to retire debt and increase social spending (Ennis and Pinto, 2005, pp. 212­213), Private Infrastructure in Developing Countries: Lessons from Recent Experience 17 while others believe that they were wasted defending the peso (Mussa, 2002 as cited by Birdsall and Nellis, 2005, p. 24). The effect on employees is also more controversial, but some recent studies suggest that employees do not suffer nearly as much as critics claim. In their in depth study of four Latin American countries, for example, McKenzie and Mookherjee (2005) found that privatization was often accompanied with workforce reductions of 30 to 75 percent. But infrastructure and utility layoffs accounted for at most 2 percent (in Argentina) and as little as 0.13 percent (in Bolivia) of the total workforce in the country. Moreover, employment in the infrastructure and utility sector often rebounded within a few years as the demand for services expanded. In Mexico, where the employment reductions were substantial, researchers were able to follow the experience of laid off SOE workers for a year and discovered that 40 to 50 percent of them found employment in the same sector at no reduction in benefits. And based on the increase in sector employment, McKenzie and Mookherjee estimate that 80 to 90 percent of the laid off Argentine workers were reemployed within four years. Finally, the private enterprises providing infrastructure services do not appear to be earning unreasonable returns, again contrary to popular perception. Sirtaine et al. (2005) found that the average return on investment fell far short of the costs of capital for 34 private companies in 9 Latin American countries. The returns were estimated from audited financial statements through 2001, and thus do not reflect the adverse effects of the Argentine financial crisis. The average historic return was negative 24 percent, largely because most of the projects were still in relatively early years when investments tend to be high and earnings low. Returns climbed to 7 percent, a little more than half of the average estimated cost of capital of 13 percent, if earnings over the remaining life of the concessions were projected to increase as fast as the historic growth of the GDP in the host country. And returns increased to 9 percent if the management fees paid to the parent companies were considered hidden dividends. Only if the firm was credited with projected earnings in perpetuity, beyond the life of the concession, did the projected returns climb to 14 percent, just above the cost of capital. These averages hid substantial variation, however, with telecommunications companies generally much more profitable than water and wastewater companies and electricity enterprises in between. Estache and Pinglo (2005) found similar results for 120 companies in 31 developing countries, although they did not adjust for management fees or include telecommunications companies. Companies in the poorest countries, in the Latin American and African regions and in railroad and water industries had the lowest returns. Perhaps the most convincing evidence that returns typically are not excessive is that many of the prominent international energy and water companies have decided to cut back their investments in developing countries. 18 José A. Gómez-Ibáñez Explaining the Discrepancy One explanation offered for the discrepancy between performance and perception is that the benefits of reform are diffused broadly and are almost invisible while the costs are highly concentrated and obvious (Harris, 2003; Birdsall and Nellis, 2005; Shirley, 2005; and Nellis, 2006). The fiscal benefits in reduced taxes, increased social spending, or macroeconomic stability may be large in the aggregate, for example, but probably amount to only a few dollars per household per month and are not linked in the public mind to infrastructure privatization. Similarly, McKenzie and Mookherjee's estimates suggest that the benefits to the average customer are on the order of 0.5 to 2 percent of their income and often depend on a complex calculus trading off higher prices against improved access or service quality. By contrast, the costs to any laidoff employees are large, even if many are rehired within one or two years, and clearly attributable to the reforms. And if customers see their prices increase without compensating improvements in access or quality, the effect is also obvious and the blame clear. Such a pattern of benefits and costs makes it easier to mobilize opponents than supporters. An equally plausible explanation is that the exceptions are often as politically salient as the average, and that there have been enough bad exceptions to lend credence to the critics of privatization (Birdsall and Nellis, 2005; Fay and Morrison, 2007; Shirley, 2005; and Nellis, 2006). While private participation often improves performance, many privatizations have not gone well. Even if the average customer and employee are better off or not greatly harmed in a particular privatization, there are sometimes groups of customers or employees who suffer significantly. Negative outliers seem to be more influential than positive outliers, moreover, since the press and the public pay more attention to conflict and failure than to harmony and success. The old saying is that bad news makes good news copy. Privatization makes an especially tempting target, moreover, since any complaints are likely to fall on receptive ears. In many countries private provision of infrastructure is seen, at least in part, as a policy imposed on the government by international financial institutions with the encouragement of multinational utility companies. And even where the policy is homegrown, the public in many developing countries is often skeptical about the fairness of market reforms and their government's ability to control large private companies, and therefore not surprised to learn that there are problems with private provision of infrastructure. All of these explanations suggest that reformers should pay as much attention to the political as the technical aspects of reform. Insuring that the benefits of reform exceed the costs is often technically challenging. But it is also important that the distribution of benefits and costs be, and be perceived as, reasonably equitable. And it is equally important that a fair distribution be maintained over the life of the project, not just at its inception. These tasks are Private Infrastructure in Developing Countries: Lessons from Recent Experience 19 made more difficult by the fact that the reformers may not have the incentives or skills to be concerned enough about the political dimensions of the reforms. As one longtime observer of these efforts explains: The technicians and advisors who design reforms are focused on and skilled in economic and technical rationality, not political sustainability. Foreign advisors and donors are unlikely to understand the demands and compromises required by local circumstances. Politicians may have too short a time horizon to care about longrun acceptance or may believe with some justification that a reform with widespread benefits will eventually be popular regardless of its flaws. (Shirley, 2005, p. 203) If private participation in infrastructure is to be both economically sensible and politically acceptable, four topics deserve greater attention: the initial terms of privatization, the scheme for regulating tariffs and service quality, the high nominal costs of private capital, and the constraints of difficult circumstances. The Initial Terms of Privatization Insuring that the initial terms of privatization are perceived as fair can be difficult for several reasons. One obstacle that often receives a great deal of attention is the tendency for SOEs to be overstaffed, which heightens the fear that labor will suffer from layoffs. The severity of this problem varies considerably, however, depending upon the sector and circumstances. Excess labor is less of a problem for infrastructure than for most manufacturing, agriculture, or services, simply because infrastructure is relatively capital extensive, so that achieving high labor efficiencies is not quite as critical. Within infrastructure, moreover, greenfield projects raise fewer problems than brownfield projects because greenfield projects do not involve an incumbent workforce and arguably increase total employment than if the project would not have been undertaken in the absence of private finance. Infrastructure industries where demand is increasing rapidly, such as telecommunications, are also less problematic since laidoff workers are likely to be quickly reabsorbed. The labor problem is most serious in industries like railroads where the network is typically too large rather than too small so that the initial layoffs are deep and industry employment is unlikely to recover. Labor problems are also serious when unemployment in the economy as a whole is high so that finding a job in another sector is difficult. A more pervasive and consistent problem is the pressure on government to capture part of the proceeds for budgetary purposes. In setting the terms for the sale of an infrastructure concession or company, governments face a tradeoff between the tariffs they will allow the private firm to charge its customers and the price the government will receive for the concession. The higher the permissible tariffs, the greater the sale proceeds to the government. The impetus for privatization is often a fiscal crisis which puts pressure on the government to 20 José A. Gómez-Ibáñez raise tariffs just before or during privatization to achieve a high sales price. And even if there is no fiscal crisis, a high price helps justify the privatization in the short term by making the government less vulnerable to charges that it sold valuable public assets for a pittance. A high sales price creates a longterm political liability, however, since consumers must pay higher tariffs for many years as a result. And the benefits the public receives from the sales proceeds are soon lost in the black hole of government budgets while the tariffs persist as an obvious irritant. Another pervasive obstacle to politically acceptable or equitable terms is the tendency of infrastructure SOEs to charge unrealistically