XX\sAJ j - - = c< biD' S ti - .1 -~~~IL D E V E L O P M E N T IN PR ACT ICE Rural Energy and Development Rural Energy and Development Improving Energy Supplies for Two Billion People T H E W O R L D B A N K W A S H I N G T O N, D.C. © 1996 The International Bank for Reconstruction and Development / THE WORLD BANK All rights reserved Manufactured in the United States of America First printing September 1996 The Development in Practice series publishes reviews of the World Bank's activities in different regions and sectors. It lays particular emphasis on the progress that is being made and on the policies and practices that hold the most promise of success in the effort to reduce poverty in the developing world. This report is a study by the World Bank's staff, and the judgments made herein do not necessarily reflect the views of the Board of Executive Directors or of the governments they represent. Cover photos: Douglas Barnes, Curt Carnemark, Richard Spencer. 0-8213-3806-4 Contents F O R E W O R D I X A C K N O W L E D G M E N T S X I A B B R E V I A T I ONS AND AC R ON Y MS X I I Executive Sumnniarv I The New Agenda 5 The Role of the World Bank 8 1 Introduction 15 The Challenges Ahead 15 Past Responses 17 2 The Rural Energy Situation 20 Effects of Biofuel Use hy the Poor 2'0 Pollution and Health 20 Ecological Damage 22 Energy Efficiency 23 The Transition to Modem Fuels 24 3 Eme-ging Practices and Policies 28 Enabling People to Choose from among Alternative Forms of Energy 29 Avoiding Unnecessary Subsidies 29 Addressing Market Failures 31) High Starn-up Costs and Risks 32 Extenial Costs and Benefits 34 Emphasizing Participation and Institutional Development 34 Participation 34 Local Institutional Development 36 Decentralization 38 V VI R U R A L E N E R G Y A N D D E V E L O P M E N T Recognizing the Central Role of Good Enabling Conditions for Development 38 4 Options for Rural Electrification 41 Progress to Date 41 Pricing and Financial Policies 44 Cost-Effectiveness and the Choice of Alternatives 48 Costs of Grid Supplies 48 Reducing Initial Investment Costs by Using Appropriate Design Standards 49 Micro-Grids Supplied by Diesel Generators 49 Electricity Supplies from Renewable Energy Sources 49 Regulatory and Price Refoims, Unbundling. and Privatization 53 Implications for Rural Electrification 53 Approaches 56 5 Innovations in Renewable Energy 58 Technical Progress in Using the Solar Resource 58 Policies toward New Renewable Energy Sources in Rural Areas 62 Program Development 64 Prices 65 Credit 65 Taxes and Subsidies 65 6 Cooking Fuels: Toward More Sustainable Supply and Use 67 Improving End-Use Efficiency with Biomass Stoves 67 Improving Charcoal Efficiency 70 Developing More Sustainable Ways to Supply Biomass 71 Agro-Forestry and Farm Forestry 71 Participatory Approaches to Forest Management 74 Improving Access to Kerosene and Gas 74 Subsidies versus Price Liberalization 74 Distortionarv Effects of High Taxes on Cooking Fuels 77 7. The Role of the World Bank Group 79 Policies and Operations since the 1970s 79 Rural Electrification 81 The Sustainable Production and Use of Woodfuels 86 Renewable Energy 88 Project Innovations and Advisory Services 90 The Way Forward: A Renewed Commitmient by the World Bank Group 91 C O N T E N T S VIl Broadening the Scope of Energy Sector Reform 92 Investments 94 Opportunities for Partnerships 96 A P P E N D I X 1 0 2 World Bank Lending to Rural and Renewable Energy Projects. Fiscal 1980-95 102 A Rural Electrification Projects/Comilponents 102 B Renewable Energy Projects/Components 10)4 C Fuelwood-Related Projects/Components 105 Methodology for Estimation of World Bank Lending to Rural and Renewable Energy Projects 107 N O T E S 1 0 9 B I B L I O G R A P H Y 1 1 0 T A B L E S I Broadening Energy Access: An Action Plan for the World Bank I1 2.1 Indoor Air Pollution from Biomass Combustion in Developing Countries 22 2.2 Rural Energy Use Patterns in Developing Countries by End Uses 25 2.3 Current and Projected Use of Biomass by Region, Selected Years 26 4.1 Urban and Rural People Connected to Electnicity in Developing Countries, by Region . 1970 and 1990 42 4.2 Appliance Use in Households with Electricity in Urban Indonesia, 1987 42 4.3 Costs and Tariffs for Electricity in Rural Areas of Selected Countries, 1993 44 4.4 The Effects of Line Length and Consumption Levels on the Costs of Rural Electrification 50 4.5 The Costs of Small Diesel Supply Systems in Pakistan and Yemen 50 6.1 Returns to Agro-Forestry Practices in Six Central American Countries 73 7.1 Bank Projects and Programs That Have Increased Rural Energy Access 80 7.2 Broadening Energy Access: A Checklist for Bank Staff 99 F I G U R E S I Rural Electrification, Increases in the Number of People with and without Service, Selected Countries and Regions. 1970-90 2 2 The Use of Biomass in Relation to GNP per Person in Eighty Countries 3 3 Actual (1970-92) and Projected (1993-2015) Costs of PV Modules 7 1.1 Per Capita Consumption of Commercial Eneigy in Selected Regions and Countries, 1992 16 Vill R U R A L E N E R G Y A N D D E V E L O P M E N T 1.2 Rural Electrification. Increases in the Number of People with and without Service. Selected Countries and Regions. 1970-90 17 2.1 Energy Efficiency of Selected Cooking Fuels 24 2.2 Use ot Biomass by Rural and Urbain Populations in FOLuIr Developing Countries 26 3. 1 Costs of Grid Electrification in Relation to Load Density. Urban and RuLal Areas 32 3 2 The Use of Biomass in Relationi to GNP per Person in Eighty Counitries 40 4.1 Energy Efficiency and Lighting 43 5.1 Actual ( 1970-92) and Projected ( 1993-2015) Costs ot' PV Modules 611 7.1 World Bank Commitments for Rural Electrification, 1976-95 82 7.2 Lending Commitments for Sustainable Supply and Use of Woodfuels 87 B O X E S I Broadening the Scope of Energy Oper-atioris 9 2.1 Time. Health, and Cooking with Woodfuels 21 3.1 Who Benefits trom Electricity Subsidies? 31 3.2 Decentralized Power Meets Village's Needs for Clean Water 35 4.1 Rural Electrification in Thailand 45 4.2 C'redit Financinig Compensates for Fligh Cost of Access in Rural Bolivia 47 4.3 A Credit Proerarn for Micro-Hydro in Peru: A Promising Beginning 52 4.4 PV Market Growth Outpaces Cirid Connections in Rural Kenya 54 4.5 Institutional Reforin in the Electric Power Sector 55 5.1 Renewable Energy in China 59 5.2 Solar PV Home Systems in Rural Indoniesiai 63 6.1 Lessons from the World's Lareest Stove ProFrams 69 6.2 Woodfuel Markets Promote Tree Planting on the Island of Cebu 72 6.3 Rural Market Approach Anchored in Grassroots Reality: The Niger Household Energy Project 74 6.4 Household Fuelwood Use Virtually Eliminated in Hyderabad 76 6.5 Taxing the Rich Inadvertently Hurts the Poor 7S 7.1 Broadening the Scope of Energy Operations 93 7.2 StIh-Saharan Power Challenge: Promoting Electricity Access to Rural and Low-Income U!sers 95 7.3 The IFC and Renewable Energy 97 Foreword M aany people do not think about where their energy comes from. They just flick a switch or start their car. For nearly two billion people in developing countries, however. the search for energy is a daily grind. These people have neither electricity nor g,as nor oil to cook their food. Women and children from these families often spend hours each day gathering dung and wood-hours they might otherwise be able to spend on productive work or education. Their health is damaged by the smoke given off by these fuels. Their environmenit mayv be damaged as they chop down trees for wood. The problems of rural energy have long been recognized. What is the World Bank doing about them'? In recent years the Bank's work in energy has largely focused on making existing energy supply and consuminyg industries more efficient, openinlg them up to competition. and encouraging private sec- tor participation. This is an important job and is far from finished. In many developing countries. for example, electricity prices are heavily subsidized. These subsidies often benefit the wealthiest households the most. They also undermine the financial health of utilities. making it difficult for them to extend services to rural areas. But the Bank should not concentrate on reforimiing existing energy markets alone. Any reform will fail in the long run if it does not benefit the whole population. Improving rural energy should therefore be seen as an important goal in itself. The exciting message of this report is that, through a combination of better technology and decades of experience. we are now able to tackle the problems of rural energy better than ever before. The report draws on the experience of many experts in government, industry, and non- governmental organizations (NGOs). We thank them for sharing their knowl- edge with LIs. Tackling the problems of rural energy will force us to challenge some old assumptions. Those kings of the energy world-coal, oil, and gas. lx X R U R A L E N E R G Y A N D D E V E L O P M E N T for example-are not the only modern forms of energy suitable for rural areas. This report describes the growing attractions of renewable technologies, such as solar power. It argues that policymakers, governments. NGOs, and the pri- vate sector should concentrate on improving the use of traditional fuels. such as wood, and not just on promoting modern energy, such as electricity. It stress- es the importance of designing policies and projects with local people rather than imposing schemes from above. It shows, for example, how the environ- ment can be protected by giving farmers responsibility for managing forests. Rural energy presents the Bank with an important challenge. It should be a key part of our work, whether we are discussing country assis- tance strategies. energy sector reform, or new investments. While the Bank can play an important catalytic role, real progress in tackling these issues is critically linked to the adoption of the needed policy and institutional reforms. mobilization of local entrepreneurial and NGO resources, and development of innovative rural energy delivery and financing mechanisms. We therefore also plan to promote regional and country workshops to discuss these issues. encourage investment in rural energy, and foster policy reform. Such an effort has started already: this report itself reflects the results of widespread consul- tations between the Bank, donors, NGOs. and recipient governments that took place during its preparation. We will regularly report back to the Board and to the public on our progress on this vital task. Jean-FranCois Riscliard Vice President Flinance and Private Sector Development The World Bank Acknowledgments This report was prepared by Douglas Barnes, Dennis Anderson, and Karl Jechoutek, under the direction of Richard Stern. The report benefit- ed from the valuable assistance of Willem Floor and Robert van der Plas. Background papers were written for the preparation of the report by William Hyde, Gregory Amacher, Robert Wirtshafter. Deng Keyun, Zhang Zhengmin. and Russel DeLucua. Zhang Zhihong and Andrew Young also made helpful contributions to the project. The report also profited greatly from extensive discussions both within and outside of the Bank. Within the Bank the report received thought- ful and thorough reviews by more than fifty staff members, all of whom made valuable contributions to the project. The paper was also improved through a dialogue established with numerous individuals and groups outside of the Bank, and we wish to thank all those who met with us, including experts in government, industry. academic institutions, and nongovernmental organiza- tions. In this regard, we appreciate the efforts of Andrew Barnett and Daphne Wysham in organizing meetings with nongovernmental organizations repre- senting a wide range of organizations and countries. This paper was edited by Norna Adams, Alice Dowsett. Daniel Litvin, and Paul Wolman. We also wish to express our appreciation to Paul Wolman for coordinating the production of the document for publication. xi Abbreviations and Acronyms ASTAE Asia Alternative Energy Unit ESMAP Energy Assessment and Eneigy Sector Management Assistance Programme GEF Global Environment Facility GNP gross national product IFC International Finance Corporation ITDG Inteninediate Technology Development Group kW kilowatt kWh kilowatt hour LPG liquid petroleum gas NGO nongovernmental organization OED Operations Evaluation Department PV photovoltaic TA Technical Assistance WDR World Development Report xii Executive Summary It is hard not to be dauntedl by the scale of the problem of providing modem energy service to the world's rural population. Ther-e are nearly two billion people without access to modern forms of energy. such as electricity and oil. This report describes in detail the plight of these two billion. Its mes- sage, however, is that there are now many ways in which their situation can be improved. For though the problem is daunting. practical and affordable pre- scriptions are available. To uliderstand the possible solutions. consider first what is known about the problem. Aroulid a third of all energy consumliption in developing countries comes from burning wood. crop residues, and animal dung. Such biofuels are mostly used in rural areas, though wood is also Lised as a fuel by the urban poor. Biofuels produce the equivalent of twice the energy of the coal mined il China or the United States in a single year. They produce energy, however, in a seri- ously inefficient way. For a start they help trap the user in poverty. Gathering fuelwoocl and dung takes time-tine that could be devoted to more productive activities such as farming. A recenit study in the hill areas of Nepal showed that even in areas with fairly good supplies of wood, women need to spend over an hour a day collecting fuels. In areas where wood was more scarce. the chore lasted about 2.5 hours a day. Moreover. most biofuels rieed to be collected in large quanti- ties. They are a highly iniefficienit means of cooking compared with fuels- such as liquid petroleum gas (LPG)-used by wealthier- households. A kilo- gram of wood, for example, generates a mere tenth of the useful heat for cook- ing delivered by a kilogram of LPG. Biofuels canl also damage people's health, because they give off smoke that contains many hazardous chemicals. Studies of rural areas show that smoke levels inside dwelliigs often far exceed safe levels recommended by the World Health Organization. Cooking with biofuels does riot cause health problems everywhere. but in houses that have poor ventilation it can be as dangerous as smoking cigarettes. A stucly in The Gambia. for example. examiined the health of 500 childrei tinder- five years old. It found that childien who were calTied 1 2 R U R A L E N E R G Y A N D D E V E L O P M E N T on their mothers' backs as they cooked in smoky huts were six times more like- ly to develop acute respiratory illness than other children. Studies of women in Nepal and India exposed to biomass smoke-but who did not smoke them- selves-found that their death rate from chronic respiratory disease was simi- lar to that of male heavy smokers. The use of biofuels can also damage the environment. The search for fuel- wood often involves chopping down local trees. As trees disappear, fuelwood has to be sought further and further away. Using dung and crop residues as a fuel reduces the amount available for use as a fertilizer for growing crops. Such problems are avoidable. As the report shows. farmers in many areas use biofu- els in sustainable ways. But in many other areas the gathering of biofuels ranks together with logging. and the clearing of land for agriculture, as a cause of deforestation. In the northern Chinese county of Kezuo. for example, people have already cut down most of the trees around the farm lands. Poorer house- holds are now turning to even less efficient fuels such as straw and dung. Without electricity, moreover, poor households are denied a host of mod- ern services such as electric lighting and refrigeration. To an extent, some of these problems are being alleviated. Electricity supplies have been extended to over 1.3 billion people in developing countries over the last twenty-five years. Yet most of these connlections have been in urban areas. In many regions of the Figure 1 Rural Electrification, Increases in the Number of People with and without Service. Selected Countries and Regions, 1970-90 Millions of people 400 30() 200 1o0 0 North Africa Latin Sub-Saharan South China East Asia and the America Africa Asia and the Middle East Pacific 0 Incremental Popuilation Served Total Populationi Increase E X E C U T I V E S U M M A R Y 3 world, including sub-Saharan Africa and South Asia. populations have grown even faster (figure I). With the total population of developing countries expect- ed to grow by more than 3 billion in less than four decades, the problems of rural energy are likely to become more pressing than ever. So what can be done to solve them? The problems of rural energy should certainly not be dealt with in isola- tion. Poverty and dependence on biofuels go hand-in-hand. As household incomes rise, people normally switch to modem fuels, if these are available (figure 2). Higher-income countries also depend much less on biofuels than do poor countries. The best schemes for improving rural energy may therefore fail if other policies prevent economic growth. Figure 2 The Use of Bioniass in Relation to GNP per Person in Eighty Countries Percentage ot bioniass in total energy uised 6(1 so n~o E1 lH 4( O C1 13I 20 F-I F-I I I I . 