noTE no. 51 ­ FEB. 2010 GRIDLINES Sharing knowledge, experiences, and innovations in public-private partnerships in infrastructure PPI in poor countries 56742 How to increase private participation in infrastructure management and investment James Leigland T o overcome huge shortfalls in access to In other words, non-IDA countries had 4.4 times infrastructure services, poor countries as much PPI-related investment per capita over need much higher investment levels and this period as IDA countries did. Moreover, in the more expertise to build, operate, and maintain wake of the global financial crisis, the role of PPI infrastructure facilities. The private sector is in poor countries appears to be diminishing rather one source for such resources, and projects than strengthening. involving private participation in infrastructure (PPI) have increasingly been used in developing PPI is nevertheless important in IDA countries countries. But PPI investment has been much compared with other sources of investment. In lower in poor countries than in better-off African IDA countries, for example, PPI accounts developing countries--and has been more for more infrastructure investment than any other affected by the global financial crisis. How source except government budgets--almost 78 can PPI projects play a larger role in improving percent more than ODA. PPI represents about 25 infrastructure service provision in these percent of the total in the region's IDA countries countries? and about 14 percent in its blend and non-IDA countries (with ODA, government budgets, and Infrastructure service levels in poor countries non-OECD financiers such as China accounting for remain extremely low, and the costs of services the rest). Globally, PPI investment had grown to high. Non-oil-producing low-income countries about 2 percent of GDP in IDA countries by 2007, in Africa, for example, would need to spend the almost twice the level in blend and non-IDA coun- equivalent of 22­36 percent of GDP annually for tries. The GDP share in IDA countries increased the next decade to reach even modest service level by nearly 100 percent in 2003­07, but the global targets like those proposed for water and sani- financial crisis has wiped out much of that gain. tation in the Millennium Development Goals. Today these countries are spending only 5­6 percent of GDP from public budgets and another What works--and what does not? 2­3 percent from external sources such as official development assistance (ODA) and the private Available data, recent research, and anecdotal sector (Foster and Briceņo-Garmendia 2010). evidence from practitioners point to the types of PPI projects demonstrating viability and sustain- ability--and producing significant efficiency In IDA countries private participation in infra- gains--in IDA countries since the late 1990s. structure (PPI) has helped improve services, especially in telecommunications.1 But the contri- bution is small compared with that in blend and Merchant telecommunications projects non-IDA countries (figure 1). The PPI Project More than 60 percent of PPI investment in IDA Database, maintained by PPIAF and the World countries in 1995­2008 was in telecommunica- Bank, indicates that in 1995­2008 total per capita PPI investment in IDA countries was about 64 percent of that in blend countries and only James Leigland is program leader for subnational technical about 23 percent of that in non-IDA countries. assistance at PPIAF. Helping to eliminate poverty and achieve sustainable development P U B L I C - P R I VAT E I N F R A S T R U C T U R E A D V I S O RY FA C I L I T Y through public-private partnerships in infrastructure 2 FIgure 1 In IDA countries, only a small contribution from private participation in infrastructure Infrastructure investment associated with PPI projects, 1995­2008 140 120 100 2008 US$ billions 80 60 40 20 0 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Non-IDA Blend IDA Source: World Bank and PPIAF, PPI Project Database. tions (compared with about 50 percent in blend watts)--because of the small size of national and non-IDA countries). Most of that investment markets and lack of regional project develop- came through merchant projects, which circum- ment but also because of poor project preparation vent many of the obstacles that limit the use of and noncompetitive procurement. As a result, in other kinds of PPI arrangements in IDA coun- Africa IPPs contribute smaller shares to overall tries. Merchant projects have no need for project PPI investment in many countries and are less preparation or procurement by government. They common than in some other regions, with only have robust revenue streams, even in poor coun- 23 IPPs operating in Africa's 39 IDA countries. tries, because they are often secured by prepay- Some types ment systems. They involve relatively low capital Port concessions costs, and most benefit from unregulated tariffs. PPI port projects appear to be viable in many IDA of PPI countries, although the best-practice PPI approach, projects bring Arguably, merchant projects are not even public- the landlord port model, is still rare. Private opera- private partnerships in the usual sense, because tion of ports has worked reasonably well in Ghana, greater gains they typically involve no public-private risk sharing Nigeria, Tanzania, and several other African IDA than others formalized in contractual agreements. But they may help strengthen popular support for more tradi- countries. Nigeria's massive port reform program led to more than two dozen concessions in this tional public-private partnerships in some countries sector, although many are being restructured. PPI by promoting the idea that private service provision port projects have been reasonably successful in offers efficiencies over government provision. They