low tariffs. In the early 1990s, on the eve of many of the privatizations, water tariffs in developing countries typically covered only one third of the costs of the service while electricity and gas tariffs covered only 60 or 80 percent of these costs. With such low tariffs the SOEs usually accumulated large investment backlogs. It was common for water and electricity to be rationed to several hours per day, for example, and for many households to be without water or electricity service because the government infrastructure company lacked the resources to expand its network of transmission pipes or lines. Under such circumstances, even a substantially more efficient private company would find it difficult or impossible to survive collecting only the old tariffs. And if the company is charged with providing more adequate and reliable supplies, or with extending coverage to unserved areas, then substantial tariff increases are inevitable. The reality is that the very low tariffs are unsustainable, but privatization becomes the scapegoat for the increases. A related obstacle is that customers are often heterogeneous, with different initial conditions and interests. Conflicts often arise because the tariffs that SOEs charge incorporate cross subsidies from some types of customers or services to others. Usually these cross subsidies are from less to more politically influential groups. For example, the charge per kilowatt hour of electricity or per cubic meter of water is typically higher for industrial users than for households or farmers. Similarly, users of longdistance telephone services were often, at least until recently, charged prices in excess of costs to keep the monthly tariff for basic residential service low. In other cases the rationale for the cross subsidies has eroded over time. Charges to connect new users to a utility network have been traditionally set below cost on the grounds that existing users benefit because the new users allow the company to exploit economies of scale in common facilities, such as electricity generating or water treatment plants. In the case of water and wastewater there was also the argument that the general public benefited because connections reduced the spread of disease. Over the years, network expansion and technological changes have weakened the rationale for crosssubsidizing new connections, especially outside of water and wastewater, but the practice persists. Private Infrastructure in Developing Countries: Lessons from Recent Experience 21 Privatization often exposes these differing interests by increasing the pressure on companies to "rebalance" tariffs so that they are more reflective of costs. Rebalancing is often required if the industry is reorganized to allow private firms to compete in providing selected activities. The introduction of competition in wholesale electricity markets or in longdistance telephony makes it harder to sustain traditional cross subsidies from industrial to residential electricity users or from longdistance to local callers. Rebalancing is also encouraged by the need to motivate private providers to expand and maintain services. A firm is more likely to expand its network, for example, if it is allowed to charge a connection fee that covers most or all of the connection costs. Altering these traditional patterns of cross subsidies creates winners and losers among consumers. For example, the traditional practice of undercharging for connections is less controversial for an SOE since the firm usually does not have the financial resources to expand its network anyway. If the SOE is replaced by a private firm that is allowed to raise connection charges to costrecovering levels, however, then existing customers with connections will be protected but many potential customers may feel aggrieved because they can not afford to join the network. Alternatively, if the private firm is required, as a condition of its concession, to expand service coverage without raising connection charges, as is often the case, then existing customers may become dissatisfied because they are now forced to crosssubsidize new customers. Either way the private firm is likely to be blamed for the dilemma. Many of these problems are illustrated by the privatization of the Argentine telephone company in 1989. By most measures the reform was an astonishing success: in the first 10 years of private service the number of fixed lines increased from 3.0 to 8.4 million, the number of mobile lines from a handful to 6.4 million, the network was converted from analog to 100 percent digital, and waiting times for new phone service and repairs dropped dramatically. Meanwhile, the real price for a typical basket of residential services remained roughly constant while the price for a typical basket of commercial services dropped by roughly half (Urbiztondo and GomezIbanez, 2002). The reform was highly controversial, however, in part because the government, concerned about the price it would receive, raised tariffs just before the sale and gave the private companies exclusive rights to provide service in their territories for five years, with the possibility of a three year extension.8 Moreover, competition from international calling services forced the government to allow the companies to rebalance rates in 1997, dropping charges for long distance calls and increasing the basic monthly subscription fee. The initial increase was probably unavoidable because tariffs had been eroded badly by inflation in the late 1980s and the SOE was losing about $1.5 billion per year on the eve of the sale. But the combination of the increase and the monopoly 8 The exclusivity did not cover mobile telephones and data transmission services, where entry was permitted. 22 José A. Gómez-Ibáñez franchise made consumers suspicious from the start, and the subsequent rebalancing shifted costs from rural, highincome, and businesses customers who used more longdistance services to urban, lowerincome, and residential customers who consumed less. The average Argentine telephone customer must have benefited enormously, even considering the initial tariff increase, but the average hid a variety of circumstances including some who probably lost because they already had a longdistance line or made mostly local calls. Some argue that the frequent emphasis in privatization on costbased tariffs is technically as well as politically unwise. It may be inappropriate to charge tariffs that cover the full costs of services, such as water, which provide important public health or other benefits to nonusers. And the pressure to recover investments that have a 40 or 50year life within the span of a 15 to 25 year concession contributes further to excessive tariffs. A shift to costbased tariffs also typically involves decreasing the charges based on usage while increasing the fixed charges for a connection, changes that shift more of the financial burden onto poor households with low usage rates. In addition, critics point out that the industrialized countries often subsidized the development of their infrastructure, so it is hypocritical to discourage developing countries from doing so. But the argument about significant nonuser benefits applies to only a few types of infrastructure. Moreover, concessionaires often inherit a substantial existing facility and usually are required to recover, through tariffs, only the costs of operations and of the incremental investments they make. The notion that governments can continue to heavily subsidize infrastructure also ignores the budgetary constraints that drive many governments to private participation in the first place. This is not to deny that subsidies and cross subsidies may be useful to even out the distribution of costs and benefits in a privatization, but the reality is that they can be used only sparingly. The single most effective way to insure that the distribution of benefits and costs can be made politically acceptable is to search out projects where private participation generates substantial real efficiency gains and to avoid those where the primary effect of private involvement is to transfer costs or benefits from one party to another. Real efficiency gains involve reducing the amount of inputs that are required to produce a given output or improving the quality and quantity of output for a given input. Argentina's private telephone companies generated real efficiency gains, for example, by cutting the number of employees and the amount of capital needed per subscriber, and by providing millions of new lines at a cost much lower than subscribers were willing to pay. Transfers, by contrast, involve changes in who pays for inputs or who receives outputs without any changes in what outputs are produced or how. If, for example, private involvement is primarily an excuse to raise tariffs enough to cover costs or to lay off excess workers who have little prospect of finding other work, then it simply transfers resources from customers (who pay higher tariffs) or laidoff workers (who lose their salaries) to taxpayers (who no longer subsidize the SOE). Private Infrastructure in Developing Countries: Lessons from Recent Experience 23 The importance of real efficiency gains is that they provide the wherewithal to compensate parties that might otherwise lose from reform. If privatization is mainly about transfers, then it is a zerosum game: no one can win without someone else losing. But if privatization generates real efficiency gains, then there is at least the possibility that all parties can be made winners. The surpluses that Argentina's private telephone companies generated with new lines, for example, might have been used to soften the blow of rate rebalancing on residential customers or to provide assistance for older laidoff workers who might face a longer period of unemployment before they found other work. It is surprising how often even experienced politicians ignore the need to make the initial distribution of benefits and costs more equal and how important real efficiency gains are to making that distribution possible. For example, the collapse of the Cochabamba water concession, one of the most famous debacles in the history of recent privatizations in Latin America, was due in part to the insistence of the mayor