0 I__ _ __ _ _ _ _ _ _ _ _ _(__ _ __ _ _ GNP per person (1988 0S$) Source: World Bank data Provided the background conditions are right, one of the most powerful ways to improve energy supplies is to ensure that the energy market is deter- mined by consumers' choices. In particular that means both that the price of energy should retlect its cost and that regulation of energy industries should encourage competition and choice. Govemnments should concentrate on ensuring that there is a level playing field for different investors in energy. whether they are public utilities, private finns. or enterprises set tip by the local community. 4 R U R A L E N E R G Y A N D D E V E L O P M E N T The opposite has been true in most developing countries. Rules and regu- lations have strangled the emeigenice of firmns other than the state-run utility. For example, it is illegal in many countries for local private or cooperative non-grid-connected generation and distribution enterprises to enter the mar- ket. Many governm11enit programs have attempted to extend energy supplies to rural areas. But too often the result has been unsustainiable public institutions promoting technologies that are unsuitable for rural consumers. Subsidies for electricity consumption are a particular problenm. In the early 1990s average electricity tariffs in developing countries were less than USą4 per kilowatt hour (kWh). even though the average cost of supply was around US I() per kWh. Such subsidies are harmful in a host of ways. They constitute a huge fiinancial draini (revenues from electricity supply in devel- oping countries fall short of costs by some US$ 100 billion every year). As a result utilities are often economically crippled, unable to finance the extension of services to rural areas. Where supplies have been extended to rural areas, subsidies often underiiiine the efforts of businesses to provide cheaper ways of generating electricity. In remote rural areas, for example. diesel engines or solar photovoltaic (PV) systems may provide electricity at a lower cost than grid supplies. But consumers will not opt for them if grid supplies are subsidized, nor will investors come forward to develop least- cost options to serve rural consumners. Overall subsidies on energy consumptioni tend to benefit rich people more than the poor. A recenit World Bank study of seven countries showed that high-incomiie households benefit disproportionately, largely because they use mo:lre electricity. In Malawi. for example. a pooI consumider onl average receives a mere US04 a year in electricity subsidies. while a rich one gets US$6.60. Some subsidies may be justified-but only if they are limited to specific and affordable goals. such as providing cheaper rates for very poor households for a fixed niaximiumn consumption per month (which can be achieved by charging wealthier households a little more than the cost of supply). Hefty subsidies for moderin cooking fuels. such as kerosene and LPG. are also commoni in a numilber of developing countries As with electricity. the results are often counterproduIctive. In Indoniesiai, for example, kerosene used for cookinig and lighting is subsidized. But riclher households reap a disproportionate share of the benefits because they can afford to buy more energy than the poor. The governm111ent of Ecuador also subsidized kerosene until recently. But the poor received little of the fuel because retailers could make more money selling it for use in vehicles. Even if subsidies reach the poor they often become unsustainable financial burdens on the state budget. Senegal's ann[ual subsidies for LPG. for example. rose from US$2 million E X E C U T I V E S U M M A R Y 5 to US$1() million between 199(0 and 1994-an increase that could have paid for thousands of desperately needed teachers. Market-opening reforms can be dramatically effective, as shown by the experience of Hyderabad in India. In 1980 only the richest 10 percent of households in Hyderabad used LPG. The proportion has since risen to over 60 percent. Meanwhile fewer households are using fuelwood even though the city's population has doubled since 1980. The main cause of the change was the liberalization of energy markets. In particular the Indian govern- ment relaxed restrictions on the production and import of LPG. As a result more middle-class households could buy LPG. a more efficient fuel than kerosene. That in turn allowed the poor to graduate up the fuel ladder from wood to kerosene. The New Agenda Liberalizing energy mar-kets. however important. may not be the complete answer. Despite the progress made in encouraging private investment in the electricity industry since the beginning of the 1990s, for example. private companies have shown little interest in extendinig electricity supplies to rural areas. They have instead plreterred to concentr-ate on more lucrative conitr-acts to generate electricity and to supply industrial and urban cLIs- tomers. There is evidence, in other words. that creating urban-based energy markets by itself will fail to provide rural electricity. There is one obstacle in particular that discoura-es companies from pro- viding supplies to rural areas: high start-up costs. Extending an electricity grid to a remote village can be very expensive, especially if only a few households are to be connected. Until more households join the network, the cost of electricity can reach USą70 per kWh. seven times the typical cost in an urbani area. Even setting up a solar electricity system for a single home can cost between US$500 and US$ 1I000. a large sum to spend in one lump. The problem here is not necessar-ily that people are unwilling to pay. Evidenice suggests that people will spen(d a significant proportion of their incomes on better energy, which impr-oves their quality of life or enables tihem to become more productive. In Bangladesh even the poorest people are connectin- to the grid when the service is available. In rural China. many people without easy access to cooking fuels are investing in efficient stoves and tree planting. The problem is that rural customer-s often cannot get affordable credit. That makes it difficult for them to pa)y the high start-up costs of improving their energy supplies. One solution may be to establish a local member-sup- ported bank to make small loans (such as the Grameen Bank in Bangladesh. 6 R U R A L E N E R G Y A N D D E V E L O P M E N T which lends mainly to women and poor people). Another is to promote com- panies that lease basic equipment to consumers, communities, and local energy suppliers (e.g., LPG distributors and small power companies). The importance of credit is illustrated by the experience of Mizque and Aiquile, two rural villages in Bolivia that are supplied by local diesel micro- grids. When the villagers were first given the opportunity to buy electricity from their local micro-grids. barely a quarter of them could afford the service. The electricity company that runs the micro-grids then decided to provide credit for the US$100 to US$125 connection charges. allowing customers to pay back the costs in small monthly installments over five years. As a result more than half of the villages' households were able to purchase electricity, even though the prices of local micro-grid power were fairly steep, at USą25 to USą30 per kWh. Suppliers can also expand their markets in rural areas simply by reducing start-up costs for rural consumers. A key aspect of the electrification scheme in Mizque and Aiquile, for example, was that supplies were limited to evenilg hours. Uniform standards and codes that prevail in many developing coun- tries and that have been designed for industrial and higher-income customers often prohibit such service adjustments. Technologies such as solar. wind, and small-scale hydropower are often ideal in rural areas and require far- more systematic attention by policymakers than they have hitherto received. Renewable technologies have significant environmental advantages relative to fossil fuels. S.anlight is also in ample sup- ply in many developing countries. Most important of all, the costs of many renewable technologies have come down significantly over the last decade. A combination of improved technology and economies of scale has pushed down the costs of wind power, solar thermal power (sunlight is used to heat air or water), and photovoltaic power (electricity is generated directly from sunlight). In the 1950s and 1960s. for example. PV cells were only used in the space industry. Their costs have fallen so dramatically since then (figure 3) that they are now used to power tens of thousands of homes in developing countries. Though electricity generated from PVs is still more expensive than electrici- ty generated from fossil fuels. PV modules can be easily installed in remote locations, circumventing the need for large investments in extending the elec- tricity grid. For this reason. PVs often make economic sense for rural house- holds or for water pumping. In recent years in Kenya, for example. 25.000 rural households have bought (unsubsidized) PV modules-more than have been connected to the electricity grid under the highly subsidized govemment rural electrification program. Still, both electrification strategies have so far reached only a tiny fraction of Kenya's rural population of 19 million. The effective exploitation of the new renewable technologies requires a "leveling of the playing field"-that is, eliminating tax, subsidy, and import E X E C U T I V E S U M M A R Y 7 Figure 3 Actual (1970-92) and Projected (1993-2015) Costs of PV Modules U.S. dollars per watt-peak 10,000- 100 c0 o0 r0 00oo0 0 000 0 ° Cost ranges required for 10 ° 0 commercial application: Small-scale 8 -° e o _ 4 applications 1 r Power generation (decentralized) at Power peak load generation I (decentralized) 0.1 I * 1 " 1 1 1 * 51 1 1 | - fl t at base load 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Note: The ranges of costs show the PV niodule costs iequired to coinipece with srnall-scale applications and with decentralized power generationi (assuming supply costs of (fSvS to tUSe l per kWh (at base load) anid USel6.5 per kWh (at peak load) The spread in the poiniii reflects the spuead in costs of different technologles Source. Ahmiied (1993). distortions that now discriminate against renewables in favor of fossil fuels; initiating national surveys of wind and solar resources, promoting credit schemes to help consumers meet the high first costs of such systems; devel- oping private supply and service infrastructures and associated training; and supporting selected demonstration projects. However quickly modem energy spreads in developing countries, dung, crop wastes, and wood will be used by tens of millions for decades to come. Encouraging the use of improved biomass stoves is one step in improving the efficiency and sustainability of the use of biofuels. Relatively simple and inexpensive improved stoves can reduce by as mtich as 30 percent the amount of fuel needed for cooking, thus cutting the time needed to gather wood. Moreover. because the improved stoves give off less smoke, they make the domestic environment less damaging to householders' health. Encouraging the use of improved stoves is easier said than done. A number of programs run by govemnments and donor agencies in the late 1970s and early 8 R U R A L E N E R G Y A N D D E V E L O P M E N T 1980s tailed, partly because they were oblivious to local customs and the local economic climate. But lessons have been learned-in particular that schemes should target those who would benefit most. For example, the Chinese National Improved Stove Prograrii (the world's largest stove scheme, which has installed 120 million stoves in rut-al households) wisely concentrated on areas with the greatest fuelwood shortages. In Tanzania a project funded by the International Development Association started up a self-sustaininig market that by 1992 had sold more than 60),000 improved stoves. It nio doubt helped that produIction and sales of these stoves were put in the hands of local artisans and shopkeepers. Fanm forestry and natural forest managemiient have long played an inportant role in alleviati wood pmpoblems in China, India, and many other countries. By providing market incentives for effective managetilemit of existing trees arid shrubs and limited planting of new ones, these practices significantly increase afforestation rates. Programs that support such market-driven approaches direct- ly address farmers' needs for fuel. fodder. mulch. medicine. arid wood. They also reduice pressure on agricultural residues and can improve soil fertility. Other ben- efits include curbing soil erosion, improving soil moisture retention, and supply- ing soil nutrients. Effective management of existing, forest resori.rces requires that local people take active responsibility and that forestry departmenits must change their role. Success depends on voluntary participation and decisionmaking as well as on fomial transfer of responsibility for forest managemLnt to the rural coniniunities. Perhaps the most important lesson learned by the World Bank and other organizations in recent years is that local inpiLt is vital to success. Local peo- ple trnderstand their situation and their problems better than outsiders. The village of Pura in south India is a prime example of this. Community biogas digesters. which produce gas from dung and farmn residues, have a h1istory of problems in India. Initial attempts to promote this technology failed in Pura too. The aim of the initial schemes was to encourage villages to use biogas as a cooking fuel, even though fuelwood is abundant in Pura. When finally asked, villagers explained they were more interested in getting clean water. So the biogas was instead used as a fuel to generate electricity. sonle of which was used in turni to power a deep tubewell pump. The result was a