Lessons of Experience No. 4: Financing Private Infrastructure Executive Summary

By Laurence W. Carter and Gary Bond (1996) Results on the Ground Five examples illustrate the magnitude and breadth of the changes affecting infrastructure services in many countries today: In 1996 a U.S.­based insurance company, Prudential Power, signed one of the first loans by an insurer to a private infrastructure project in a developing country. The sixteen-year, US$40 million loan was made under IFC's syndicated loan umbrella to an unrated, greenfield project: the US$1.4 billion Sual power plant in the Philippines. In August 1994 the government of Côte d'Ivoire signed a nineteen-year agreement to purchase power from the first private power project in Sub-Saharan Africa. Financing closed five months later. The 100 MW plant was generating power by April 1995, achieving 83 percent availability in its first six months against an 80 percent target. In 1993, IFC helped finance a power plant innovatively structured to reduce country risk: the generators in Guatemala's first private power plant were installed on a barge, which could be towed away in the event of nonpayment. The demonstration impact was rapid: by 1996 most other Central American countries had independent private power plants. In the 1993 privatization of a 30 percent stake in Hungary's telecoms company, the government stipulated that the network be expanded by 15 percent per annum, to reach 35 lines per 100 population by 2002. The company is investing rapidly to meet this target: nearly US$1 billion in 1995­96 alone. In December 1995, the government sold another 37 percent to the strategic investors for US$852 million. Infrastructure privatization partly explains why Hungary received over ten times the average foreign direct investment per capita for Central and Eastern Europe in 1990­94. The 1993 concession to manage water and sewerage services for six million people in Buenos Aires was awarded based on a 27 percent tariff reduction. Within two years 400,000 new water and 250,000 sewerage connections were made (many in low-income areas), drinking water quality improved, average repair response times fell from 180 hours to 48 hours and Buenos Aires experienced its first summer without water shortages in fifteen years. Ten years ago, none of these events might have been deemed remotely possible. Today more countries than ever are introducing competition and private participation in infrastructure (PPI). IFC has helped to finance many of these pioneering projects, approving US$3.1 billion of financing to 148 projects worth US$29 billion in forty countries by June 1996. This report follows a paper published in September 19941 and draws some updated lessons from IFC's evolving financing and advisory experience with private participation in infrastructure. The Upside Three positive trends underlie recent experience with private infrastructure financing in developing countries: More volume, countries, and sub-sectors. Total estimated financing of new private infrastructure projects in developing countries doubled between 1993 and 1995, from about US$17 billion to over US$35 billion. In the two years to June 1996, IFC financed projects for the first time in, among other countries, Bangladesh, China, Côte d'Ivoire, the Dominican Republic, Honduras, Jordan, Latvia, Panama, Tanzania, Uruguay and Vietnam. Sectoral firsts included an airport, a mass transit project and several toll roads. More privatization of existing assets. In 1994, the latest date for which figures are available, governments sold US$10.1 billion worth of infrastructure assets in seventy-five companies in thirty countries. New financing sources. Insurance companies are starting to finance PPI projects in developing countries. And official financing agencies, including export credit agencies, have set up programs to provide financing or guarantees to private infrastructure projects. Commercial bank financing has expanded in the last two to three years as more banks become familiar with PPI; IFC's expanding syndicated loan program is one indicator. Local financing is expanding in some countries, although slowly and from a low base. Some sizeable domestically owned infrastructure companies are beginning to emerge and invest substantial sums. Progress Is Uneven Much of this activity remains concentrated in a few countries and in the power and telecoms sectors. For example, in 1993 just nine countries (Argentina, Colombia, Hungary, India, Malaysia, Mexico, Pakistan, Philippines and Thailand) accounted for 99 percent of international private infrastructure loans. In 1995 the same countries, plus Indonesia and Turkey, accounted for 97 percent of the total. Only ten to fifteen countries have so far made significant progress in privatizing and market restructuring. More have started to liberalize infrastructure services but have made limited progress so far. And some countries are still just talking about PPI. Why? Managing private entry to infrastructure is a complex, politically charged process. Furthermore, political costs may occur immediately, while the economic benefits of infrastructure liberalization emerge with a lag. The uneven record of progress reflects the differing degrees to which governments have succeeded in overcoming these political hurdles. Private infrastructure projects differ from many other private investments in that governments are intimately involved, as regulators, buyers, or suppliers. Governments thus need to help shape the conditions so that transactions can close successfully. Many approaches to PPI are being used; these offer investors opportunities, but also pose risks. Some problems arise because all sides are relatively inexperienced. Governments may have unrealistic expectations, or mismanage the difficult process of awarding concessions. Delays in securing government commitments may prevent projects from achieving financial closure. Or, following successful financing, political pressures may lead governments to revoke concessions, fail to adjust tariffs, or restrict access to foreign exchange. Accordingly, risk management of PPI projects has both technical and political dimensions: even well-structured PPI projects may fail without government commitment. The importance of political commitment means that what is achievable varies enormously between countries and over time. Successful transactions help policies to evolve Political commitment differentiates the PPI leaders from other countries. But policy changes need to be translated into new investments and improved services if they are to garner broad support. IFC's experience suggests that well-structured PPI projects undertaken transparently create local constituencies for further liberalization. Many