of Cochabamba that the concessionaire be required to build the Misicuni dam and a connecting tunnel and aqueduct. Cochabamba had chronic water shortages so new sources of supply were needed, but a less expensive option was to extract more water from the existing reservoir system. The mayor was warned that building the Misicuni dam would require substantial tariff increases but he had been campaigning for the dam for years and must have seen the privatization as an opportunity to achieve a longsought goal. The inefficient investment in the dam made it hard to insure that users would be better off. Moreover, the concession called for immediate tariff increases, long before the dam would be complete, and for the expropriation of small wellfed water cooperatives that had developed on the periphery of the city because the public water company had not expanded its network. When the intensity of opposition to the tariff increases became apparent, the mayor joined the project's critics (Finnegan, 2002; Bechtel, 2005). Systems of Regulation A second source of controversy has been the systems created to regulate tariffs and service quality. The initial terms of the privatization establish a poor or a favorable starting point, but it is the regulatory system that must cope with that legacy and with any new problems that arise. The challenge is to design regulatory systems that can resolve disputes in a manner that is regarded as fair by both investors and consumers. At the risk of much simplification, two different approaches to regulating private infrastructure providers have been employed: contractual and discretionary (GomezIbañez, 2003). Under the contractual approach the government awards a private firm a longterm contract to improve and/or manage an infrastructure facility, usually through a process of competitive 24 José A. Gómez-Ibáñez bidding. The contract attempts to describe completely the obligations of the private firm to the government and vice versa during the life of the concession, and can not be changed unilaterally by either party. There is a regulatory agency, but its job is limited to monitoring the firm's compliance with the contract and applying any sanctions the contract calls for. The great advantage of the contractual approach is that it can prevent the concessionaire and the government from behaving opportunistically by fixing their obligations in the contract. Absent a contract, the government might press the concessionaire to offer very low tariffs knowing that the concessionaire's infrastructure investments are durable and immobile and so can't be easily moved to another community. Similarly, absent a contract the concessionaire may charge high tariffs, knowing that competition is unlikely if the infrastructure has the characteristics of a natural monopoly. In addition, awarding the contract by an open and competitive tender assures customers that the tariffs and other terms of the concession are fair. The main drawback of this approach is the risk that the contract will prove to be incomplete in that it fails to foresee an event that makes the agreement unworkable for one or both parties. In such cases, the parties face a difficult choice of whether to renegotiate the contract even though they are still vulnerable to one another or to live with the contract's unsatisfactory terms until it expires. The discretionary approach assumes that it is too difficult to draft a complete contract and instead establishes a regulatory agency that has the discretion to unilaterally change tariffs and other terms of service. The agency is typically insulated somewhat from pressures to behave opportunistically by, for example, separating it from the ministries that supervise other aspects of infrastructure policy, appointing the regulators who head the agency to fixed terms and making them removable only for specified causes, and funding the agency from a special surcharge on utility bills instead of through the normal legislative appropriations process. The agency remains accountable through its authorizing statute, which establishes the issues it must consider and the procedures it must follow in setting tariffs. And there is usually some procedure for consumers and firms to appeal the agency's decisions to a third party, often the courts or another independent tribunal. But the guidelines in the statute are usually broad, so that the agency has substantial scope in interpreting them. The great advantage of the discretionary approach is flexibility: there is no need to foresee all the possibilities and incorporate appropriate contingencies because the regulatory agency has the discretion to deal with the unanticipated. One disadvantage, however, is that the task of determining fair tariffs and other terms is technically very difficult, especially since the firm has more information about its costs than the regulatory agency.9 Of equal concern is the possibility that the agency will not use its discretion wisely but will instead be "captured" 9 The contractual approach avoids this problem by using the competition for the contract to insure that the tariff and other terms are fair. Private Infrastructure in Developing Countries: Lessons from Recent Experience 25 by "special" interests that prevent it from striking an appropriate balance between consumer and investor interests. And these drawbacks loom large in countries with a shortage of technical skills or a tradition of cronyism or corruption. The contractual approach has been used widely in developing countries because it appears so simple and fair, but it has suffered from two problems. One is that it has sometimes been difficult to attract a reasonable number of bidders, a difficulty that seems to have become more common as the controversy over private participation has grown. A second, and more fundamental, problem is that we have seriously underestimated the difficulties of writing workable contracts and the political costs of renegotiation. In a seminal study of over 1,000 infrastructure privatizations in Latin America and the Caribbean between 1982 and 2000, Guasch (2004) found that 74 percent of all water and sanitation contracts and 55 percent of all transportation contracts had to be renegotiated, most within only a few years of their signing. (Renegotiations were far rarer in telecommunications and electricity, but largely because the contractual approach is less common in those sectors.) Some of the early failures were due to inexperience in writing contracts in that they involved events that should have been anticipated or contingencies that were obviously inadequate. The problem has persisted, however, in part because the desire to finance investment typically leads to concessions of 10 to 30 years and it is extremely difficult to foresee all the important developments that will occur during the life of such a long contract, especially in a rapidly developing economy. It has proven particularly hard to write politically acceptable and economically practical contingencies for the major macroeconomic and foreign exchange crises that seem to plague developing countries periodically. Finally, there is the strong suspicion that many concessions deliberately submitted optimistic bids, anticipating that the government would renegotiate once the firm was in trouble rather than go through the disruption of retendering the concession. The renegotiation of concession contracts has been an enormous political liability for privatization, especially because citizens of developing countries are often mistrustful of their governments. Most contracts include provisions for arbitration or renegotiation to cope with unforeseen contingencies, but the clauses are typically modeled after those found in contracts between two private parties and do not recognize the public and political nature of disputes involving infrastructure concessions. In a dispute over a private contract the two parties negotiate directly, but in dispute over a concession contract the government negotiates with the firm on the consumers' behalf. And because it is hard to design a renegotiation that is as open and transparent as competitive bidding, the public's suspicions are often aroused and the legitimacy of the privatization program undermined. The discretionary approach is less common in developing countries but is used in activities, such as telecommunications and electricity distribution, where 26 José A. Gómez-Ibáñez technologies and demands are changing so rapidly that it is obviously difficult to write a complete contract. The record of discretionary regimes is much harder to evaluate if only because there is no explicit contract to breakdown. International institutions and consultants have been understandably concerned that the discretionary regulatory agencies have the technical skills and independence needed to do their jobs. And there is some evidence that skills and independence matter. In their study of 34 Latin American private utilities, for example, Sirtaine et al. (2005) found that the rate of return the firms earned tended to be closer to their cost of capital if the regulatory agency had a healthy budget and was governed by a commission with several members rather than a single person. But this emphasis on technical skills and independence has led some regulatory agencies to neglect the need to build and maintain popular support for the privatized industry. Controversial decisions are sometimes taken with only token public input and with little effort to build public acceptance, leaving the impression that the regulatory system is out of touch with, or even dismissive of, popular concerns. The opposite problem is also common: when the regulator views his primary role as defending consumers against monopoly, leaving the firm with so little popular support that it is difficult to approve realistic tariffs. The importance of understanding the political dimensions of regulation are illustrated, again, in the privatization of Argentina's telephone company. Certain decisions were deferred at the time of privatization, including when to end the private firms' period of exclusivity and how to rebalance tariffs as exclusivity ended. The architects of the reform presumably felt that the initial increase in tariffs and privatization