double gain for the villagers: clean water and electricity sLrpplies. The Role of the World Bank In the I 980s the finances of electricity utilities in a number of countries were deteriorating sharply. As a result the Bank turned toward encouraging countries to liberalize their energy markets. introduce transparent forms of regulations, and attract private investment. Two Bank policy papers published in 1993 set E X E C U T I V E S U M M A R Y 9 BOX 1 BROADENING THE SCOPE OF ENERGY OPERATIONS What the Board Approved in 1992: Bank Commitment lending Promotion of clean technologies Countries Sector reform 4- * Transparent regulation * Commercialization/corporatization * Private involvement * Importation of services * Market pricing * Demand management 4- * Added dimensions-a renewed comm11itment to * Extend modern energy supplies to unserved populations * Promote sustainable supply and( use of biofuels lintroduce new and renewable energy technologies by • Promoting commercial pricing and private involvenmenit in distribution - Providing incentives for extensioni of service * Supporting agroforestry and biofuel progrinis * Encouraginig local initiative anid open markets out this policy reform agenida. Sinice then. the Bank has been learning a great deal about how to improve rural energy development. At the same time the economics of renewable energy have also been changing rapidly. As a result the Bank is now pursuinig a new set of solutions to supplement the established ones of liberalizing energy markets (box I ). Sonie of the Bank's operationis in recent years have borne the stamp of this new knowledge. The sustainable management of forest resources has been 10 R U R A L E N E R G Y A N D D E V E L O P M E N T impr-oved by enipowerinig local populations to gain revenue from supplying wood to nearby fuel markets. Forestry assistance has led the way, followed by natural resource management and energy operations. Innovative delivery of renewable energy to consumers has been spear- headed by lending operations in India and Indonesia that focus on developing the ability of retailers and other intermediaries to reach rural markets. Technical assistance co-funded by bilateral donors is identifying policies and projects in many countries of Sub-Saharan Africa. Asia, and Latin America that pave the way for using new decentralized technologies to provide afford- able energy. However, this is just a beginning. The emerging strategies that promise to make a real difference for rural populations require a new and powerful com- mitment by the Bank and its partners. The Bank is committed to devoting sig- nificantly more attention and resources to the alleviation of the "energy pover- ty" of large portions of the developing world's rural populations. An Action Plan to achieve this is in place (table I ). The Bank's Action Plan (see table I) calls for a strong effort to accelerate the opening of rural energy markets, help consumers to have a choice, and put in place better systems to deliver and finance rural energy. The Bank intends to base its efforts to provide better access to energy for rural people on five main principles: * Provide for consunter choice. A better choice of affordable energy sources should be provided to rural consumers. Informed consumers will choose the most cost-effective solution. according to their preferences. * Ensure cost-reflective pricing. Distortions in prices that are created by subsidies and taxes should be eliminated. They create a disincentive for entrepreneurial solutions to rural energy supply, and give consumers the wrong signals. * Overcome thle high-first-cost barrier The obstacle of the high initial cost of obtaining energy needs to be removed. Credit mechanisms, lower-cost equipment. and lower service standards can all contribute to achieve this. * Encourage local participation. Participation of local communities. investors, and consumers in the design and delivery of energy services is essential. Decentralized approaches need to be part of the solution, includ- ing systematic local capacity building. * Implement goold sector policies. These are the basis for bringing better energy access to rural populations. Energy sector reform should include E X E C U T I V E S U M M A R Y 11 the opening up of the rural energy market. Macroeconomic policies should not discriminate against rmral energy. The role of the government should change from central planning to supporting markets. The task of improving energy supplies for two billion people remains daunt- ing. What has changed is that we now have solutions more firmly in our grasp. Table 1 Broadening Energy Access: An Action Plan for the World Bank 1. Develop Regional, Counltry, and Local Owner.shIip anzhd Commitment to Efforls to Broaden Energv Access The Bank should continue to conduct regional workshops, road shows. and train- ing seminars in Latin America. Africa. and Asia to initiate and reinforce planning for effective and affordable rural energy supply. For example. Bank activities should sustain and build on the regional dialogues such as the recent Johaninesburg power reforn symposium, which acidressed the requirements of expanding electricity supply and use in Sub-Saharan Africa. The institution should also earmark country-focused assistance as a follow-up to regional work- shops. as in the ongoing upstream work in Cameroon, Mali. and other countfies. On the local level. an example of productive assistance is reflected in a joint Chinese/ESMAP technical assistance project. which inclided training of county- level staff in economic analysis of energy projects. Finance technical assistanice and training projects to assist countries in developing policies. institutions, and programs designed to promote rural energy. Most of the Bank' tradlitional energy clients-notably the power and oil and gas companies- are ill-equlipped to puisue pro-rains involving provision of energy in rural and poor areas. Indeed, their current programs and policies often undermine rural energy markets. It is essential, then. that both the energy "establishments" and the new institutions involved in developing rural energy build the necessary capacity. 2. Srstematically "Operationalize " Rural Enetg! in Batnk Assistance Programs The Bank's energy sector assistance instrumilenits need to make a more consistent and systematic effort to address the problems of people's access to energy ser- vices. The course of normal Bank work provides many opportunities to promote eneroy access for rural and poor people. but at present those openings are not sys- tematically exploited. Bank assistance efforts can help promote wider energy access by helping client countries to "level the playing field" for existing and new energy service providers-for example, by setting import tariff levels on energy products at reasonable levels. Iimiting distorting taxes and subsidies for house- hold fuels, eliminating or regulating monopolies to allow wider entry into the business of rural energy supply, andl so oni. continued 12 R U R A L E N E R G Y A N D D E V E L O P M E N T Table I Broadening Energy Access contiumed In addition, "operationalizing' rural energy developmsent in the Bank's policy dia- logues and reviews conducted with client countries will help broaden the base of expertise within the Bank and will help the Bank arid its clients develop creative and effective approaches. For example, in India an ESMAP project involvinig renewable energy led to the development of a new type of renewable energy loan; in Kenya, policy dialogues on the distortioniary effects ot' import tariffs have helped to create a mor-e attractive environimenit for private photovoltaic (PV) iiia-keters. As they do for power sector- rei'or-ril. staff appraisal reports for all energy projects shoruld describe the country's and the Bank's strategy for rural energy develop- merit even it thie operation does riot include direct SUppOrt for rural enlrgy. It is not enoughi to assume that policies to address (tie problems of large energy indus- tries will indirectly benef'it rural and porr peoplc. The Bank shiourld help its clienits develop cross-sectoral strategies that take advan- tage ot the well-estikblished tiid(ling that cnergy projects are niost effective when coordinated witli efforts to promlote rirral developimienit through agriculturc. heatlti, and educationi projects. The importance of this approach is illrstrated by the Brazil rural developmenit pro.wCt, in which communiities are rsirig social developmenit funds to purchase PV and wind systemiis for lighting arid televisions at comiriiiiity centers. Local demilanid for these systemiis in turn has encouraged the developmenit in Brazil of conirinlercial distributioni networks for renewable eneiry products. To encourage coirUllities to see energy as part of the menIu of rural development choices. energy services and componienits should be typically includl- ed on the Bank's list of' approved uses of' social developmenit fuLids. along with schools, roads, health clinics, and so on. * Bank staff should pay more systenilatic attentioni to the energy access problems of people in rurral areas in country assistance strategies (CAS). In this context, the CAS is importanit for coordinatirig energy assistance with other sectors. 3. Promol' Bor.; Prac I'ce.s tiid I,,,ro. ioato in Prop'ie i De'sigr nm/rd Imrpler'mentition * The Bank can hell) to develop ininovative projects arid project componelits for rural eriergy. Efforts sliotrld be made to include conilporrents in energy projects withi the goal of incorporating the best practices irito the Bank's lending prograni. Examples are Niger and Chad Fuelwood Resotrrce Manageriierlt Projects. India Renewable Energy Project. arid Indonesia Solar Horiie Systemis. * Micro-finance for rural energy will be an important part of the strategy to ease access. This will inivolve working withi local financial institutionis and NGOs to establish rural energy as an eligible item for small-scale finialnciln. A link with the CGAP progranil will he established. * If access is really to be expanded to rural and poor- people. a consistent effort is needed to devise mearis to case or defer the all-importanit l'irst costs of energy colrrilrled E X E C U T I V E S U M M A R Y 13 Table I Broadening Energy Access continued access (e.g.. conilectiolis, appliances, and equipment . Ininovative experiments in financing, leasing. and delivering systemiis must be designed and piloted. The Bank should develop standard analytical tools for project appraisals of innovative rulal energy projects as well as ftor traditional rural electrification projects. 4. Disseminate Inn ovations and Be.st Practice Make available inlformationi on good praictice in delivering and financing rural energy-including leasing and credit schemes. eid rural electrification, micro- grid systems. renewable energy' systeiis. anid natural forest maiagemeent for fuel- wood production. Conduct research on impact of rural energy and energy policies on rurl developienit. Best practices can he seen in the Thailand Rural Electrification Program. which lowers grid elect6ification costs for rural people by using local donations of land and labor in the construction of electricity net- works, and by providing modest utility cross-subsidies that help keep the f'irst costs of coninection low. 5. Partnershipis w. itih Donors. NGO.x. and Othier Organication.s * The Bank should expand its work withi bilateral donors. NGOs and fou.ndations in providing technical assistance. This is an area in whichi the Bank can exell global leadership in direct support otf innovative and new projects and learn from best practices. - One way of' broadening the base of involvement and coordiniating efforls is to promote joint implementationi of'projects in collaboration with groups outside the Bank. including bilateral donors. NGOs, and foundations. The Bank has already undertaken a successful rural electrification pr-oject in Bangladesh with the help of the U.S. National Rural Electrification Cooperative Association (NRECA). New initiatives include the lauliching of a fiinancing vehicle for solar home svs- tems being conducted in cooperationl with a major founidation. T The Banlk ciain also provide technicial assistance to make fund(s available tor onlending to NGOs and voluntary and grassroots organizations that can catalyze investments in communities and organlize communLities to catry out eneigy-relat- ed programs. One sucIh program. proposed in the Haiti Forest and l'arks Protection Project, involves assistance for implementing an ilmproved charcoal stoves project through CARE. 6. Special Itnitiative oan Afiica * Sib-Saharani Africa has a particular nee(d to address the issues of serving dis- persed rulal eneigy dernand- The lack of populition density in rural areas requires innovative solutions for electricity and cooking fuels. * An initiative will designi a program of assistance to African clients that will aim to reforimi the power sectors. find solutiolns for improving energy access for the conlinued 14 R U R A L E N E R G Y A N D D E V E L O P M E N T Table I Broadening Energy Access conlinued 80 to 90 percent of the continent's unserved population, and facilitate develop- ment of local delivery and financing mechanisms. * Electricity delivery will be looked at in terms of designing ways to put in place decentralized. locally financed solutions (includinig renewables and solar home systems) and the relited reforms required in the existing power systems. * Access to cooking f'uels will be enhanced through programs emphasizing fuel- wood management, stove efficiency, charcoal efficiency, and transition to mod- ern fuels. * Assistance in policy reform will include opening of inarkets, elimilation of price/tax/tariff distortions, and facilitatinig entry of competitors. 7. Improve the Bank's Capacihv to Deal witJi Rurail Energy Issues * Training and retraining of staff to deal with rural energy issues will remain an important nee of the Bank. This can be accomplished through training courses, internal workshops on best practices, and opportunities to work in agencies deal- ing with rural energy issues. 