countries where early PPI projects were financed two to three years ago have since liberalized infrastructure services further. IFC's Recent Experience In the two years to June 1996, IFC's Board approved nearly US$1.5 billion of financing to sixty-four projects costing US$13.8 billion. Fifteen projects were approved in Argentina alone, reflecting the breadth of the country's infrastructure liberalization and privatization efforts. Other highlights include five power projects approved in Pakistan, reflecting a strong policy framework; a water and sewerage concession in Brazil; a Chilean railroad; regional basic telecoms networks in Bangladesh and Indonesia; cellular networks in Jordan and Bolivia; and ports in Panama, Vietnam, and China. PPI advisory work to governments on divestiture has expanded sharply. Between fiscal 1993 and 1995, IFC started nine PPI stand-alone advisory mandates; in fiscal 96 a further eight mandates were signed. The work has focused increasingly on the most politically sensitive transactions, where the value attributed to IFC's neutrality as part of the World Bank Group is highest. Recently signed mandates have included several in the water sector (Philippines, Gabon, and India), and countries where perceived risks are high (Haiti, Kenya, Ecuador, and Uganda). IFC has also provided more general policy advice, often in conjunction with the World Bank or the Foreign Investment Advisory Service (FIAS). FIAS has sponsored several round tables to facilitate dialogue between investors and policymakers on policy and regulatory obstacles to private infrastructure. In 1996 the round tables were held in central and eastern Europe and southern and eastern Africa. Experience in Low-Income and Risky Environments Low income Over 60 percent of IFC's PPI approvals have been in low or lower-middle income countries. Thirty-five projects costing US$6 billion have been financed in Côte d'Ivoire, Tanzania, Uganda, Zaire, Zimbabwe, Bangladesh, China, India, Nepal, Pakistan, Sri Lanka, Vietnam, and Honduras. Country risk Financing PPI projects is difficult-but nevertheless possible-in countries perceived by investors as risky. Over a quarter of IFC's approvals have been in countries with little access to international private capital, including Honduras, Latvia, Tanzania, Dominican Republic, Panama and Côte d'Ivoire. Investors have focused on projects offering short payback periods (cellular), projects which generate foreign exchange revenues (ports), or projects with strong sponsor and government support (for instance, some power plants). Project Performance By early 1996, forty-eight IFC-financed projects had completed construction and started operating. Although this sample is still small and new (especially in the context of the fifteen- to thirty-year agreements common in such projects), the evidence suggests that PPI projects yield operational efficiencies and good construction performance: Construction. On average the projects were 3 percent under budget and five months late (22 percent of the expected period). Although they are not directly comparable, a much larger sample of publicly financed infrastructure projects had cost overruns of 10-23 percent and time overruns of 54-68 percent. Good early operational performance. For example, several power projects have exceeded the performance targets in their contracts. Positive stakeholders. Successfully financed PPI projects have usually been followed by more infrastructure deregulation and privatization, suggesting that initial PPI projects are meeting consumer and government expectations. These "demonstration effects" are working across countries as well as between sectors. Unexpected events have (mostly) not been serious enough to derail projects. Although unexpected events have affected several projects, a combination of good project structuring, adjustments by investors, and flexibility by lenders and governments means that nearly all projects have remained operational. Nevertheless, concessions were disputed in two projects and the government is paying its utility's obligations in another project. The financial performance of IFC's PPI portfolio has been satisfactory. Achieving Financial Closure Financial closure is important because it allows investment to proceed and encourages further policy changes. IFC's experience suggests: More projects are closing, but closure times show a mixed picture. Although average closure periods fell from seventeen months in fiscal 94 to eleven months in fiscal 96, this hides wide disparities at the project level. Several high-profile projects have failed to close. Varying speeds. Closure has been faster for smaller projects, in countries with prior PPI experience or strong political commitment, in projects with experienced sponsors providing strong support, and in projects generating foreign exchange revenues. Delays have resulted from difficulties in resolving issues of risk allocation among project participants. This has taken longer where: investors, lenders, or government officials were inexperienced; political changes affected government's commitment; anti­PPI interests were stronger than anticipated; or noncreditworthy buyers or suppliers were involved. Improving the Policy Framework There is no single blueprint for improving infrastructure services. PPI options are defined by the extent of political commitment, the strength of opposing groups, and investor interest. Successful early transactions can create a positive dynamic of more investor interest, consumer support, and further government liberalization. For sustainable transactions to take place, the policy framework must meet the interests of governments (acting on behalf of the public), sponsors, and lenders. IFC's experience suggests that private entry is more likely to be successful and sustainable if governments use experienced and impartial advisors, and if transactions are transparent. Transparency involves: Clarity. Governments that adopt clear procedures for awarding and operating concessions are likely to get a strong investor response. Pakistan's policy framework for power generation projects is one example. Predictability. The government's role in implementing its commitments predictably is important. This might include, for example, undertaking tariff adjustments, meeting fuel supply commitments, or purchasing land. Competitiveness. Project fundamentals need to be able to withstand public scrutiny. This applies in the longer term, not simply at the time of the transaction. Equally, however, governments need to recognize that sponsors of early projects may demand a premium for the costs and risks of being first. While fully functioning regulatory frameworks and international competitive bidding are desirable, they may not