were controversial enough for the time being, but they left the regulatory agency to deal with very difficult problems. Initially the agency was independent, but in 1996 it was converted into an office that reported directly to the Secretary of Telecommunications. The private companies petitioned for rate rebalancing as early as 1994, before the earliest date when exclusivity was scheduled to end, arguing that unexpected competition from international callback services was undermining their profitability. Public hearings showed how controversial the companies' proposal was, but when the regulatory agency released its own rebalancing scheme two years later, it initially refused to hold a public hearing and then, under the threat of lawsuits, held the hearing in a provincial capital. The agency argued that the previous hearings had been held in Buenos Aires and it was important to obtain views of other consumers, but critics contended that the location was chosen deliberately to avoid urban customers who would suffer most from the reforms. It is conceivable that the agency or the Secretary had been corrupted by the companies, and the lack of regulatory independence didn't allay those suspicions. But the staff was, to all appearances, technically competent and genuinely convinced that some form of rebalancing was inevitable. It would have been challenging enough to design a rebalancing proposal that was both Private Infrastructure in Developing Countries: Lessons from Recent Experience 27 economically realistic and politically acceptable. But any chance the agency had for selling such a scheme was lost by its insensitive behavior. A hybrid between the contractual and discretionary approaches has emerged in many developing countries. In this scheme a concession contract is awarded competitively for a term of 10 to 20 years but the contract specifies the tariffs only for the first 5 or 10 years. After that initial period, tariffs are readjusted periodically, often every 5 years, by a regulatory agency that enjoys substantial discretion. The same regulatory agency usually monitors contract compliance during the initial period so that it has a chance to develop expertise before it must set tariffs. The hope is to combine the best of the two approaches: the simplicity and fairness of competitive tendering with the provision for discretion when it is likely to be needed. However, the hybrid is also vulnerable to the weaknesses of both approaches since the contract must be reasonably complete during the initial period after which the regulatory agency must exercise its discretion wisely. A more radical option that deserves greater consideration is to dispense with regulation altogether. Private provision has been more successful in telecommunications than any other sector, and especially in the difficult regions of subSaharan Africa and the Middle East. The success is due to many factors: investors in mobile systems have less capital at risk (in part because customers are required to provide their own handsets), companies can easily cut off service to customers who don't pay their bills, and tariffs are less politically sensitive since telecommunications is considered less of a necessity as water or electricity. But a key component has been the willingness of many governments to regulate mobile telephony only loosely if at all, especially in countries where entry has been allowed and two or three mobile operators have grown large enough to provide effective competition for one another and the old fixedline company. The level of competition, while not perfect, may also be strong enough to allow regulation to be relaxed or abandoned in some other forms of infrastructure. Seaports, freight railways, and airports are the most obvious sectors where regulation might be largely dispensed with, particularly in large countries. Where regulation is necessary, more might be done to reduce the chances and political costs of contract renegotiation. Experience suggests that the risk of contractual disputes can be reduced significantly, for example, if sponsors take the time to be certain that the contract is clear on major issues and is based on the best available information so that there are fewer surprises. It also helps if the contract prohibits renegotiation for minor problems and if the bidding is transparent and very aggressive bids are scrutinized (Guasch, 2004, pp. 19­21 and Boston Institute for Developing Economies, 2006). In the case of discretionary schemes, it is important that the agency has reasonable resources and independence, but also that the regulators understand that true independence requires political skills and sensitivity. Finally, with either approach, it is helpful if a key investor in the company has experience and 28 José A. Gómez-Ibáñez intends to stay in the sector and the region for the long term, so that the firm's expectations are realistic and it behaves responsibly when problems arise. The High Cost of Private Capital A third challenge is the high cost of private capital. Private capital is often nominally more expensive than public capital, which makes it difficult for a private firm to hold down tariffs reasonable, especially since infrastructure is so capital extensive. Estache and Pinglo (2005, p. 59) estimate that the weighted average cost of capital (debt plus equity) for a typical private infrastructure provider around the turn of the century was 15 percent in a lowincome developing country, 11 percent in a low/middleincome developing country, and 10 percent for a middleincome developing country. The interest rates on sovereign debt for these countries were presumably several percentage points lower, however, since rates on sovereign debt reflect only the country risk and not the systematic risk inherent in a utility investment. And the interest rates on a loan from a development bank were probably as little as 2 to 3 percent during this period, while the special concessional loans available to the poorest countries from the World Bank amount to grants since they carry no interest and have very long repayment terms.10 Economists argue that the nominal difference in the cost of private and public capital is at least partly an illusion because public finance imposes hidden costs on taxpayers. Under private finance, private investors absorb the loss if the project proves to be a financial disaster and cannot repay its debts. But if the project is financed by sovereign debt or a loan to the government from a development bank, then the taxpayer must cover the loss if the project can not pay. This argument assumes that the private investor assumes all or most of the commercial risks, which is not always the case. Many of the takeorpay contracts for new public power plants that developing countries signed during the 1990s were structured similarly to sovereign debt, for example, in that the government assumed both the demand and the foreign exchange risks (Wells and Ahmed, 2007). Even where the assumptions hold, moreover, the argument that the difference in interest rates is an illusion is not likely to be understood by the general public. But the public is sure to notice if utility tariffs rise when private debt and equity replace public debt. These problems are compounded because private investors in infrastructure are often foreigners rather than nationals. Foreign involvement stems in part from a desire for technical expertise that is not available locally. Consortiums bidding for concessions are usually required to include a member with a track 10 Estache (2005, p. 20) reports that taking into account commitment and other fees, the effective interest rates on normal development loans from the World Bank was 44 basis points over LIBOR with other development banks ranging from 35 to 154 basis points over LIBOR. Private Infrastructure in Developing Countries: Lessons from Recent Experience 29 record of operating similar services, for example, and that often means a foreign firm in countries where there is little recent local experience with private infrastructure. But foreigners also dominate because the capital markets in many countries are insufficiently developed or too illiquid to raise sufficient funds, especially for larger transactions. The thinness of local capital markets is often due not just to the low income levels of the countries but to weak legal protections for private property, contracts, and the rights of minority shareholders. This dependence on foreign capital often comes at a cost, the most obvious being the potential for resentment and misunderstanding on both sides. Foreigners may suspect that they will be treated poorly simply because they are foreign, and locals often come to resent the fact that foreigners own and profit from essential local services. The potential for misunderstanding is sometimes heightened by inexperience, as many of the European and American infrastructure companies that went abroad in the 1990s had never invested in developing countries before. The involvement of foreigners also generates additional issues to quarrel over, notably the treatment of foreign exchange risk. Most foreigners want to repatriate their profits eventually. Since swaps and other risk management instruments are not available at reasonable cost for the currencies of many developing countries, investors and lenders often request exchange rate guarantees or the indexing of utility tariffs to foreign currency movements. If a subsequent exchange crisis forces the government to devalue the local currency, as in Asia in 1997 or Argentina in 2002, consumers find it hard to understand why the foreignowned utilities expect to be made whole while the rest of the country's households and businesses suffer. Political risk poses a particularly difficult problem. Investors prefer to avoid risk, all else being equal, so the riskier an investment, the higher the return investors require. Infrastructure investments in developing countries are usually perceived as risky, if only because the services are politically sensitive and the facilities durable and immobile. It can be hard for an investor to withdraw his capital in the event of a dispute with government over tariffs or other issues. Yet giving the investors a substantial premium for perceived