8. Monitoring of Progres.s inA chieving Objectives The following activities will be important: * Preparing an aninutal status report that reviews the progress in implementing the action plan. This would include reviewinY the extent to whicih rural energy has been included in Bank assistance and the effectiveness of the instruments tised, such as promotion of structural reforms designed to open the rLral sector to new entrants, grid niral electrification. renewable energy, fuelwood, and so on. - Monitoring the investments and policy work in the areas identified in the paper as retlected in the CAS. - Assessing the recruitment and retraining activities for bringing Bank staff skills in line with goal of increasinig activity in rural energy project development. * Assessing the effectiveness of regional and country workshops. Note: CARE Cooperative for Americai Relief Everywhere CAS Country Asistance Strategy C(CAP Cornsltative Group to Assist the Poorest ESNIAP Energy Sector Managemient Assistiance Progranimne NGO Nongovernmenital Orgatizalions NRECA National Rurail Electrification Cooperatie Associatioin C H A P T E R O N E Introduction Policies on energy and the environment are overwhelmingly pre- occupied with the production and use of modern energy forms. Thus they focus on private versus public ownership, global warming. acid deposition, urban smogs, and so forth. While all these matters are extremely important, and the energy. industry. policymaking, and environmental communities, including the World Bank Group, pay a great deal of attention to them, anoth- er aspect of energy production and use gives rise to developmental and envi- ronmental problems of equal, if not greater, importance: half the populations of developing countries lack access to modem energy forms. Approximately 2 billion people depend on traditional fuels-animal dung, crop residues, wood. and charcoal-for cooking, and an equal number are without electricity.1 We also know that fuelwood and dung are inefficient energy sources-gas. for example. is fully ten times more efficient for cook- ing-and that their use leads to environmental damage through the stripping of forests and woodlands, and to respiratory diseases and premature death for millions of people through smoke inhalation. Dependence on such fuels is also an economic hardship. as in many parts of the world people spend excessive amounts of timte gathering and cooking with these fuels. While using traditional fuels more efficiently and in ways less damaging to the environment and to people's health is possible, this is not happening in large regions of the world, while the shift to modern fuels (which most people in developing countries would surely count as a ble.ing) is not progressing as rapidly as it could. The Challenges Ahead The total population of developing countries will grow by more than 3 bil- lion people in less than four decades and by 5 billion within the lifetimes of 15 16 R U R A L E N E R G Y A N D D E V E L O P M E N T children now in school. Thus energy policies will not only have to grapple with the task of meeting the needs of the 2 billion people curTently without service, but with new demands resulting from population g-rowth. The alter- native, a situation in which billions more people continue to lack access to modern energy forms and to depend on fuelwood and dung for cooking. would be socially iniquitous, environmenltally unsustainable, and a failure for development. In addition, developing countries need to meet the rapidly growing demands for modern or commercial ener-y from households and businesses that already have service. Their per capita consumilption of cojmmercial ener- yy is extremely low relaltive to that in the industrial countries. For example. per capita consumption of coinmercial energy in the United States is 8 tons of oil equivalenlt energy per year. or 80 times mIlorte thani in Africa, 40 times more than in South Asia. 15 timies more thani in East Asia. and 8 times more than in Latin America (fioure 1. 1). Electricity consumption is also much higher in industrial countries, about 13.000 kilowatt hours (kWh) per capita per year. compared with about 600 kWh per capita in developing countries. The financial anid manager-ial requir-ements of serving those lacking access, of keeping tIp with populaktion growth, and of meeting increased demand( on the part of those already served are (Iriving the current movement to liberalize the ener-gy industry in maniy countries. Figure 1.1 Per Capita Consumption of Commercial Energy in Selected Regions and Countries, 1992 Index: US = 100) I00 I n0 80 60) 44.6 45. 46.0 49.1 L 40 1.3 2_6 Sub- Soutil East Latini Germans Japan U.K. France LI.S. Saharan Asia Asia Amcrica Atrica .Sournct. World Banik ( 1994h. Slani ticail A [inexes). I N T R O D U C T I O N 17 Past Responses How might rural energy demands be met in ways that are sustainable? A good starting point for answering this question is to consider what has been done already and what can be learned from experience. Rural energy problems have long been recognized, and for more than two decades the electricity industry. forestry and agricultural agencies. development organizations, nonigovernimienital organizations. and many oth- ers have made innumerable efforts to address them. Few, if any. ideas and approaches have not been thought of or tried. India's interest in rural elec- trification, for instance. goes back three generations. In the past 25 years, developing countries have extended electricity supplies to more than 1.3 bil- lion people (500 million in rural areas. 800 million in urban areas), equiva- lent to almost twice the population of the Organization for Economic Cooperation and Development countries. Whatever the shortcomings of electricity supply industries in developing countries, this is an impressive achievement. and testifies to their commitment to expanding service. In Africa and South Asia, however, the efforts to expand service still lag behind population growth. which underscores (lie immensity of the task that still lies ahead (figure 1.2). Figure 1.2 Rural Electrification, Increases in the Number of People with and without Service, Selected Countries and Regions, 1970-90 Millions of people 2()0 _ _ __ _ E!li ,(0 _I_. North Africa Latin Sub-Saharan South Chuna East Asia and the America Afr-ica As~ia and the Middle Eas~t Pacific t ni.r.menital Population Served Total P'opulationi Increase 18 R UR AL ENE R G Y AND DE V EL O PM ENT Many countries have tried to provide incentives, some successful, others counterproductive. to improve the supplies of LPG and kerosene for cooking and lighting. As concerns biofuels and renewable energy, countries have experinented with numerous approaches: introducing woodstove programs, promoting agro-forestry and farm forestry (the former involving intercrop- ping of trees or shrubs and traditional crops and the latter involving cultiva- tion of trees alone as a crop), demonstrating sustainable ways to use local for- est resources, and encouraging the use of biogas from dung and crop residues. They have also attempted to introduce new ways of utilizing solar, wind, and hydroelectric energy resources on a small scale. Practically all these efforts were pioneering, and much can be learned from both the successes and the disappointments. The World Bank has also had significant experience with rural energy, and has supported rural energy programs for more than twenty years. Its policy paper on rural electrification was published in 1975 (World Bank 1975), and was followed by thirty-four projects in eighteen countries amounting to US$2 billion, plus several general electrification projects that had rural electrifica- tion components. In addition, the Bank financed more than forty fuelwood- related projects during 1980-95, often as components of agricultural and forestry development projects, and twenty-onie renewable energy projects. with a total commitment of more than US$4 billion. The joint United Nations Development Programme/World Bank Energy Assessment and Energy Sector Management Assistance programs, begun fifteen years ago, facilitated exten- sive policy discussions on the energy sector (including the rural energy sec- tor) and project preparation activities in nearly 120 countries. In addition to investment and policy advice to improve the rural energy situation directly, developing countries also need a financially sound and well-managed modern energy supply industry to meet the energy needs of the millions currently unserved. The 1980s, however, saw a marked deterioration of finances and managerial efficiency in the industry in numerous countries. often with deleterious consequences for their public revenues (given the fre- quent recourse to subsidies). This eventually culminated in the recent surge of interest in regulatory reform and the liberalization of energy markets around the world. The Bank's own policies on the subject were laid out in its policy papers on electric power and energy efficiency (World Bank 1993a,b), approved by the Bank's Board in 1992. and in a number of subsequent publi- cations and seminars. Important as efforts to reform and liberalize the energy sector are, they leave several questions unaddressed. How will fundamental shifts in policy affect the energy supply situation for the rural--and also the urban-poor?2 In all countries, rich and poor, public utilities have historically undertaken I N T R O D U C T I O N 19 almost all rural electrification programs. Is service provision likely to become more widespread in the new institutional and economic settings. as many believe it will, or will the low-income markets for electricity and other mod- em fuels, such as LPG and kerosene. be neglected without public leadership? Will services be affordable'? Who will be concerned about using traditional fuels in more sustainable and healthier ways'? Will the reforms help with the introduction of solar energy technologies? More generally. how will public policies on energy and the environment ensure that the world's most inipov- erished people are not left out'? The purposes of this paper, therefore, are to take stock of the efforts of the developing countries and of the support the development community has pro- vided. to see what we can learn from experience to date, and to identify ways by which the World Bank Group can help its member countries address the problems discussed. To this end, several drafts of this paper have been shared with and critiqued by many people and organizations that have been involved in developing policies and projects in this area. As the title implies, the paper is mostly about rural energy supply and use. However, the problems it discusses are not confined to rural areas. Millions of people in urban areas also lack access to modem energy forms and depend on supplies of wood. charcoal, and in some countries dried dung, for cook- ing-another manifestation of urban poverty and a source of damage to rural and urban environments, as well as to people's health. The policies discussed in the following chapters are often equally applicable to unserved urban pop- ulations, for whom the costs of service extension are much lower. CH APT ER T W O The Rural Energy Situation Approximately one-third of all energy consumption in developing countries derives from the burning of wood, crop residues, and animal dung (biofuels). By some estimates. it amoiiunts to around 1,000( million tons of oil equivalent energy per year, more than three times the energy of the coal mined in Europe in a single year and twice the energy of the coal mined in the United States or China. Most of this energy is used in rural areas, which account for about 60 percent of the population of the developing world, or up to 70 per- cent in the low-income economies. Consumption of fuelwood and charcoal in urban areas is also large in many countries, and results in deforestation and environimilental damage in the suiTounding countryside. with fuelwood even- tually havinig to be trucked over large distances. This is especially true in African countries. where the costs of distribution and of acquiring appliances often inhibit the use of gas and electricity. Effects of Biofuel Use by the Poor Aside from the economic hardship associated with gathering and cooking with biofuels. the indoor air pollution created by such fuels is a health hazard. particularly to women and children (see box 2I). In addition, collection of bioftiels frequently leads to ecological damage to forests, woodlands. and farmilands. and biofuels are generally energy-inefficient. Polilutio9, al(d Health As the 1992 World Development Report (WDR) noted (World Bank 1992). studies of smoke from the use of biofuels in rural areas (Smith 1987. 1988: 20 T HE R UR AL ENE R G Y SITU AT IO N 21 Smith and others 1993) have founid levels of solid particulate matter that reLg- ularly exceed the safe levels cited in World Health Organization guidelines by several orders of magnitude (table 2. 1). Cooking can expose women and chil- dren to such levels for several hours a day. and has serious health effects that have only recently been studied systematically, even though they are often just as senious as the effects of cigarette smoking. Carbon monoxidle emis- sions niay give rise to amiibienit concentrations that interfere with the body's norimial absorption of' oxygen. Estimates indicate that smoke contributes to acute respiratory infectionis that kill some 4 million infants and children a year. Recurrent episodes of such infections show up in adults as chronic bron- chitis and emphysema. which can eventually lead to heart failure. Studies in Nepal and In(dia of nonsmiiokinig women who are exposed to bioinass smoke have found abnornially high levels of chroniic respiratory disease. with inor- tality from this condition occurring at far earlier ages than in other populations and at rates comparable to those of male heavy smokers. BOX 2.1 TIME, HEALTH, AND dren, though the provision of COOKING WITH WOODFUELS improved woodstoves along with household education and extension When woodfuels are scarce, the time programs can help remedy this. A people spend collecting fuels reduces study of 500 children under five years the time they can devote to produc- of age in The Gambia found that girls tive agricultural activities. A recent who were carried on their mothers' survey in the hill areas of Nepal, for backs as they cooked in smoky huts example, found that even in regions had a risk of acute respiratory illness with relatively good supplies of fuel- six times that of other children. wood, women still need to spend Studies in Papua New Guinea and more than an hour a day collecting India show that nonsmoking women biomass. and the time they devoted who have cooked on biomass stoves to agriculture was correspondingly for many years exhibit a higher preva- less compared to people not depen- lence of chronic lung disease than dent on these fuels. In the more those who have had lower levels of deforested areas where fuels are exposure to cooking smoke. scarcer, the time and effort women In conclusion, the quality of life of expended were even greater, with women and children can be improved about 2.5 hours per day being spent by improving access to biomass and collecting fuelwood. fodder, and providing improved biomass stoves. grass. Surveys in Africa dating back to the 1970s have similar findings. Source: Kumar and Hotchkiss (1988); The use of biofuels also has an Smith (1987, 1991). See also Cleave adverse effect on the health of (1974) for surveys of people's use of women and children, especially chil- time in Africa. 22 R U R A L E N E R G Y A N D D E V E L O P M E N T Ecological Damnage The costs to the environnment of biofuel use in terms of increased deforestation, soil erosion, and reduced soil fertility have also attracted much attention. The consumption of fuelwood and dung is not the only cause of these problems- logging and clearing land for agriculture often cause greater damage-but it is, nevertheless, a source of environmental damage and cannot be ignored (see World Bank 1978, 1991). A study on Ethiopia (Newcombe 1984) found that where tree cover losses were severe, all the natural cycles through which nutri- ents were returned to an initially rich topsoil had been breached: first through the losses of trees themselves, and then through losses of grasses. crop residues. and dung when they were used for fuel instead of being used to fer- tilize the soil. Another consequence is that the soil retains less moisture. which results in reduced crop yields. Investigators have found that farms with good tree cover-where farmers have planted trees as windbreaks or shelterbelts. for example-have yields 20 to 50 percent higher, depending on local climates, terrain. and ecosystems, than those without good tree cover (Anderson 1987: Doolette and Magrath 1990: Gregersen. Draper. and Elz 1989; Spears 1986). Table 2.1 Indoor Air Pollution from Biomass Combustion in Developing Countries Concentration of .suspended Location and' parwticulate nuttter as nmultiple vear of study Measurement period of WHO peak guideline" China, 1987 Cooking II The Gambia, 1988 Average over full day 4-11 India. 