Indonesia and Vietnam, although they contribute may also ease the introduction of prepayment in relatively small shares of PPI-related investment. such sectors as water and electricity, where it has These projects appear to work well because ports sometimes met with customer resistance. generate hard currency revenues, tend to play a critical role in a country's transport network Independent power producers (making it difficult for governments to tolerate Independent power producers (IPPs) are relatively massive inefficiency), and have experienced strong common in IDA countries, typically accounting growth in container traffic since the early 1990s. for 30­35 percent of all PPI-related investment commitments in such countries as Indonesia and Toll road projects Vietnam, slightly less in Bangladesh, and close Toll road projects have proliferated in India (a to 50 percent in Pakistan. The IPP arrangements blend country), where highway projects made up often include power purchase agreements denom- more than half of all PPI transactions concluded inated in hard currency to protect operators from in 1990­2006 and accounted for more invest- foreign exchange risks. IPPs in African IDA coun- ment commitments in 2003­07 than any other tries tend to be small and relatively expensive sector except telecommunications. High traffic (Kenya's four IPPs are all less than 75 mega- volumes and methods for lowering private part- PPI in poor countries 3 ners' risks and costs (such as annuity concessions make systems efficient and commercially viable. and viability gap funding) appear to have helped Passenger services are often politically necessary spur the growth in India. Other countries, such as but require subsidization. Cross-border services, Vietnam, are considering grant support programs while necessary to create scale economies, often similar to India's. But in African IDA countries complicate the preparation of PPI arrangements. (except Nigeria) low traffic volumes and limited And overstaffing, often tolerated by governments as funding for risk and cost reduction have severely a form of employment generation, can make priva- limited the use of toll roads. tization of these systems costly and politically diffi- cult because of the need for massive retrenchments. A big Water projects increase in Across country income groups, PPI-related invest- ment is much lower in water and sanitation than Key factors in success PPI is needed in other sectors. Traditional brownfield concessions What factors appear to be linked to the success to close the in this sector, involving private responsibilities for distribution and high levels of investment, have of PPI projects in IDA countries? And how can financing gap become rare since the Asian financial crisis. But PPI donors and international financial institutions projects aimed at operational efficiency and service help ensure that these factors are in place? quality rather than private investment appear to be proliferating even in IDA countries (Marin 2009). Available project opportunities Such projects often involve investment, but it is Many IDA countries have few real opportuni- usually covered by government and donor funding ties for bankable PPI projects because revenue (as in the affermage contracts in West Africa and streams are generally too small to attract mean- some design-build-lease projects) or annual cash ingful private risk taking. Successful PPI programs flows (as in projects in Côte d'Ivoire and Mali). maximize demonstration effects by focusing on the few project opportunities with a strong chance Small-scale PPI for success--and leaving the rest for funding by Little systematic information is available on government budgets, ODA, or non-OECD finan- small-scale PPI. But studies estimate that in some ciers. Donors and international financial insti- developing countries, particularly African IDA tutions can help by avoiding preemption of countries, tens of thousands of small-scale private potentially bankable PPI projects by ODA or non- service providers are active in the water and elec- OECD financing or attempts to pursue such proj- tricity sectors. In Nairobi, for example, commu- ects where modest efficiency improvements offer nity-managed water schemes serve at least 60 easier opportunities for quick wins. percent of households in urban areas and even larger shares in slum and periurban areas. Most enabling environment operate like micro merchant projects, without A strong business climate is critical to the success government risk sharing or regulation. of PPI projects because it helps facilitate affordable project finance. India's investment-grade credit To make the most of these providers, the challenge rating, achieved while the country was still classi- is to legalize them, and thereby control costs and fied as low income, has played a big part in its PPI quality, and support skill development (to help success. Strong central government leadership in them understand, for example, that cost recovery easing restrictions on doing business--and showing is necessary for access to finance)--all without commitment to PPI--is critical in improving the bureaucratizing service delivery and dampening business climate. Donors and international finan- the entrepreneurial spirit of operators (Baker cial institutions can help by providing more 2009). While the development community long upstream capacity-building assistance to support hoped that reformed utilities would make these reforms in the enabling environment for PPI. small-scale service providers unnecessary, they are now beginning to be recognized as permanent Project preparation solutions in some countries. PPI projects in IDA countries commonly require upstream preparation--involving sector, policy, rail concessions and legal and regulatory