political risk may be selfdefeating if it fuels popular outrage about the returns the investors require. Finding ways to reduce real or perceived risks would seem better than simply passing them on in the form of higher returns and tariffs. Unfortunately, several schemes introduced in the last 20 years to make foreign investment in developing countries safer do not seem to have worked well. Political risk insurance has become more widely available, offered by the foreign investors' home government and by international financial institutions such as the World Bank (through MIGA, the Multilateral Investment Guarantee Agency) and the Asian Development Bank. In addition the World Bank's International Financial Corporation (IFC), the European Bank for Reconstruction and Development (EBRD), and other regional banks have invested in private 30 José A. Gómez-Ibáñez infrastructure projects that they believe are attractive but might not be funded without their involvement and implicit endorsement. But the most important innovation has been the spread of bilateral investment treaties that give foreign investors the option of international arbitration if they feel mistreated by their host government. Many of these treaties call for arbitration at the International Center for the Settlement of Investment Disputes (ICSID), which is housed at the World Bank, and allow foreign investors to collect on damages awarded by attaching the government's overseas assets. To the extent that MIGA, IFC, and the regional banks are more adept at evaluating political risk than less experienced investors, their insurance or participation may provide a useful signal that risks are lower than they appear. But if others invest primarily because they believe that the involvement of the World Bank or the regional banks is a deterrent to aggressive government behavior, then those investors may be tempted to behave irresponsibly by taking a tough line in any future controversy. In addition, access to international binding arbitration does not seem to have reduced conflict: the ICSID has been swamped with 69 disputes initiated by private infrastructure providers between 1997 and 2007, many in the wake of the Asian and Argentine financial crises, and the resulting backlog will take years to resolve. More important, Wells and Ahmed (2007) argue that the arbitration procedures are so rigid and onesided as to be unworkable in the long term. ICSID arbitrators have been stricter in enforcing contract terms and calculating damages than the courts of developed countries are in similar domestic contractual disputes. And while foreign private investors can take developing countries to arbitration and seize their overseas assets for not complying with promises made, the developing countries can not take the investors to arbitration or seize their overseas assets when the companies fail to keep their part of the bargain. One obvious solution to the higher nominal costs of private finance is to use as much lowcost public financing as possible. Public finance is unlikely to be sufficient to finance all the investments needed, however, so that some reliance on private finance is inevitable. The World Bank and other donors made the shortage worse by their reluctance to lend to infrastructure SOEs during the 1990s, when enthusiasm for privatization was at its height. That policy has been relaxed and development agencies are now more willing to lend to governments to finance assets that private concessionaires might build or manage, thus hopefully getting something of the best of both worlds. Mixing public finance with private management creates some complications: for example, it is harder to write a contract that holds private managers accountable for the performance of the firm when the design and timing of investments is not under their control. But mixed finance can help hold utility tariffs down. Another possibility that deserves more attention is to reduce the reliance on foreign financing by strengthening the domestic capital markets of developing countries. Some countries have developed local markets for infrastructure debt. Private Infrastructure in Developing Countries: Lessons from Recent Experience 31 Banks are not ideal financers of infrastructure because they usually insist on floating rates and relatively short terms. But local banks have played an important and apparently healthy role in financing infrastructure in a number of countries, such as Thailand. And in Mexico, banks were able to sell infrastructure bonds directly to depositors during the early 1990s. Pension funds and insurance companies are more able to offer fixed rates and long terms, and several countries, including Chile, have managed to persuade pension and insurance regulators that infrastructure debt is not too risky an investment. Debt from banks, pension funds, and insurance companies is attractive because they usually have enough at stake and enough sophistication to monitor their investments carefully. Local equity has been raised for many infrastructure projects as well, although usually it is held by a few firms or individuals rather than widely. Widening the equity market would be helpful not only to increase the potential pool of capital but also to potentially strengthen popular support for private companies. But widespread participation in equity markets requires laws and regulations to protect minority shareholders and courts that are willing to enforce them, both ingredients often lacking in many developing countries. Finally, anything that can be done to reduce political risk will be helpful since political risk greatly increases capital costs. Foremost among the possibilities are insuring that the distribution of benefits and costs in the initial privatization and subsequent regulation are fair. Adapting to Difficult Circumstances A final challenge is to adapt to circumstances where private provision of infrastructure is particularly difficult. The current fashion among scholars of economic development is to focus on the institutions required for successful reform. The argument that institutions matter is hard to deny: private participation in infrastructure is bound to be more difficult where courts lack the independence and integrity to enforce contracts, for example, or where corruption makes investors or consumers wary that they will get a fair hearing before a government regulatory agency. But it is hard to know how to adapt private participation to circumstances where institutions are weak, beyond limiting the investments at risk or abandoning the approach altogether. Both the contractual and discretionary approaches to regulation depend on the integrity of government, for example. Integrity is probably easier to maintain in the commercial courts than in a discretionary regulatory agency, if only because all sectors of the economy have a stake in the performance of the courts while the regulated infrastructure companies and their customers are the only ones with a stake in the performance of the specialized regulatory agency. But in a very weak institutional environment sustaining private participation is likely to be difficult unless the investors are connected to the political elite, have relatively little 32 José A. Gómez-Ibáñez capital at risk, or enjoy some special advantage in technology or other areas that protects them from takeover. Other basic characteristics that affect the prospects of private participation in infrastructure include the level of income, population, and geography of the countries involved. Private participation has proven to be much harder in poor, small, and largely rural countries. All regions have countries with these characteristics: Bolivia, Ecuador, Honduras, and Nicaragua in Latin America, for example, or Cambodia, the Lao People's Democratic Republic, and Papua New Guinea in Asia. But the largest concentration of such countries is in subSaharan Africa. SubSaharan Africa has an average per capita income of $496, slightly higher than South Asia ($441) but well below Latin America ($3,888) or Eastern Europe and Central Asia ($1,998). The countries in the region are, on average, small and rural: only 3 of the 47 countries have populations above 40 million,11 while another 7 have populations between 40 and 15 million.12 The remaining 37 countries have populations of less than 15 million, and an average population of 4 million.13 Only 33 percent of the population in subSaharan Africa lives in urban areas, a figure comparable to that in Southeast Asia (39 percent) but less than half that of Latin America (77 percent). It is no coincidence that subSaharan Africa received less than 4 percent of the private investment in infrastructure in developing countries between 1990 and 2003 or that half of the investment in subSaharan Africa went to a single country--South Africa--with the remaining 46 countries dividing up the rest. Moreover, 65 percent of the investment in subSaharan Africa went to one sector: telecommunications. The second most important sector, energy, received only 22 percent of the investment, mostly in the form of new generating stations built by Independent Power Producers (IPPs) and backed by takeorpay contracts with stateowned electricity companies. There has been relatively little private investment in electricity distribution and transmission and almost none in water and sanitation, although not for the lack of trying. In the electricity sector, for example, approximately half of the countries in subSaharan Africa have attempted to involve the private sector in the management of their distribution companies, but in most cases either the effort was aborted before the contract was awarded or the contract was cancelled after only a few years (Benoit, 2005). The number of success stories in electricity or water distribution can be counted on one hand.14 11 Nigeria with 121 million, Ethiopia with 69 million, Democratic Republic of Congo with 51 million, and South Africa with 45 million (total population 286 million). 12 Tanzania with 36 million, Kenya 32 million, Uganda 25 million, Mozambique 19 million, Madagascar 17 million, Côte d'Ivoire 16 million, and Cameroon 15 million (total population 160 million). 