1987-88 Cooking with - wood 75 (15-minute peak) - dung 90 (15-minute peak) - charcoal 25 ( 15-minute peak) Kenya 1987 Average over full day 5-8 1972 Overnight (space heating) 12-34 Nepal. 1986 Cooking 9-38 Papua New Guinea, 1975 Overnight (space heating) 1-39 Zimbabwe. 1990 Cooking (two hours) 6 Brazil. 1992 Stoves with flues < 0.4 Note: The studies are noit completely comparable because of different measuiremiienit milethods. a. The WHiO peak (98th percentile) guideline recomnitends that a concentration of 231) micrograms per ciubic meter not be surpassed by more than 2 percent (seven days) of a year. Source: Smith (1988). T HE R UR AL ENE R G Y SITU AT IO N 23 Recognition of the linkages between biomass use and the productivity- many would argue the sustainability-of agriculture has done much to revive interest in the once time-honored practice of agro-forestry. Development prac- titioners now understand that they should not view the "fuelwood problem" and its resolution in an isolated way, but as part of the larger problem of ener- gy supplies, poverty alleviation, and the protection of natural resources in rural areas (see chapter 6). More people now also recognize that the use of biomass need not be inherently negative. In principle, biomass fuels can be supplied-from forests, woodlands, or farmilands-and used in sustainable ways. One can cite several examples from Kenya and India of fanners turning to agro-forestry autonomously to respond to local demand for wood and to improve the local ecology. Biofuels are traded and are an ilmportant source of cash income for many of the world's rural people, and growing fuelwood can yield a compar- atively high return. In Africa alone, the production and marketing of biofuels represents a US$5 billion business that provides gainful employment to more than 400,000 people. Nevertheless. some regions in densely populated countries have already passed the point of sustainable production of biomass fuels. In areas of China and India. the combination of intensive use of land and rural population growth has already transformed many forests into farmland, leaving only mar- ginal lands to supply trees and shrubs. In the northern Chinese county of Kezuo, for example, people have already cut most of the trees around the agri- cultural lands and are now turning to less efficient fuels-straw and dung- while wealthier households are using coal. Deforestation has other undesir- able effects on energy production: in Yongchun County, Fujian Province, China, lands and forests had become so degraded by 1983 that siltation had reduced annual hydroelectric production from the 1960 level of 5.000 hours to only 2,200 hours. Energy EfficiencY Figure 2.1 presents data on the efficiency of various cooking fuels. Biofuels are generally much less efficient for cooking than modern fuels such as liquid petroleum gas (LPG) and kerosene. An exception is biogas. This is derived from digesters of dung and farm residues. and both China and India have done much to develop biogas and encourage its use among people in rural areas. The least efficient fuels are agricultural residues, leaves, and grass. With few exceptions, people use these fuels because they are available from the local environment at no cash cost, not because they value them as convenient cooking fuels (Agarwal 1983). Many parties have worked hard to try to raise 24 R U R A L E N E R G Y A N D D E V E L O P M E N T the efficiency with which biofuels are used by introducing improved stoves. often with positive results. Urban and some rural households generally pLirchase wood. charcoal. coal. and kerosene. These fuels have higher energy values per unit of weight than wood and are generally used in more efficient stoves. In addition, the level of heat output of kerosene stoves can be adjusted relatively easily. so kerosene is more convenient for preparino a wide variety of dishes. Households with the highest incomes use gaseous fuels such as LPG. LPG burns cleanly and effi- ciently. it is convenient, and it has an easily adjustable heat level. Figure 2.1 Energy Efficiency of Selected Cooking Fuels LPG Biogas Kerosene Etficient charcoala Charcoal Household coal Elficienit wooda Wood Cr'op residues /) 5 1( I 5 20 25 30 Megajoules of heat delivered to the cooking panl per kilogram of fuel Note: IThe valnes in this tiguire are derived irom a combination ot the tttels energy content and the etficien- cy wi1li1 which the fiels are typically burined for cloking in developing coullntries. a. Uised in an etlicieit slove. Soutne: Flo(orandivaniderPlas 1991].ESMAP(forthcotiing) The Transition to Modern Fuels As their incomes grow. rural people begin to use modern fuels more exten- sively. Table 2.2 sumimiarizes some typical changes in patterns of energy use by households and in agriculture and small industrial enterprises. The initial dependence on biofuels in the home eventually gives way to the use of elec- tricity for lighting and fossil fuels for cooking. In agriculture and industry. diesel engines and electricity replace manual and animal power for a vari- ety of purposes. Where rural electrification from the grid is not available or is too costly, diesel generators may be used instead. More recently, photo- voltaic systems have become an attractive option for small-scale electricity T HE R UR AL ENE R G Y SITU AT IO N 25 supplies for homes, businesses, refrigeration and lighting in health clinics, and water pumping. However, as figure 2.2 shows, the transition to modem fuels is likely to take some time. In the lower-incomiie developing countries, high percentages of rural people in particular, but also of the urban poor, continue to rely heav- ily on biomass.3 Even residenits of countries approaching the lower-middle incomiie range. such as Ecuador and Uruguay. still consurne large amounts of biomass energy. As regards the regional use of biomass, in Africa, about 85 percent of the total energy used is in the form of biomass, in South Asia this figure amounts to 60 percent, and in East Asia and Latin America to 25 to 30 percent. Table 2.3 provides estimates of the current shares of biomass in total energy con- sumption and of likely levels in 20()0 and 2010. The actual rate at which the transition to modern fuels will occur will depend on countries' economic per- formanice and development policies: the extent to which people cun-ently use Table 2.2 Rural Energy Use Patterns in Developing Countries by End Uses Household in(olne Etdl Else w) Me(ditut Hijh Household Cookine Wood, residues. Wood, residues, dung. Wood. kerosene. & dung kerosene & biogas bioyas, LPG. & coal Lighting Candles & kerosene Candles, kerosene. Kerosene. electricity. (sometimes mlolne) & gasoline & gasoline Space heating Wood, residues, & Wood. residues, & Wood. residues, dung. dung (often nonie) dung & coal Other appliances None Electricity & Electricity & storage cells storage cells Agriculture Tilling Hand Animal Animal. gasoline, diesel (tillers & tractors ) Irrigation Hand Animal Diesel & electricity Post-harvest Hand Anitiial Diesel & electricity processing Industry Milling Hand Hand & animiial Hand, animal, diesel. & mechanical & electricity Process heat Wood & residues Coal. charcoal. Coal, charcoal, wood. wood. & residues kerosene, & residues 26 R U R A L E N E R G Y A N D D E V E L O P M E N T biomass, coupled with population growth. suggests that a large number of people will depend on biofuels for many years to come. Even in East Asia and the Pacific, a region that has experienced significant econorric growth and a major increase in the use of commercial energy. biomass use still accounts for 33 percent of energy supplies and is expected to decline only by some 50 per- cent during the next fifteen to twenty-five years. Figure 2.2 Use of Biomass by Rural and Urban Populations in Four Developing Countries Percentage of population (00 80 60 40 20 0 Zalllbia Philippines Llrugpay Ecuador RLal Urban Note: Pe c-apita groshs naional product for the four counitries is a' follows iin 1993 L .5. dollars): Zambia - LS$2')( Philippines - L35S77(' Uruguay - US$3,341) Ecuador - ULS$;17(0 Source: Barnes and othe's ( 1994a.: World Banik ( 1993c. 19941b). Table 2.3 Current and Projected Use of Biomass by Region, Selected Years (percentage of total energy used) Region 1990 2000 20/10 Sub-Saharan Africa 85 83 8( Sooth Asia 60 52 43 East Asia and Pacific 33 26 20 North Africa and Middle East 27 2-3 19 Latin America attd Caribbean 26 22 19 sotoe: WVorld Bank estimaltes T HE R UR AL ENE R G Y SITU AT IO N 27 The conclusion must be, therefore, that energy policies will need to be as concerned about the supply and use of biofuels as they are about modern fuels, and so have two aspects, not one. They must create the necessary coIn- ditions so that modein fuels-primarily electricity from the grid, renewable energy sources, liquid fuels, and LPG-can be efficiently supplied to large populations that still lack them. Second. they must support ways to use bio- fuels more efficiently and sustainably. a task that is not confined to those working in the energy sector, but must also involve those working with agri- culture and forest management. C H A P T E R T H R E E Emerging Practices and Policies The transitioni to mooderin fuels and the sustainable provision andl use of biofuels will both depend on the quality of the enabling conditions for devel- opment. When human resource programs are in place and enabling conditions for investmenit are favorable. policies to extend energy services to those without them, whether rural or urban. are much more likely to succeed. Under such cir- cumstances. the markets for energy will be stronger, and suppliers of electiici- ty. photovoltaic (PV) sets. liquid petroleum gas (LPG). kerosene, woodstoves. biofuels, and so folth will have better financial returns both to investment and to labor. Such situations give rise to a "virtuous circle," because with increased rewards, efforts to expand services fLurther will grow. Thus development poli- cies can accomplish a great deal independenitly of policies that focus directly on the rural energy problem, though both types of policies are necessary. There are five general principles that experience has shown to be relevant to the provision of both modern and traditional fuels. namilely: * Enabling people to choose alternative forms of energy * Avoiding unnecessary subsidies * Addressing market failures * Emphasizinig participation and institutionlal development * Recognizing the cential role of enabling conditions. We consider these before examining the specifics of policies toward each type of energy. 28 E M E R G I N G P R A C T I C E S A N D P O L I C I E S 29 Enabling People to Choose from among Alternative Forms of Energy In economic terms. givinig, people choices and allowing t1eni to express pref- erences means liberalizing the pricing of and regulatory policies toward ener- gy supply and use. and allowing people to express their willingness to paTs for service. In the past. interventionist financial and regulatory policies restricted people's ability to inake choices and ultimately undermined the provision of services. Notable examples are the restrictions often imposed on electricity prices and, in some countries. on LPG and kerosene prices. By reducing ener- gy companies' reveniues. such restrictions often had the cotLnteiproductive effect of limiting their financial capacity to invest and extend services. Evidence froin rural areas suggests that people are willing to spend a sig- nificant portion of their incomes on higher quality energy services that improve their quality of life or enable them to become more productive. In Bangladesh. for example. when service is available, even the poorest people are coninectinig to the grid. In Kenya, where the possibility of timely connec- tion to gridl service is remote. higher-income rural households are buying solar PV systems. In rural Chiia, people without easy access to energy for cooking are investing in efficient stoves and in tree plantiig. In India, during the 1970s and 1980s people shifted from usinig animal power for irrigation to diesel or electric pumps. In some parts of rural Bolivia, those without access to elec- tricity use LPG lanterns as an alternative to candles. Experience shows that if services are available and reliable, rural people ar-e willing to use many different types of energy. anid that given a choice, they genierally select the combination appropriate for their daily demands. Avoiding Unnecessary Subsidies Man' governments have sought to promote the provision of affordable mod- erni energy services in rural areas by subsidizing particular foorilis of energy. Rural electrification is perhaps the main example. although governmenlts have sometimes offered subsidies for kerosene, LPG, solar energy, woodstove pro- grams, and so on. Although well intentioned, such policies have often proved to be counterproductive for the following reasons: They worked against consumers making least-cost choices, and in doing so also underminled investors' efforts to provide alternative energy forms. Commonl' cited examples are the choice between diesel enginies, solar PV systems. and grid-supplied electric motors for irrigation pumping and small-scale power suipplies. When load densities are high or connection 30 R U R A L E N E R G Y A N D D E V E L O P M E N T distances are short, the least-cost alternative is generally grid-supplied electricity, but subsidized investment programs of electric utilities have too often led to grid supplies being extended to small loads when diesel, solar energy, or in some areas wind energy, would have been cheaper. * Widespread recourse to unnecessary subsidies has frequently proved to be fiscally unsustainable. When coupled with price restrictions, subsidies eventually limiited energy companies' investmiient programs (see chapter 4). This has not only happened to some countries' rural electrit'ication programs, but sometimes to the suppliers of fuels such as kerosene and LPG for cooking. * Subsidies have discouraged efficient energy use. Subsidies have also, for the most part, gone to higher-income households (box 3.1). Subsidies lhave often gone to farms and commercial entelprises willing and able to pay for the full costs of service. For example, in Indonesia. subsi- dized kerosene for cooking and lighting is available to anyone who wants to buy it, but middle-class and better-off households, who can afford to buy more energy than the poor. reap a disproportionate share of the benefits. Until sev- eral years ago. the government of Ecuador subsidized kerosene, a fuel used predominanitly by the poor. Hlowever, the poorest households ended up receiv- ing only limited amounts, because retailers could make higher profits by divertings supplies to those using them for vehicles. Large subsidies have also been ineffective in reaching the poor through rural electrification programs, particularly when decreased cash flows undermine power companies' ability to maintain the quality of service and to extend service to remote areas. Addressing Market Failures Limited subsidies or tinanicial support can be justified for some programs in areas where education and training may be needed or where the costs and risks of start-up are high.4 No country so far has succeeded in providing uni- versal electricity service without some form of public support or cross-subsi- dies in the tariff structure. A typical approach that most industrial countries followed was for the government to mandate a regulatory requir-ement or to stipulate that a utility had to expand services in rural areas and towns, but could recover the costs through an overall increase in average electricity tar- iffs. As the costs of service are higher in rural areas, such policies led to some