reforms. This upstream Rail concessions have generated much smaller preparation is expensive, time consuming, and, shares of PPI investment in IDA countries than because it may not lead to deal closure, ultimately in blend or non-IDA countries. While common risky. Countries with access to adequate funding in Africa, they usually rely on donors or host for project preparation, from government budgets governments for capital investment. Rail conces- or from donors and international financial insti- sions face unique problems. In poor countries huge tutions, have better chances of success with PPI rehabilitation investments are often required to projects. But in Africa, where estimates suggest 4 that 10 percent of a project's total investment is Postscript: the financial crisis needed for project preparation (rather than the 3­5 percent normally cited for industrial econo- By late 2009 the global effects of the financial crisis mies), total donor funding available for PPI project were easing, and PPI investment was once again preparation is only a fraction of what is needed. on the increase in a few middle-income countries. But the recovery is taking place through larger projects in countries able to substitute budgetary PPI priorities and potential resources (or directed loans from national devel- opment banks) for commercial finance (Izaguirre How much PPI investment is needed to close the 2010). Pipelines of smaller PPI projects in poorer infrastructure financing gap in IDA countries? countries have not yet recovered and are suffering In Africa investment from all sources amounts from the higher cost of project finance and shorter to less than half of what is needed annually to debt maturities, which are expected to affect the meet modest service delivery targets after 10 global project finance market for some time. By the years (Foster and Briceņo-Garmendia 2010). PPI end of 2010 government funding support for PPI investment would have to increase by four to five projects is likely to be nearing exhaustion in middle- times to fill that gap. Greater operating efficiency income countries, increasing their need for what- and more official assistance will also be needed if ever commercial finance is available. Poor countries IDA countries are to have any hope of achieving will be further crowded out of the PPI markets. And minimally acceptable levels of service delivery in because of the procyclical nature of ODA, they are the next quarter century. likely to have less of that for infrastructure invest- ment over the next few years as well. With attention to the kinds of success factors mentioned above, a doubling of precrisis PPI invest- After several years of modest gains, IDA countries ment might be possible over the next 10 years. But will probably see PPI investment levels recede again. such an increase will add value only to the extent These countries need support now more than ever that it goes to the kinds of projects that can do if they are to make the best use of PPI in helping most toward meeting infrastructure service needs: to close the massive gaps in infrastructure finance. · Brownfield concessions in energy and transport, Note urgently needed to cope with rehabilitation 1. IDA countries are those eligible for concessional credits from needs stemming from chronic underfunding of the International Development Association (IDA) because of operations and maintenance, but difficult to do poverty (currently, with gross national income per capita of less than $1,135) and lack of creditworthiness. Blend countries are because the investment needs are so large those eligible for funding from both IDA and the World Bank's nonconcessional window, with eligibility defined on the basis of · Greenfield electricity generation projects (IPPs), middle- or low-income status, debt sustainability, and institutional quality. Non-IDA countries are ineligible for concessional funding. especially critical in Africa References · Management and lease (or affermage) contracts, Baker, Judy L., ed. 2009. Opportunities and Challenges for Small- which appear to help improve operating effi- Scale Private Service Providers in Electricity and Water Supply. ciency and service quality without requiring Washington, DC: PPIAF. risky capital investment by private partners Foster, Vivien, and Cecilia Briceņo-Garmendia, eds. 2010. Africa's Infrastructure: A Time for Transformation. Washington, DC: World · Small-scale PPI, already playing a major Bank. GRIDLINES part in infrastructure service delivery in many IDA countries Izaguirre, Ada Karina. 2010. "Assessment of the Impact of the Crisis on New PPI Projects--Update 5." PPI Project Database, World Bank and PPIAF, Washington, DC. Gridlines share emerging knowledge on public-private partnership and give an over- Marin, Philippe. 2009. Public-Private Partnerships for Urban Water view of a wide selection of projects from Utilities. Trends and Policy Options Series, no. 8. Washington, DC: various regions of the world. Past notes can be PPIAF. found at www.ppiaf.org/gridlines. Gridlines are a publication of PPIAF (Public-Private Infrastructure Advisory Facility), a multidonor technical assistance facility. Through technical assistance and knowledge dissemination PPIAF supports the efforts of policy makers, nongovernmental organizations, research institutions, and others in designing and implementing strategies to tap the full potential of private involve- ment in infrastructure. The views are those of the c/o The World Bank, 1818 H St., N.W., Washington, DC 20433, USA authors and do not necessarily reflect the views or PHone (+1) 202 458 5588 FAx (+1) 202 522 7466 the policy of PPIAF, the World Bank, or any other P U B L I C - P R I VAT E INFRASTRUCTURE ADVISORY FACILITY generAL eMAIL ppiaf@ppiaf.org WeB www.ppiaf.org affiliated organization.