always be possible, particularly in the early stages of promoting PPI. In certain circumstances-the first project in a country with higher risk, or a small project where there is urgency or limited knowledge of the market-directly negotiated projects may be appropriate. IFC has financed several such projects. Long-term agreements are bound to face unforeseen events and changing circumstances. Although it is too early to draw on experience yet, mechanisms such as dispute procedures in contracts or regulatory appeal procedures may be important in helping sponsors, governments, and lenders to maintain viable contracts over long periods. Financing Patterns Two financing structure parameters identified in the 1994 IFC infrastructure report remain broadly the same in this updated sample: An average debt-equity ratio of 58:42 suggests that substantial equity commitments are being made to attract debt, or lenders are ensuring that projects are not overleveraged. Projects where market risk is not a major issue, such as power generation projects backed by power purchase agreements, are generally more highly leveraged. About two-thirds of project costs are financed from foreign sources. Why isn't this ratio falling, given increased local financing in some countries? Simply, IFC continues to finance projects in risky countries, where access to domestic long-term finance is normally very limited and reliance is heavier on foreign financing. Managing Project Risks Efficient risk allocation and mitigation is central to bringing infrastructure projects to financial closure and to providing appropriate incentives during construction and operation. Efficient risk allocation occurs where risks are assumed by the party best able to manage them. Projects may still be financeable if some risks are not allocated according to this principle, but costs-and ultimately tariffs-will be higher. Sponsors and lenders expect higher rewards for assuming higher risks. Mobilizing debt is particularly sensitive to having adequate risk management mechanisms in place. With at-risk debt supplying over half of project costs, achieving financial closure can be stalled if the risk mitigation requirements of lenders are not met. Project financing has hitherto been a fairly specialized field. For PPI to continue spreading, more government officials, investors, and lenders will need to become familiar with the risk mitigation and management techniques used in successfully closed projects. IFC's approach to environmental risk management has evolved over the past two years to encompass greater transparency and consultation in environmental assessment. IFC's experience shows that many PPI sponsors-and their financiers-consider that sound environmental management makes good business sense. Conclusions and the Future Several conclusions have emerged from IFC's experience to date: The uneven pace of private infrastructure growth in developing countries is due to varying degrees of political commitment to liberalization in the sector. Soundly structured investment opportunities have attracted private financing. Private entry and competition in infrastructure are yielding significant gains in construction and operational efficiency. There is no single PPI method, although economic benefits generally increase with the volume of assets under private control and the extent of competition. The options depend on a country's creditworthiness, the extent of political commitment, and investor interest. Nevertheless transactions are more likely to achieve financial closure and be sustainable if they are transparent in the broadest sense: clear, predictable, and competitive. Well-structured PPI projects can be financed in countries with low income or high risk, or both. Some projects are easier to finance in such circumstances: smaller projects, those not facing severe market risk, those with strong sponsor and government support arrangements, and those earning foreign exchange. There is often a link between financed transactions and further policy changes. These demonstration effects apply within and among countries. Although it is early to conclude, most operational projects have not been jeopardized by serious noncommercial risks. There is some evidence of increased local financing, yet IFC's sample suggests that foreign financing continues to account for over half of project costs in developing countries. Several fundamental changes appear likely in infrastructure services, particularly in the countries that have liberalized and privatized the furthest. Many of the projects being financed today reflect the transitional state of partly reformed infrastructure sectors. They are likely to be quite different from the types of investments that will be occurring when reforms are complete. Three trends seem likely: Stronger competition within infrastructure markets. This has happened already in telecoms and transportation services and is quickly spreading to other sectors such as power. Competitive infrastructure markets have many forms, but there are probably some common features that they will share. First, there will be a clear separation of the roles of regulator and operator. Second, project sponsors will increasingly be those willing to take both implementation and market risk. Third, governments who expose investors to market risks will come under pressure to reform markets that supply such projects, such as state-owned fuel supply monopolies. And fourth, local investors are likely to play a larger role in investments than at present. More projects will be at the subnational level. If the privatization of transport, water, and waste sectors is to accelerate, the creditworthiness of municipal and provincial governments will need to be assured. Ultimately this requires the reform of local finances. In the meantime, there may be some scope to design credit enhancement and risk-sharing packages in the more attractive projects and regions. Domestic savings are likely to play a bigger role in financing private infrastructure projects. Developing domestic capabilities to manage, participate in, and finance private infrastructure projects is important to broaden the constituency of PPI, enlarge the pool of funding, and mitigate foreign exchange risk. In industrialized countries, and increasingly in the more mature reformed developing countries, one of the largest sources of financing for investment is the utility's own cash flow. But additional funding will have to come from the domestic capital markets. This will require a strong macroeconomic framework and a solid financial infrastructure, as well as attractive investment opportunities. 1. Financing Private Infrastructure Projects: Emerging Trends from IFC's Experience, IFC Discussion Paper No. 23 (1994).