cross-subsidization from urban consumers, though without undermining the utilities' financial positioil. E M E R G I N G P R A C T I C E S A N D P O L I C I E S 31 BOX 3.1 WHO BENEFITS FROM A recent Bank study looked at the ELECTRICITY SUBSIDIES? distributional impact of electricity sub- sidies in seven countries. Electricity The shift from subsidized tariffs to subsidies are significant in all seven, economic pricing is urgent in several ranging from 0.2 percent of gross transitional and emerging market domestic product (GDP) in Malawi to economies. Many governments have 2.7 percent in Poland. The higher the historically set prices far below cost household income, the greater the recovery levels through subsidies subsidy realized, with the extent and cross-subsidies to serve social depending on the level of electricity and political objectives. Generally, used. However, as high-income however, the wealthiest benefit the households have more electrical most. The subsidies often amount to appliances, they benefit more than a significant portion of gross national poorer households. Thus a poor con- product (GNP) and lead to excessive sumer in Malawi receives US$0.04 a demand, higher investment require- year in electricity subsidies, while a ments, fiscal stresses, and inefficient rich one receives US$6.60. energy use. Source: Gutierrez (1995). Electricity Subsidies as a Percentage of GDP in Seven Countries Poland Zambia Ni geri a Kenv a Bolivia VeneZuela Mualawi O 1 2 3 Percentiage share of GDP Most countries also recognize the wisdom of supporting the development and use of new and innovative technologies, for example, by providing tax incentives, and sometimes investment grants, for solar energy technologies, which have considerable promise as commercial and environmentally attrac- tive sources of energy (see chapter 5). Public doubts about the capacity of unfettered marklets to meet the needs of the lowest income groups are often important in policies. The use of "lifeline" rates (that is. low tariffs) for low levels of electricity consumption is not an 32 R U R A L E N E R G Y A N D D E V E L O P M E N T uncomnnoni practice, and such rates are effective il properly targeted. Experience has shown that they can be implemented without undermining industries' finan- cial integrity (see the example of Thailand in chapter 4). Aside from the social justification of such polices as lifeline rates. are there any ground rules for the application of subsidies and other folms of public SUpport'? Experience and economic principles both suggest that they work best if they are targeted directly on-and are limited to-addressing market failures. In tihe case of rural energy provision and use in developing countries. two circumstances merit attention: high start-up costs and risks and external costs and benefits. High Start-uip Costs anid Risks Programs to extend electricity to rural areas and towns and to establish solar energy programs are perhaps the best-known exaimples of energy sector activ- ities that entail high start-up costs and risks. The costs of grid extension can be high initially because of the lumpiness of the investimients needed to reach rural areas and towns for the first time. Costs are often USą4() to USą70 or mor-e per kilowatt hour initially, but can declinie rapidly toward the levels found in urban area as load densities rise (see figure 3.1 ). Average costs dur- Figure 3.1 Costs of Grid Electrification in Relation to Load Density. Urban and Rural Areas U.S. centis per kilowatt hour 60 50 40 30 low Load Densitv High Load Density Rural Urban .Source: Statf calculations based oil pr(oject daLta (see also table 4.4 E M E R G I N G P R A C T I C E S A N D P O L I C I E S 33 ing a program's early years may greatly exceed the marginal costs of expand- min service. and some allowance for this in the tariff structure and financing arranigements may be justified to encourage use. The situation is similar in relation to the costs of connection and the pur- chase of appliances and of solar home systems. Setting up a community grid can cost tens of thousands of dollars. and the expense of connection to grid electricity is a pervasive problem, with initial connection costs being as much as US$1,000 in some cases (not including the additional costs of internal house wiring). A solar home system costs between US$500 and US$1.00() or more, depending on the system's configuration, imllport tariffs. and profit mar- gins. Stoves and cylinders necessary to burn LPG are also a significant invest- ment for the rural poor (see chapter 6). Another start-up cost for many pro- grams is that of providing education and training, which was an important fea- ture of Kenya's PV program, which is discussed more fully in box 4.4. The problems posed by start-up costs and risks arise partly from a real or perceived lack of creditworthiness of low-income consumers, and partly from a shortage of long-term credit. Although many researchers have found that rural savings (like people's willingness to pay for service) are often high- er than commonly thought (Von Pischke, Adams, and Donald 1983). the lack of affordable credit can be a barrier to meeting the initial costs of more mod- ern energy. However. emerging innovations in credit delivery systems, such as the Grameen Bank in Bangladesh (a member-supported bank mainly com- prising poor families and women), offer some promising approaches to eas- ing the credit situationl in rural markets (Khandkher, Khalily. and Khan 1995). Leasing arrangements for equipment such as PV sets, such as are being considered in India, are another possibility. While micro-financing and other programs have been directed imlainly at nonenergy uses, they seem well suited to financing energy investmenits. such as small-scale generating and distribution equipment and appliances. Electricity companies can also provide credit by including connection and service fees on consumilers' bills and spreading the costs over several years. The five-year installiiieiit paylenit for the coninection charge operated by the Bolivia public utility is a good example (see box 4.2). Similarly, and perhaps with the aid of a participator-y program, wholesalers or retailers of household PV systems, LPG, and improved appliances could, in principle, also become useful conduits for consumer credit. Nontovernmenital organi- zations, cooperatives, extension services, and other community organiizations could play a role in desigiiing and overseeing such programs. Private industries in the rich countries frequently absorb start-up costs themselves, and also extend credit to their customers in anticipation of coIll- mercial rewards later when markets develop. Examples of this abound in 34 R UR AL ENE R G Y AND DE V EL O PM ENT appliance markets. While the same may eventually occur in developing coun- tries as they liberalize their economic policies. it cannot be assumed that it will. Public policies to address the problems posed by high start-up costs are both necessary and desirable. E-xternal Costs and Beniefits Public support is also merited to address indoor and outdoor air pollution and the damage to natuial resources arising from the use of traditional fuels for cooking. Economists have long recognized that the most effective way to address environmental problems is to tax or regulate pollution. But taxing or regulating rural households in developing countries for cooking with wood and dung is not a practical proposition, and other policy instruments are more appropriate-for example, working through rural education and extension services to inform people about the fuels' health effects and initiating related efforts to encourage the use of more efficient woodstoves with flues. Not all externalities are negative, however, and in relation to energy sup- plies, rural (and urban) energy policies now need to take into account the pos- itive externalities of new renewable energy technologies. The costs of solar technologies and solar-derived technologies, such as wind, are declining steeply for several reasons: learning by doing, the opportunities for innovation provided by new investment, and scale economies in the manufacture and provision of support and marketing services. Given the considerable economic potential and environimlental advantages of these technologies, both for small- and( large-scale applications, the idea of subsidies to develop them further is fully consistent with the economic prin- ciples of public policy (Arrow 1962; Baumol 1995). The World Bank Group has also accepted this argumenit in relation to Global Environmient Facility projects (see GEF 1995. 1996. various years). Emphasizing Participation and Institutional Development As with other endeavors, development practitioners now recognize that par- ticipation by community organizations and local social units is crucial for the success of rural energy policies. To many community workers. scholars, vol- untary organizations, and others who pioneered participatory approaches. this has long been self-evident; their efforts are now beginning to bear fruit. Participation The World Bank Participation Source Book (World Bank 1995b) summarizes the main features of the participatory approach arid draws on case studies E M E R G I N G P R A C T I C E S A N D P O L I C I E S 35 from many countries to provide pointers toward good practices. Cooperatives, nongovernmental organizations, and local community orga- nizations can be highly effective vehicles for supportino the delivery of energy services of all kinds and for managing natural resources. Such grassroots orga- nizations are familiar with and understand local resources and needs. and are often willing to assume responsibility for implementing policies and projects. For example, in Karnataka State. India. residents of the village of Pura. supported by the Karnataka State Council for Science and Technology and the Centre for the Application of Science and Technology to Rural Areas, now administer household electricity and water generated by large community bio- gas digesters (see box 3.2). Subsequent chapters of this report present fulrther exainples of the participatory approach. Even the participatory approach, however. does not guarantee success; given its nature it must be tailored to the local situation and the task at hand. Experience has shown that Linless projects are related to social groups or actors interested in their outcome and capable of carrying them out, they are unlikely to succeed (see Cernea 1992). Several World Bank-financed com- BOX 3.2 DECENTRALIZED POWER ested in obtaining clean and reliable MEETS VILLAGE'S NEEDS FOR water supplies near their houses. CLEAN WATER Because grid electricity supplies were unreliable, the community decided to Near Bangalore, south India, the vil- establish a system of biogas produc- lage of Pura supplies household elec- tion for fueling a five-horsepower tricity and water through large com- diesel generator. Electricity from the munity biogas digesters. Although generator was supplied to house- community biogas digesters have a holds through a micro-grid, and also history of problems in India, the key to powered a deep tubewell pump. All Pura's success was in listening to households participating in the pro- what villagers viewed as their main gram received a tap with clean water problems and needs. in front of their houses, eliminating Original attempts to promote com- long walks to the local tank and sig- munity biogas systems in Pura failed nificantly improving their health. Each because they were directed at substi- household is charged a fixed rate for tuting biogas for wood as a cooking the water tap and each electricity fuel. Abundant wood resources in connection. Some households now Pura make fuelwood collection rela- have both a grid and a village con- tively easy, and therefore villagers had nection, which some villagers refer to no incentive to maintain the system. as "people's power." Subsequent discussions with vil- lagers revealed they were more inter- Source: Ramani. Reddy, and Islam (1995) 36 R U R A L E N E R G Y A N D D E V E L O P M E N T munity woodlot or community forestry projects, for example, proved to be disappointing for this reason (Ceniea 1992). Undertaken in the late 1970s and the 1980s. these projects were intended to reduce fuelwood shortages and pressures on local forests and woodlanids by involving communities in plant- ing local woodlots. However, even though they were based on local inquiry and sought to have high levels of local participation, planting rates and wood- lot maintenance often fell below expectations. The underlying mistake was to view communities as units of social organization or economic agents, when in actuality, the interests of community subgroups frequently differed and worked against collective actions; community land was limited and the tenure of common lands was uncertain; the local systems of authority had uneven influence over subgroups' decisions; the distributional arrangements for shar- ing the products of the common endeavor were disputed; and. more general- ly. the communiities were -eographiical residenitial groupings, were not orga- nized as joint producers, and were not systems onto wliichi external econom- ic activities could be grafted (Cernea 1989. 1992). In contrast. when project designer-s identify the appropriate social unit. the chances of success are improved enormously. An excellent example is afforestation based on farm- or agro-forestry. in which the farmers themselves plant trees on farmilands in association with agriculture. Such projects began to receive more support in the 1980s. and have often exceeded the project designers' expectationis, largely because *as an enduring social unit able to sustain forestry development, the far-m family is a powerful social resour-ce" (Cernea 1989, p. 53). Participation in the development of woodstove pro- grams by women's grouLIps is another successful example. Local Inistitutionatl Development Community action may however be effective when people from several vil- lages and towns in a particular region are seeking electricity services. To this end, various approaches are possible, namely: * Establishing an electricity cooperative, a model followed in parts of the United States and encouraged in the 1960s and 1970s in several Latin American countries. Bangladesh's rural electrificationi program has recently used cooperatives and has achieved considerable success in terms of the number of connections achieved. cost recovery (close to 100 percent), and service quality. Sending deputations from the communities to the regional or national electricity utility to request service. perhaps in conjunction with offers of community self-help to reduce costs. E M E R G I N G P R A C T I C E S A N D P O L I C I E S 37 Establishing village-level or community ownership of a local system where grid supplies are not available (box 3.2) Historically, however, providers have used other imeans of making elec- tricity available to villages and towns. In the first few decades of this century. it was often supplied by a local enterprise willine to invest in a small genera- tor and establish a small, local grid to retail the electricity service or by an agro-industrial plant thi had surplus power available. In other cases, local governments established public companies to provide electricity service to their communities. either using their own genlerators or medium-voltage liniks to a grid. Joint public-private partnerships were another arrangement com- monly followed in both industrial arid developing countries. As villages and towns expanded following migration fromii rural areas and economic growth took place, the area's public or