13 The 37 smallest countries had a combined population of 166 million persons out of the total 605 million persons who lived in subSaharan Africa in 2002. 14 The usual list of successes includes the water services in Côte d'Ivoire (which has been operated as a private concession since 1960), Mozambique, and Senegal; and electricity in Uganda. Private Infrastructure in Developing Countries: Lessons from Recent Experience 33 Poverty makes private participation difficult in part by increasing problems of access and affordability. Access to infrastructure services is a much more significant problem in subSaharan Africa than in the rest of the world. The disparity is greatest in electricity: in 2002 only 15 percent of households have access to electricity in the average subSaharan country compared to 31 percent in the average South Asian country, 79 percent in Latin America, and 88 percent in Middle East and North Africa. But subSaharan Africa is also behind other regions in access to clean water sources and sanitation, and telecommunications (Estache, 2005). As for those who are connected, affordability is a potentially more serious problem. One study estimated that 55 percent of households in Africa and India would have difficulty paying electricity and water bills of US$10 per month, an amount roughly sufficient to cover costs, whereas only 35 percent of households in East Asia and 10 percent of households in Latin America would experience similar difficulties.15 In a poor country the infrastructure companies are also much more economically and politically prominent, which further complicates reform. The companies are often among the largest and most profitable enterprises in the economy, particularly in countries where the private sector is dominated by smallscale agriculture and traders. Since the utilities are among the few places where large amounts of cash are concentrated, they become attractive targets for governments in need of resources. In Africa it is common for government agencies not to pay their electricity, water, or telephone bills, for example, only the most obvious of many means to siphon off utility resources to support government activities. And because the utilities are among the few large enterprises, they offer some of the only wellpaid professional and managerial jobs in the country. Private participation becomes more problematic because it is harder for the government to commit to a defined share of the utilities' resources. Furthermore, private participation becomes more controversial if expatriates assume some of the best jobs in the country, usually earning many times more than the nationals they replaced. Finally, poverty makes countries more dependent on and resentful of foreign aid and advice. Many of the early adopters of private infrastructure in Latin America and Asia--such as Argentina, Brazil, Mexico, and India--did so because macroeconomic or sectoral crises convinced the governments that reform was necessary. While crises have played a role in encouraging local support for reform in subSaharan Africa as well, the more decisive factor seems often to be pressure from multilateral or bilateral aid agencies. In the 1990s particularly, private participation was usually imposed as a condition for further loans or grants to the infrastructure sector, and adopted grudgingly by national governments. The pressure from aid agencies helps the governments avoid 15 The threshold for affordability was set at 5 percent of income based on the observation that very few poor households spend more than 5 percent of their income on water or more than 6 to 8 percent of their income on electricity (Foster and Yepes, 2006, p. 17). 34 José A. Gómez-Ibáñez responsibility for the performance of the sector, since they can always blame outside forces. And the lack of local conviction means that the policy reforms are poorly implemented and closely monitored by critics, ready to seize on the problems that usually develop. Many of these problems are magnified if the population is small and the geographic area is large. Small populations and low densities make it harder to exploit economies of scale in management and the physical plant. A small national company can not justify or afford highly specialized or expert staff, a disadvantage that a large international company can overcome, but at the risk of provoking local resentment. Moreover, demand is often insufficient to support efficientscale plants, particularly if the population is dispersed. In electricity, for example, the minimum efficientscale gasfired generating station is around 100 to 200 MW, which is as much as the entire generating capacity of some of the smallest countries. And the customers in secondary cities are often served by isolated distribution systems powered by very small dieselpowered generators. Similarly, a modern container terminal requires a volume of 200,000 to 300,000 twentyfoot equivalent units (TEUs) per year, when Nigeria, the most populous country in subSaharan Africa, generates only about 400,000 to 500,000 TEUs per year. The inability of many small countries to support a minimum efficientscale electricity plant or container port on their own, much less to foster competition among several electricity generators or ports, increases the costs of their utility services. The small population also strains the pool of talent available to administer reform. The task of formulating infrastructure policy or regulating a private utility does not decrease in proportion to the size of the country, so that small countries always have a more difficult time staffing policy and oversight bodies. In a small country, moreover, the group with the education or experience to fill the key policy and regulatory jobs is typically relatively small and its members well known to one another. The situation makes it especially difficult to pick a regulator who will be perceived as independent since most candidates usually have a history of working closely with (or strongly against) either the private officials in charge of the regulated company or the government officials in charge of infrastructure policy. Many of these problems are illustrated by the privatization of Mali's electricity and water company, Energie du Mali (EDM), which provides electricity to 13 percent of the country's 13 million residents. Mali's electricity is expensive because the country is landlocked and diesel fuel has to be trucked many kilometers from the coast. In addition, 20 of the 32 communities served by EDM are so remote that they are not connected to the national grid but served by small isolated distribution systems. The country's hopes for cheaper electricity have historically rested on developing its hydro resources, and foreign donors have used loans for the dams as leverage to force a series of reforms on EDM. EDM built a small dam in 1967 and a larger one in 1980, both with French Private Infrastructure in Developing Countries: Lessons from Recent Experience 35 technical and financial aid. In the mid1980s the World Bank and other aid agencies discovered that the turbines in the second dam were not being adequately maintained, and conditioned loans for rehabilitation on management reforms at EDM. These reforms failed to take hold so when Mali, Senegal, and Mauritania wanted to build a third and much larger dam at Manantali, the key donors met in Paris and agreed that Mali would receive its share of the funding only if it hired a private company to manage EDM, raised tariffs, and increased collection rates. A fouryear management contract was awarded to a consortium headed by a large French utility company, Saur, in 1995 but the effort collapsed after three years amidst disputes between Mali and Saur about why EDM's performance had not improved. The donors then insisted that EDM be offered as a 20year concession, so that the private mangers would have greater control and responsibility for performance. The concession was awarded through competitive bidding to Saur again in 2000 only to see Saur withdraw in 2005. The immediate cause for the concession's breakdown was a major flaw in the tariff adjustment formula that became apparent in 2002. The formula called for a significant increase in electricity tariffs that year even though Manantali's turbines were coming on line, allowing EDM to replace expensive diesel with cheap hydro power. The tariff formula might have been renegotiated, however, had it not been for deep resentment and poor behavior on all sides. The donors were partly responsible for imposing reform, although they had some internal allies and their desire not to see aid wasted was understandable. Saur bore a big share of the blame by insisting for too long that the contract should be honored despite the obvious injustice of the tariff formula and then by not being very forthcoming with data needed to set a reasonable tariff. But the company made an easy target for politicians and became an issue in the presidential elections, especially since Mali's electricity tariffs were still among the highest in West Africa and the compensation for the 12 expatriate managers was reported to amount to onequarter of the payroll for the over 5,000 Malians who worked for the company. The technical staff in the Ministry of Mines, Energy and Water was also unsympathetic, possibly because many of them were exEDM officials who thought they could do at least as well as the expatriate managers. The regulatory agency, although nominally independent, was headed by an exEDM general manager and who seemed to compete with the politicians in attacking the company publicly, which left the firm completely isolated with no local support. Not surprisingly, fact finding and mediation sponsored by the World Bank and other donors failed. One obvious solution to the challenges faced by the poorer, smaller, and more rural countries is to promote regional cooperation in the provision of infrastructure. The attraction of this approach is, of course, the potential to reduce costs by exploiting economies of scale. In transportation, some types of regional infrastructure are emerging through the normal competitive process as, for example, Abidjan, Tome, and Lagos vie to become the leading container port 36 José A. Gómez-Ibáñez in their corner of the African coast. In other sectors, regional infrastructure is