private comilpaniies also grew, and were the pre- decessors of the larger mullicipal-and sometimes regional-electricity com- panies we still see in some regions. Inteiconnection then followed, initially to pool reserve capacity. and later- to reduce costs further through trading power and energy and through the scale economies that could be obtained by increasing the capacity of power stations. All this took place. if not always in a competitive environmenit, at least in a situation where good financial returns to investimient could be, and generally were. being achieved. In developing countries, public ownership succeeded local private or quasi-private ownership of electricity distribution in the 1950s and 1960s, and took place somewhat earlier in most industrial countries. Since then. the prin- cipal source of service expansion has been the direct extension of supplies from a national electricity authority under public ownership and regulation. With the privatization and unbundling of the electricity industry in many countries, the question arises as to whether these historical models will reap- pear or, where countries clo not pursue full privatizationl. ownership will be in the hands of municipalities that operate on commercial principles. This ques- tion is difficult to answer, because no country has followed this path before. and it raises new issues, not only about ownership but about risks. taxes, reg- ulation, and the roles of local or regional governments (see chapter 4). A monopoly element would still be present in any public or private involvement in distribtLtion. though the companies would be more visible than their nationialized predecessors, and thus more accountable locally. However, this scenario assumes that companies are willing to invest extensively in extending distribution networks to villages and towns. Yet despite the progress made with institutionial reform in the electricity inIdtlStrV since the beginning of the 1 990s. private companies have so far shown comparatively little interest in distribution and much more in generation1. Will local or regional govemments have the capacity to attract private investment in distribution on the scale 38 R U R A L E N E R G Y A N D D E V E L O P M E N T required? We are at the beginning of a new and complex process of institu- tional development. the path of which is not yet clear. In the meantime, service expansion needs to proceed. Chapter 4 pursues the discussion of options. Even when well-tailored local energy approaches are pursued by cooper- atives or municipalities. they run the danger of failing because of inadequate training of personnel and inadequate maintenance and repairs of systems. Local diesel generators and micro-hydro schemes have failed for such rea- sons. and, as chapter 5 points out, so have some PV systems, even though they are inherently low-maintenance operations. A solution often suggested is for communities or cooperatives to share the needed infrastructure of services, a practice that can be facilitated by regional or central leadership. Decentraalization In a broad sense, participation by local people in choosing and designing local investments is moving countries toward decentralized economic develop- ment. Thity years ago a Nobel Laureate, the late Sir Arthur Lewis, drew attention to the importance both of decentralizing financing and of building Up local governmenit units. He also noted that public services (schools. water and electricity supply systems, local roads, and health centers and hospitals). which in mature economies are rin by local authorities. were in many coun- tries still being run by central governments and financed from central rev- enues, a situation that persists in most developing regions (Lewis 1969). Local participation and the development of local institutions in this broader sense is desirable fiom both the demand- and supply-side perspectives. On the demand side, it helps to reveal people's preferences among different services and may also reveal other things. For example, many studies have found that people's willinigness to pay for electricity and other services is far higher than had been assumed (Whittington, Lauria, and Mu 1991; World Bank 1992). Similarly, as noted earlier, rural savings are also often greater than assumed, which makes the problem of people raising enoLIgh moniey for initial connection somewhat less severe. On the supply side: decentralization increases local accountability and provides incentives for more efficient provision of services. In sum, then, the general aims of seeking greater local participation in rural energy programs are to ensure that such programs are far more respon- sive to people's needs than they often have been in the past. Recognizing the Central Role of Good Enabling Conditions for Development The dependenice of so many people who lack access to modern energy forlmls on biofuiels is above all a manifestation of poverty, and unless development E M E R G I N G P R A C T I C E S A N D P O L I C I E S 39 policies in general promote economic growth and development on a broad basis, not only will the rural energy problems described in this report persist. they will become worse as developing countries' populations grow. Even well-conceived investments in rural energy may falter, not because they are intrinsically wrong, but because economic conditions may be work- ing against them. For example, in rapidly dleveloping agricultural regions. the provision of electricity helps to raise the productivity of local agro-industrial and commercial activities by supplying motive power, refrigeration, lighting, and process heating. In turn, increased earnings from agriculture and local industry and commerce raise households' demands for electricity. However, when development efforts fail because of. say. poor crop pricing and market- ing policies, electricity supplies may be able to do little to remedy the situa- tion. nor will electricity or other modern fuels be in great demand. If it is to serve a useful purpose, electricity needs a "market." as do other energy forms such as l PG and renewables. Thus it is no coincidence that investments to improve energy supplies in rural areas are more successful whien development is proceeding and incomes are rising. The higher a country's per capita income, the greater the use of modein fuels (figure 3.2). When per capita annual incomes are US$300 or less, 90 percent or more of the population typically depend on fuelwood and dung for cookiig. Once annual per capita incomiles reach US$1,000 to US$1,500. substitution to modern fuels is almost complete. With techinological progress and reductions in the costs of supplying modern f'uels, the income levels at which the transition takes place have been declining for more than a century. This can be illustrated by comparing the experiences of the United States and the Republic of Korea. In the United States, the transition from being 90 percent dependent on woodfuels to elim- inating them almost entirely took nearly seventy years, from 1850 to 1920). In Korea, the same transition took about thirty years. and wats almost com- pleted by 1980. For each 1.0 percent increase in per capita incomes. roughly 0.6 percent of the population turn to inodern fuels. Based on projections of per capita income in Global Economic Prospects and the Developing Countries (World Bank 1995c), the population served by modern fuels cotIld rise fiom around 2 billion in 1990 to more than 4 billion by aroulid 2010. and if' this happens, the number dependent on traditional fuels would decline by 0.5 billion or mole during the period. The biggest gains would be in Asia and Latin America, but there is no reason why Africa could not make good progress, given a good enabling environmenit. Looking at the relationship between per capita incomes and fuel use in this way is, of course, a simplification. When development policies are such that 40 R UR AL ENE R G Y AND DE V EL O PM ENT incomes in a region are growing. both public and private investments in ener- gy supply, in natural resouice management, in soil erosion control, and so forth are all more likely to "work" better and show good returns. When assess- ing the prospects for policies and investmenits focused directly on improving energy supplies, therefore, one should also ask whether local health and edu- cation programs are being put in place; whether complementary infrastruc- ture, such as rural roads, water supplies, and sanitation is being given proper attention; and whether macroeconomic and pricing policies are such that the investments will function and will serve useful purposes. Rural energy poli- cies and investments are an adjunct to good general development policies, and their chances of success are greatly increased when the answers to such ques- tions are positive. Figure 3.2 The Use of Biomass in Relation to GNP per Person in Eighty Countries Percentage ot biorinass itl tot;l Qnergy used 1()() ________________ E:: v ql [D 60} CT F-1F- ElF 40 2( ~O [2 [ 2 iF O O - 0 ~~~1,0()0 1(10()() loo.>)n GNP pei- pers,on 1 1988 115$) SourIZLE Wnr,ld Bank data. C H A P T E R F O U R Options for Rural Electrification Rural electrification is generally thought of as grid electrification. and this is indeed the most comimlon and most desired means of supply. Where load densities are low, however, diesel generators. renewable energy (solar energy. micro-hydro. wind, and small biomaass-fired generators), and "hybrids" of such options are miiore cost-effective. Progress to Date Developing countries' efforts to extend electricity supplies to their popula- tions have been impressive. During 1970-90. nearly 1.3 billion people were newly suipplied with electricitv from national grids, of whom 800 million were in urban areas and 500 million in rural areas (table 4.1). In all regions the share of the population served rose. China accounted for nearly half of the increases, but the extent of service in Africa remains low. DIurinig this period, populations grew by 1.5 hillion, and in low-income economies outstripped the growth of people with service: in Sub-Saharan Africa, the nuimber of people with service grew by 55 million, total popula- tion grew by 220( million; in South Asia. the number of people with service grew by 250( millioni, total population grew by 320 million. Low-income households in developing countries use electricity mainly for lighting, television, radio, and ironing; as incomes rise. refrigerators and other appliances become affordable. Table 4.2 provides data from a survey of elec- tricitv use in urban households in Indonesia. where the pattern was similar to that found in rural households in other counitries. Unless one CouInts fans, whicil are quite popular, air conditioning. which is one of the biggest sources 41 42 R U R A L E N E R G Y A N D D E V E L O P M E N T Table 4.1 Urban and Rural People Connected to Electricity in Developing Countries, by Region, 1970 and 1990 (percent) Re,/ian R19 al Region 197z) 1990 /970 1990 Northi Africa and Middle East 65 Xl 14 35 Latin America and Caribbean 67 82 15 40 Sub-Saharan Africa 28 38 4 8 South Asia 39 53 12 25 East Asia and Pacitfic 51 82 25 45 All developig coauntries 52 76 18 33 Total served (in millions) 320 1,100 340 820 Nole: The'e esaitiiie, aire aonIv approximation' Smiurce World B ink project anid sector reprls. other Imatenils. aInd ILnrveVS 'If eleCi1ici ty slatis ics hy the Wo,rld Bank's regional siaff in Asia and Latiii Aitterica. Table 4.2 Appliance Use in Households with Electricity in Urban Indonesia, 1987 (percent) In_ icme cle.s.s AuIinirv 1,o Middle High Lighltinig 1(0 10( 10( Televisioni 31 63 83 Ironingi 21 5 1 77 Economilic activity 5 6 9 Refrigeration 1 6 9 Water pumping 1 4 26 Air conditioniing 0 I 1 Cooking 0 2 8 kWhlhousehold/milonth 24 47 130 Source: Peskin and Bazrnes (1994). This paper was a background paper fIo lie World Developrnent Report 1994 (World Bank 1994bl. O PT IO NS F OR R UR AL EL EC TR IF ICA T IO N 43 Figure 4.1 Energy Efficiency and Lighting Fluiorescenit l 11 40 Electric incanidesceni 12 Kerosene mantle 0. Carndlc~ .1 Kerosene wick I (J 5 Itt IS 201 25 3(0 35 4 Kiloluimen hours/kilowatt hour Swinrr e: Yarn der Pls .and de Graaftf 1988). of electricity demand in the rich countries, is virtLially absent. Electricity is also rarely used as a cookitig fuel, for which it is both expensive and energy inefficient. One sometimnes overlooked feature of electrification programs is that they may improve the efficiency with which energy is produced and used. A par- ticularly dramatic case is that of electric lighting, in which even the incan- descent lamp represents a 50- to 100-fold increase in energy efficienicy rela- tive to the kerosenie wick lamp (figure 4.1 ). The fllorescent lamp, which is well suited to solar home systemns. raises energy efficiency several hundred times. The quality of the light electric incanidescent or fluorescent bulbs pro- vide is also vastly superior to that of kerosene lights or candles. Although electricity fromil central grids, local grids, or renewables can be expensive in absolute terms, the combination of higher efficiency and better quality service is attractive to those rural consumers able to afford it, and may also reduce the real costs of the service (for example. li-ht) itself. On the island of Java, Indlonesia. for example, the use of electricity compared with kerosetie led to a seven- to tenfold incr-ease in lighting for newly con- nected consumers, the cost of electricity per kilolumen of light oLItput being about 5 percent of the costs per kilolumen of light provided by kerosene lamps. In tiost countries, the biggest sources of demand for electricity in rural areas and towns are often farms, agro-indt.stries. and small commercial and manufacturing establishmenlts (for iTigalti on pumilping. water supplies. crop processiny. refrigerationi, and inotive power). Typically 60 to 80 per- cent of the electricity supplied in developing agricultural regions is used for such purposes. Thus rural areas and towns often use electricity for purposes that are socially and econoicallv valuable. They do so often enough to justify the 44 R U R A L E N E R G Y A N D D E V E L O P M E N T investmilent from an econoinic per-spective. The main qLiestions about grid electrification in rural areas and towns relate less to its usefuliess, than to financing and costs and whethier alterinatives are mior-e cost-effective. As noted in the introduction. current moves toward market and regulatory reforiml in the industry are also resulting in new questions beingv raised about the role of public policy in the expansion of service. Let us consider these questions in turn. Pricing and Financial Policies Practically all developing coullntl-ies have set tariffs Poa- electricity supplies to rural areas at below the costs of supply, even makinig generous allowances for declines in costs as load density increases (see figurlle 3.1 ). This is the same policy that the high-incomile industrial countries have used in the past. Table 4.3 shows the rural tariffs fo7r several countries compared with the long-run marginal costs of supply. In all cases tariffs are significantly below the esti- mated costs of supply, althoujgh several coulitries do at least cover fuel, oper- atingi. and mnaintenanice costs. A common policy is to have "uniforimi national tariffs,." that is. the same tariffs in rural and urban areas, even though average actual costs can be two or three times higher in rural than in urban areas. In cases where the overall fiianicial position of an electric power utility has been sound. in that it has been able to recover- the rural subsidies froni general tariffs. theni even ambitious irLral electrificationi programs have been successful- Table 4.3 Costs and Tariffs for Electricity in Rural Areas of Selected Countries, 1993 U.S. cents per kilowallt ho(ur Gentel,tit &l' Av era eg It/ait nssion tNiOfl Di.tribtttibtto etlgicd/tltt-atl ConeI^rv Fuwe C(7jC1tt tV (aptftcit( fiotul to7,ite Bangladesh 4.6 9.2 1(.6 24.4 16.0 India 2.1 5.8 8.7 16.4 0.5 Indonesia 3.8 4.1 9.8 17.6 5.8 Malaysiai 2.3 8.8 4.4 15.5 7.2 Philippinles 5.0 2.8 7.5 15.3 9.2 Thailand 5.0 3.8 8.3 17.1 7.0 Note: The tigures in th titable relate to tile lonig-nutl Inaginal cot f(t electricity une in agricAilture att low coltage. Somei e: WOrlId Batik t 1994a:. O PT IO NS F OR R UR AL EL EC TR IF ICA T IO N 45 BOX 4.1 RURAL ELECTRIFICATION ensure revenue collection; paying IN THAILAND careful attention to customer service, marketing, and community involve- Thailand's rural electrification program ment; and employing a system of began during the early 1970s. cross-subsidies from large to small Between 1975 and 1994, the number customers. Retail tariff rate structures of electrified villages increased from 20 were designed to charge larger users to 98 percent. During the same period, higher rates than small users. With the population covered outside the respect to residential classes for both Bangkok metropolitan area increased urban and rural customers, the retail from 18 to more than 80 percent. tariff structure consisted of a fixed Reasons for the program's success charge and several increasing blocks include thorough planning for careful as shown below. system expansion; keeping costs low; using a unique billing program to Source: Tuntivate and Barnes (1995). Residential Tariff Schedule U.S. ą/kWh 9 7 I 151-400kWh 6 at 8.90 per kWh - - . Momt hn400kWH 4 . 