likely to require deliberate planning by the countries and donors, as in the case of the Manantali dam or the proposal to build highvoltage electricity transmission lines connecting gasproducing states on the coast of West Africa with their gas poor neighbors. Regional systems require cooperation and raise understandable concerns about dependence and reliability, but the prospect of lower costs can be a powerful lure. Regional infrastructure can be compatible with private participation, especially if it is advantageous to have the common facility managed by an independent party. The Manantali dam, for example, is managed on behalf of Mali, Mauritania, and Senegal by Eskom, South Africa's electricity company. An alternative but complementary approach that governments and donors are beginning to explore is very small scale infrastructure. At first glance small scale systems would seem to increase a poor country's problems rather than solve them. Small scale suggests higher unit costs. If people with the skills needed for reform are stretched thin at the national level, moreover, then they are likely to be even scarcer at the local level. Indeed, it is notable that many of the highprofile infrastructure disputes mentioned earlier involve municipal or other subnational governments. Not only do subnational governments have fewer skills and resources to resolve disputes, but they also have a narrower perspective than the national governments, and thus may be less concerned about the chilling effect that a serious dispute may have on investments in other sectors or parts of the country. But small scale is not necessarily high cost if the problem is to increase access to infrastructure services in secondary cities or remote communities, as is often the case in poor and rural countries. An isolated electricity system sacrifices economies of scale in generation, for example, but saves on the cost of long distance transmission lines. And when systems are small they sometimes can be built to different standards with, for example, water pipes on the ground or buried shallowly instead of in deep trenches. As the community grows and the system expands it can be gradually upgraded and eventually, in the case of electricity, connected to the national grid. A key advantage of small systems is that they might be built and operated by local firms and financed with local capital, thus easing the resentments and complications of foreign investment. Small systems have very low capital requirements not just because they are small but also because the technologies used are often involve a lower ratio of capital to operating costs. A small diesel generator has relatively high operating costs for fuel and maintenance but low initial costs, for example, and pipes above ground or in shallow trenches are harder to maintain but avoid the investment in excavation. One sign that this strategy holds promise is that small private utility systems are surprisingly common, even in the absence of government efforts to promote and support them. Statistics are hard to come by since these small providers are Private Infrastructure in Developing Countries: Lessons from Recent Experience 37 not typically subject to effective government regulation or reporting requirements. Based on a review of 400 infrastructure studies and reports for discussions of small providers, Kariuki and Schwartz (2005) estimated that there were approximately 7,000 small electricity providers in 32 developing countries and 10,000 small water providers in 49 developing countries. Many of the small water providers operate standpipes or tankers distributing water generated by conventional utilities, but others have independent water sources and small distribution networks. The small electricity providers usually operate small diesel or hydro generating plants and distribution networks, or sell and service home diesel or solar generating systems. Many of the systems are cooperatives, particularly in countries with a tradition of cooperatives like Kenya, Colombia, and Bangladesh, while others are owned and operated by individuals or families. Small private providers are most common in poor and rural countries and, as one might expect, in small towns or in settlements on the periphery of cities where the conventional utility has failed to extend its network. Despite their numbers, the small providers still don't reach many people. Kariuki and Schwartz (2005) estimate, for example, that small electricity providers serve only 10 to 50 million customers and mostly in Asia. Assistance on finance, engineering, and regulatory issues is probably needed if small systems are to be more widely implemented, but so far there have been only a few pilot efforts of this type. Kenya is experimenting with means of encouraging microfinance institutions to lend to small private utilities, for example, and Nicaragua is providing government aid (Mehta, Virjee, and Njoroge, 2007; Ley, Martinez, and Foster, 2006). But some of the efforts appear cumbersome and perhaps overly committed to the use of sustainable technologies, such as solar or hydro power, at the expense of providing affordable and quality service. Moreover, while financial aid is undoubtedly helpful, technical and regulatory assistance may be equally valuable. On the engineering side, the national government or some other institution might recommend designs for systems that are safe yet easy to build and maintain. On the regulatory side, the government might draft and distribute model concession contracts and maintain an arbitration panel to resolve disputes between communities and operators. The Rural Utilities Service of the U.S. Department of Agriculture provides a model in that it offers technical assistance as well as loans and grants for small cooperative and forprofit electricity, water, and telephone systems in the United States. The agency is the successor to the Rural Electrification Administration, which was established in 1935 at the height of the depression and is considered one of the great successes of President Roosevelt's New Deal. This strategy might leave the companies serving the largest urban areas, such as EDM, still in public hands, while private providers run both the regional and the very small systems. The efficiency of the companies in the large urban areas probably would be improved by private involvement, but experience tells 38 José A. Gómez-Ibáñez us that private participation in these firms is very difficult to establish and maintain, and it may be possible to get some improvement in their performance as public enterprises (GomezIbañez, 2007). More promising is the possibility that, over time, the capacity of local private infrastructure providers in small towns will develop to the point where they can take over the larger systems. Private Infrastructure in Developing Countries: Lessons from Recent Experience 39 References Andres, Luis, Vivien Foster, and Jose Luis Guasch (2005). "The Impact of Privatization on Firms in the Infrastructure Sector in Latin American Countries." World Bank, Washington, DC. ------ (2006). "The Impact of Privatization on Performance of the Infrastructure Sector: The Case of Electricity Distribution in Latin America." World Bank Working Paper WPS 3936. Washington, DC. Bechtel (2005). "Cochabamba and the Aguas del Tunari Consortium." Accessed on the Bechtel Website. Benoit, Philippe (2005). "Viewpoint: A Snapshot of PPI in SubSaharan Africa's Power Sector." Draft January 12. Birdsall, Nancy, and John Nellis (2005). "Privatization Reality Check: The Distributional Impact of Privatization in Developing Countries." Pp. 1­32 in John Nellis and Nancy Birdsall, eds., Privatization Reality Check: The Distributional Impact of Privatization in Developing Countries. Washington, DC: Center for Global Development. Boston Institute for Developing Economies (2006). "Impact of Privatization in Africa: Synthesis of Eight Cases." Report for the World Bank, October. Brune, Nancy, Geoffrey Garrett, and Bruce Kogut (2004). "The IMF and the Global Spread of Privatization." International Monetary Fund Staff Working Papers 51(2): 195­219. Carrera, Jorge, Daniele Checchi, and Massimo Florio (2005). "Privatization Discontent and Its Determinants: Evidence from Latin America." Manuscript, March. Chisari, O., Antonio Estache, and C. Romero (2000). "Winners and Losers from Privatization and Regulation of Utilities: Lessons from a General Equilibrium Model of Argentina." World Bank Economic Review 13(2): 357­ 378. Clarke, George, Claude Menuard, and Ana Maria Zuluaga (2000). "The Welfare Effects of Private Sector Participation in Guinea's Water Supply." World Bank Policy Research Paper no. 2361. Washington, DC. Denisova, Irina, Markus Eller, Timothy Frye, and Ekaterina Zhuravskaya (2007). "Who Wants to Revise Privatization and Why? Evidence from 28 Post Communist Countries." Center for Economic and Financial Research, New Economic School, Moscow, Working Paper no. 105. 40 José A. Gómez-Ibáñez Ennis, Huberto M., and Santiago M. Pinto (2005). "Argentina's Privatization: Effects on Income Distribution." Pp. 179­217 in John Nellis and Nancy Birdsall, eds., Privatization Reality Check: The Distributional Impact of Privatization in Developing Countries. Washington, DC: Center for Global Development. Estache, Antonio (2005). "What Do We Know about SubSaharan Africa's Infrastructure and the Impact of its 1990s Reforms?" Manuscript, version 4, May 24. Estache, Antonio, and Maria Elena Pinglo (2005). "Are Returns to Private Infrastructure in Developing Countries Consistent with Risks Since the Asian Crisis?" Journal of Network Industries 6(1): Estache, Antonio, and Martín A. Rossi (2002). "How Differences in Efficiency of State and Private Water Companies Vary." World Bank Economic Review, 16(1): 139­148. ------ (2005). "Do Regulation and Ownership Drive Efficiency of Electricity Distribution? Evidence from Latin America." Economic Letters 86: 253­257. Estache, Antonio, Vivien Foster, and Quentin Woodon (2002). Accounting for Poverty in Infrastructure Reform: Learning from Latin America's Experience. WBI Development Series. Washington, DC: World Bank. Estache, Antonio, M. Gonzales, and Lourdes Trujilo (2002a). "Efficiency Gains from Port Reform and the Potential for Yardstick Competition: Lessons from Mexico." World Development 30(4): 545­560. ------ (2002b). "What Does Privatization Do for Efficiency? Evidence from Argentina's and Brazil's Railways." World Development 30(11): 1885­1897. European Bank for Reconstruction and Development (EBRD) (2007). "Public Private Partnerships and EBRD." Manuscript, March. Fay, Marianne, and Mary Morrison (2007). Infrastructure in Latin America and the Caribbean: Recent Developments and Key Challenges. Directions in Development series, no. 37899. Washington, DC: World Bank. Finnegan, William (2002). "Leasing the Rain." The New Yorker, April 8, pp. 43­53. Foster, Vivien, and Tito Yepes (2006). "Is Cost Recovery a Feasible Objective for Water and Electricity? The Latin American Experience." World Bank, Working Paper WPS3943. Washington, DC. Gassner, Katherina, Alexander Popov, and Nataliya Pushak (2007). "An Empirical Assessment of Private Sector Participation in Electricity and Water distribution in Developing and Transition Countries." Manuscript, June. GómezIbáñez, José A. (2003). Regulating Infrastructure: Monopoly, Contracts, and Discretion. Cambridge: Harvard University Press. Private Infrastructure in Developing Countries: Lessons from Recent Experience 41 ------ (2005). "EDM (Energie du Mali)." Kennedy School of Government Case Program, case no. 1811.0 ------ (2007). "Alternatives to Infrastructure Privatization Revisited: Public Enterprise Reform from the 1960s through the 1980s." World Bank, Working Paper no. 4391. Washington, DC. Guasch, José Luis (2004). Granting and Renegotiating Infrastructure Concessions: Doing it Right. World Bank Institute Development Studies. Washington, DC: World Bank. Harris, Clive (2003). "Private Participation in Infrastructure in Developing Countries." World Bank Working Paper no. 5. Washington, DC. Harris, Clive, John Hodges, Michael Schur, and Padmesh Shukla (2003). "Infrastructure Projects: A Review of Cancelled Private Projects." In Public Policy for the Private Sector, World Bank Group Private Sector and Infrastructure Network, note 252, January. Washington, DC: World Bank. Izaguirre, Ada Karina (2007). "2006 Data Results from the PPI Projects Database: Preliminary Results." World Bank, presentation slides. Washington, DC. Kariuki, Mukami, and Jordan Schwartz (2005). "Small Scale Private Service Providers of Water Supply and Electricity: A Review of Incidence, Structure, Pricing and Operating Characteristics." World Bank Policy Research Working Paper no. 3727. Washington, DC. Latinobarometro (2005). "Lationbarometro Report 2005, 1995­2005: A Decade of Public Opinion, 176,544 Interviews, Ten Waves in 18 Countries." Santiago: Latinobarometro. Ley, Debora, Herminia Martinez, and Robert Foster (2006?). "Nicaraguan Renewable Energy for Rural Zones Program Initiative." Manuscript. Macedo, Roberto (2000). "Privatization and the Distribution of Assets and Income in Brazil." Carnegie Endowment for International Peace, Global Reform Project, Working Paper no. 14. McKenzie, David, and Dilip Mookerjee (2003). "The Distributive Impact of Privatization in Latin America: An Overview of Evidence from Four Countries." Economía 3: 161­218. ------ (2005). "Paradox and Perception: Evidence from four Latin American Countries." Pp. 33­84 in John Nellis and Nancy Birdsall, eds., Privatization Reality Check: The Distributional Impact of Privatization in Developing Countries. Washington, DC: Center for Global Development. Mehta, Meera, Kameel Virjee, and Serah Njorge (2007). "Helping a New Breed of Private Water Operators Access Infrastructure Finance: Microfinance for Community Water Schemes in Kenya." PublicPrivate Infrastructure Advisory Facility, Gridlines note no. 25. 42 José A. Gómez-Ibáñez Mussa, Michael (2002). Argentina and the Fund: From Triumph to Tragedy. Washington, DC: Peterson Institute for International Economics, Policy Analysis in International Economics series no. 67. Nellis, John (2005). "Evolution of Enterprise Reform in Africa: From Stateowned Enterprises to Private Participation in Infrastructure--and Back?" World Bank, photocopied, draft of April 15. Washington, DC. ------ (2006). "Privatization: A Summary Assessment." Center for Global Development, Working Paper no. 87. Plane, P. (1999). "Privatization, Technical Efficiency and Welfare Consequences: The Case of the Côte d'Ivoire Electricity Company." World Development 27(2): 343­360. Ros, Augustin (1999). "Does Ownership Matter? The Effects of Telecommunications Reform on Network Expansion and Efficiency." Journal of Regulatory Economics 15: 65­92. Ros, Augustin, and A Banerjee (2000). "Telecommuincations Privatization and Tariff Rebalancing: Evidence from Latin America." Telecommunications Policy 20: 233­252. Shirley, Mary M. (2005). "Why is Sector Reform So Unpopular in Latin America?" The Independent Review 10: 195­207. Sirtane, Sophie, Maria Elena Pinglo, J. Luis Guasch, and Vivien Foster (2005). "How Profitable Are Infrastructure Concessions in Latin America? Empirical Evidence and Regulatory Implications." Washington, DC: Public Private Infrastructure Advisory Facility, Trends and Policy Options series, no. 2. Urbiztondo, Santiago, and Jose A. GomezIbañez (2002). "Argentina's Telecommunications Industry and the Economic Crisis of 2002." Kennedy School of Government, Case Program, case 1678.0. Wallsten, Scott (2002). "Does Sequencing Matter? Regulation and Privatization in Telecommunications Reforms." World Bank, Policy Research Working Paper 2817. Washington, DC. Wells, Louis T., and Alexandra de Royere (2005). "Aguas Argentinas: Settling a Dispute." Harvard Business School, case no. 9705019.t. Wells, Louis T., and Raqif Ahmed (2007). Making Foreign Investment Safe: Property Rights and National Sovereignty. Oxford: Oxford University Press. Private Infrastructure in Developing Countries: Lessons from Recent Experience 43 Eco-Audit Environmental Benefits Statement The Commission on Growth and Development is committed to preserving endangered forests and natural resources. The World Bank's Office of the Publisher has chosen to print these Working Papers on 100 percent postconsumer recycled paper, processed chlorine free, in accordance with the recommended standards for paper usage set by Green Press Initiative--a nonprofit program supporting publishers in using fiber that is not sourced from endangered forests. For more information, visit www.greenpressinitiative.org. The printing of all the Working Papers in this Series on recycled paper saved the following: Trees* Solid Waste Water Net Greenhouse Gases Total Energy 48 2,247 17,500 4,216 33 mil. *40 inches in height and 6­8 Pounds Gallons Pounds CO2 Equivalent BTUs inches in diameter The Commission on Growth and Development Working Paper Series 34. International Finance and Growth in Developing Countries: What Have We Learned, by Maurice Obstfeld, August 2008 35. Policy and Institutional Dynamics of Sustained Development in Botswana, by Gervase Maipose, August 2008 36. Exports of Manufactures and Economic Growth: The Fallacy of Composition Revisited, by William R. Cline, August 2008 37. Integration with the Global Economy: The Case of Turkish Automobile and Consumer Electronics Industries, by Erol Taymaz and Kamil Yilmaz, August 2008 38. Political Leadership and Economic Reform: the Brazilian Experience, by Henrique Cardoso and Eduardo Graeff, September 2008 39. Philippines Case Study: The Political Economy of Reform during the Ramos Administration (1992­98), by Romeo Bernardo and Christine Tang, September 2008 40. Making Difficult Choices: Vietnam in Transition, Martin Rama, based on conversations with H. E. Võ Vn Kit, with Professor ng Phong and oàn Hng Quang, November 2008 41. Policy Change and Economic Growth: A Case Study of South Africa, by David Faulkner and Christopher Loewald, November 2008 42. International Migration and Development, by Gordon H. Hanson, January 2009 43. Private Infrastructure in Developing Countries: Lessons from Recent Experience, by José A. Gómez-Ibáñez, January 2009 Forthcoming Papers in the Series: From Financieristic to Real Macroeconomics: Seeking Development Convergence in EEs, by Ricardo French-Davis (January 2009) Setting Up a Modern Macroeconomic-Policy Framework in Brazil, 1993­2004, by Rogério L. F. Werneck (January 2009) Electronic copies of the working papers in this series are available online at www.growthcommission.org. They can also be requested by sending an e-mail to contactinfo@growthcommission.org. B eginning in the late 1980s many developing countries turned to the private sector to provide basic infrastructure and utility services, such as highways, railroads, water, wastewater, electricity, gas, and telecommunications. Recent Commission on Growth and Development studies suggest that private involvement often benefited customers and reduced Montek Ahluwalia Edmar Bacha government fiscal problems without harming employees or enriching private Dr. Boediono providers excessively. There were enough high-profile failures, however, to Lord John Browne discredit this reform in many quarters. Private involvement is likely to be more Kemal Dervis ¸ successful if it generates real efficiency gains rather than simply transferring costs Alejandro Foxley among parties, if the systems of regulating the private companies are politically Goh Chok Tong sensitive as well as technically competent, if the costs and constraints of private Han Duck-soo capital are not excessive, and if we are willing to adopt more modest and gradual Danuta Hübner schemes in difficult circumstances. Carin Jämtin Pedro-Pablo Kuczynski José A. Gómez-Ibáñez, Professor, Harvard University Danny Leipziger, Vice Chair Trevor Manuel Mahmoud Mohieldin Ngozi N. Okonjo-Iweala Robert Rubin Robert Solow Michael Spence, Chair Sir K. Dwight Venner Ernesto Zedillo Zhou Xiaochuan The mandate of the Commission on Growth and Development is to gather the best understanding there is about the policies and strategies that underlie rapid economic growth and poverty reduction. The Commission's audience is the leaders of developing countries. The Commission is supported by the governments of Australia, Sweden, the Netherlands, and United Kingdom, The William and Flora Hewlett Foundation, and The World Bank Group. www.growthcommission.org contactinfo@growthcommission.org