4at l0 pa kWH I l . . , I I I I I 0 100 200 3(( 400+ Moonthlv kWh Mlocks Tariff schedule includes a fixed charge and the increasing block rate: less than 150 kWh. a fixed charge of US$2(0 for the lirst 5 kWh or less; more thani 150 kWh. a tixed charge of US$3.56 foir the first 35 kWhI or less. Lifeline rates for the first 35 kWh were designed to make service affordable for poor people who used stiall amnounts of electricity, primilarily for lighting. Because of such Iimited ttse by the poor, lieline rates did not adversely affect the company's, ftiancial perfoirmance. 46 R U R A L E N E R G Y A N D D E V E L O P M E N T ly implemented. The policy is not ideal, but Thailand's experience, summarized in box 4.1. shows what can be accomplished. The program was started twenty years ago and has achieved connection rates of more than 75 percent for the whole country. with high participation by the rural poor. An interesting aspect of this program is the low lifeline rates offered to low-income consumers: as people's incomes grow and their consumption increases. they move into a high- er tariff category, eventually reaching USą 10 per kilowatt hour (kWh). The tar- iff thus gradually reduces the overall subsidy as incomes and demand grow, and the high costs of stail-up recede into the past-a highly desirable feature. Long-term subsidies for grid-supplied rural electrification are not sustain- able in most developing countr-ies. For example, as the Centre for Urban Poverty Alleviation, an Indian nongovernmental organization (NGO). has commentecl: "The highly subsidized and free power to agriculturists in some Indian states is an extreme example of political largesse in which there are no winners, only losers. The foremost loser is the State Electricity Board, not because of financial losses, but because it perpetuates the debilitating process which makes the Board less and less capable of fulfilling any part of its man- date. Urban and industrial consumers (who pay) suffer because the system is plagued by load shedding and other inefficiencies. The agriculturists hardly get the power they need. The country loses on all counlts.' More generally, cross-subsidies have five disadvantages. (Note that they do not arise in connection with the relatively small amounts, typically about 5 percenit of total consumption, needed to support such policies as lifelile rates.) They are as follows: • They underminie the utilities' financial viability, and thus their ability to provide service. In the Philippines (before the recent reforms of the elec- tricity sector) and India, financial losses diminished the utilities' ability to support the expansion of service, not only in rural areas, but also in urban areas. In addition, financial weaknesses eventually led to deterioration of the maintenance of the power plants and electricity networks. * The costs are high. In irLral areas, per capita consumption for household uses typically ranges from 25 to 50 kWh per year. and for nonhousehold uses is some two to three times this amount, bringing the total to some 100 to 200 kWh per year per person. a figure that is somewhat higher in the middle-incomie countries. Meanwhile subsidies are frequently USe5 to USą I0 per kWh, sometimes higher, and so may amount to US$5 to US$20 or more per personi served per year. These are substantial costs to bear in light of both of the hulidreds of millions of people already receiving service (table 4.1 ) and the yet greater numbers who will be demanding it in future. O P T IO NS F OR R UR AL EL EC TR IF ICA T IO N 47 * The prospects of alternatives-such as solar energy. small hydro and wind generator sets, and diesel-powered autogenerators--are all undennined. In everyday terms, the playing field is not level, a source of much concem to the numerous businesses, small-scale and large, otherwise able to offer elec- tricity services from such sources to people living in rural areas anid towns. BOX 4.2 CREDIT FINANCING the evening hours. This scheme COMPENSATES FOR HIGH COST enabled the number of households OF ACCESS IN RURAL BOLIVIA able to purchase electricity service to double or more, making connection When people in the rural villages of rates comparable to those achieved Mizque and Aiquile in Bolivia were in villages served by the central grid given the opportunity to purchase (such as Vacas; see the column in electricity service from local diesel boldface in the table). micro-grids, high costs prevented all Even though the households of but 25 percent of households from Aiquile and Mizque still had to pay accepting the offer (the cost of the ini- about five times more than the house- tial connection to the system is high, holds of Vacas on a kilowatt-hour and the cost per kilowatt hour of ser- basis for electricity service-service vice is high in such isolated micro- that was limited to the evening hours grid systems because of the high cost at that-a majority valued electrifica- of operation compared with central tion enough to subscribe when an electric grids). affordable financing scheme became The electricity companies that oper- available. The example shows that ate the micro-grids serving Mizque people are willing and able to pay and Aiquile then decided to finance prices that cover a substantial portion the connection charges of about of the cost of electricity supply and US$100 to US$125, allowing their suggests how important credit can be customers to pay back the costs in in expanding access. small installments over five years. The customers also had to agree to Source: Torres (1993). receive electricity service only during Credit Access and Electricity Adoption in Three Villages Crledlit ',o ofl hu (11ouselloldL, Price Vi/age Sv.etn lC(eTs credit w/ electricix ($/k Wli) Aiquile Diesel Yes 51 55 0.25 Mizque Diesel Yes 56 59 0.30 Vacas Grid No 0 54 0.06 48 R UR AL ENE R G Y AND DE V EL O PM ENT * Private investment in distribution is discouraged. This can. in theoryv be over- come by tax incentives or direct subsidies from public revenues. but again. one must ask whether the state can alford this and whether greater claims on public revenues do not exist, for example. from education and health. * Innovationi is discouraged. Such innovation might. for example. include providing smlall-scale supplies from renewable energy sources. More often than not, subsidies are not needed anyway. Rural consumers are willing to pay more if suppliers are willing to incorporate the capital costs of start-up and connection in the tariffs. The cases of Pura (box 3.2) and Bolivia (box 4.2) provide good illustrations of people's willingness to pay. In Bolivia. the costs of energy supplied by diesel were five times the costs of electricity supplies in urban areas, but when the capital costs of connection and supply were incorporated into the taifff (as a forimi of credit), connection rates were high. In many countries people have purchased diesel engines at cost ftor in-igation pumping. crop processing. and off-grid electricity supplies (Child and Kaneda 1975). Recently. rural households in Kenya have been pur- chasing PVs at cost. inclusive of import duties. to provide electricity when grid supplies were not available. An ideal policy is one in which prices recover the present value of costs over the long-term, that is, prices are equal to the discountecl unit cost (see Turvey and Andersoni 1977). In general. this will mean that prices will be below average costs for some time unltil demand and load densities have increased sufficiently. but above the long-run marginal costs of supply so as to provide for an eventual recovery of the initial costs of extension. Such a policy has the advantages of including a credit componenit, of being finan- cially viable, and of making it attractive to the company to expand service and promote use. As noted earlier, there is no reason why such a tariff cannot include lifeline rates without compromising the utility's financial position. Cost-Effectiveness and the Choice of Alternatives Electrification programs have centered almost exclusively on grid electrification. While this is cost-effective for high-density loads. planners have often over- looked the alternatives. The costs of electrification can be reduced by working with lower service standards. but other options are available (Munasinghe 1987). Costs of Grid Supplies Table 4.4 shows the costs of extending distribution to unelectrified areas. The incremental costs of expanding grid supplies usually have the following coin- O PT IO NS F OR R UR AL EL EC TR IF ICA T IO N 49 ponents: (a) the capital and fuel costs of generation: (b) the capital costs of reinforcing the transmission and subtransmission nietworks; (c) the extension of medium-voltage transmission networks; and (d) the establishment of a low- voltage distribution network and household connectiotis. Such costs vary from country to country and among regions in a single country. and the costs of meeting peak demands are considerably higher, but the table illustrates how costs vary in a typical situation. Grid supplies are usually a cheaper option il areas with high load densities and high load growth near the grid. Reducinig linitial Investment Costs bx Usinig Appropriate Design Standards The highi initial costs are a major balTier to serfvice extensioni, but can be reduced appreciably by using designi standards suitable for areas with lower demiand (once demand rises, standards can be increased). The demands of most rural consumers range from 0.2 kilowatts (kW) to 0.5 kW; however, the minimum service connectioni ratings in developing-country utilities are typically 3 to 7 kW. which raises costs. For the same reason, the costs of installation and wiring provided by the utility are high. Simplifying wiring codes and using load lim- iters (circuit breakers) for lower levels of consuimptioni can reduce costs signif- icantly. In addition, using cheaper poles and involving local people in works and mainiteniance will also reduce service costs. When service is being provided to millions of people, the aggregate of such economfies is considerable. Micro-Grids Supplied bY Diesel Generators Decentralized, isolated distribution systems have been comimonll for several decades in remote population centers, and in most developing countries predate the establishment of grids. For example. such systems were serving numerous villages and towns in northerin Ghana before the grid extensions in the 1980s. Box 4.2 on Bolivia provides another example. The costs of such systems typi- cally range from US20() to USą60 per kWh. Table 4.5 provides data for two schemes, each serving fifty consumers, one in Yemen. the other in Pakistan. Maintenance and hiigh fuel costs have been long-standing problemiis with diesel gellerators. The systems are often in remote locations. and the difficulties of purchiasing imported spare parts and fuel have often made themn Lunreliable. Electricitv Suppliesfiromn Reniewable Energy Soturces The cost estimnates in table 4.5 are ftairly typical of smiiall diesel plants. They also show why energy from solar, wind, and micro-hydio schemes has become 50 R U R A L E N E R G Y A N D D E V E L O P M E N T Table 4.4 The Effects of Line Length and Consumption Levels on the Costs of Rural Electrification (costs in U.S. cents pier kWh) Cost 0or;i/7oniCl (jiUit costs Totals' Generation and Transmission Fuel 3- 4 Capital 2 - 3 Transmission & subtransmission reinforcement 3 Subtotal (rounded) 10 Medium-voltage extension and low-voltage distribution 3 km spur line, 20 households 45 55 3 km spur line. 50 households 20 30 1 km spuir line. 20 houselholds 15 25 1 km Spur line. 5(0 households 7 17 Hieh-densitv rural loadIs 2 12 na. Notl applicable. a. Total, inicltide the getteratioti attd tratimiiton costs shown itt the rIpper rows, plus the extension and distri- hutitn costs showwn fUr cach otf the five cases of line letigthiand Igiad density. Assiniptionis: Medi uni-voltage linies. USS10.(KY) per kilolmeteL. iow-voltage distnbtitilt. lUS$5.0(11 per kilomilete.r. US40 per kilovolt alipere tor si inple pole-top distributioni tranisforniers; conlsutitptiotl levels of 35 kWh per tontlth per household: 2( mleters if lovw-voltage circuit per houtse: 10 percent clicoutt rate; twentv-fie yeear lifetitnie for circuits. fifteen years for tIrattsforiters. The above figures ire adjasted for the higher losses experieticed int serving ruiral ener- gy demtatids. Sotto-: Arun Santghvi. persottal cotnttttInication. Table 4.5 The Costs of Small Diesel Supply Systems in Pakistan and Yemen Svstcin rle;telt v PoNkisnro Yement Number of consumers 50 50 Consumption/consumer/motith 2-5 kWh 25 kWh Size of generator 20 kW 20 kW Investment cost US$14,250 US514,250 (including network and civil works] Fuel cost per liter $ 0. 17 $ 0.22 Electricity costs US$0.35/kWh US$0.5 I/kWh Sitavcr: Meunier (1993i. cited in Foley t 1995i. attractive in rejions where the solar insolation. wind regime, or hydro resources are suitable. A report by the former Office of Technology Assessment of the U.S. Congress (OTA 1992) found that the all-inclusive unit costs of elec- O PT IO NS F OR R UR AL EL EC TR IF ICA T IO N 51 tricity were as low as USą 12 per kWh for micro-hydro. depending on the site. USą45 per kWh for PVs, and USą25 per kWh for small wind sets (the costs of the latter two technologies have declinied significantly since). Electricity for local distribution can also be generated from such fuels as biogas or biomass. depending on local availability of resources (see box 3.2 on the case of Pura). Micro-hydropower can be one of the cheapest options for providing elec- tricity to riral areas that are too far away from the grid to be connected to it, and can sometimes also supply the grid. This is certainly true where local capacity to manufacture turbines exists, as in China and India. In India, a pro- grain to finance micro-hydro systems privately will both serve local demand and feed into the central grid system. Another aspect of micro-hydro is the care needed when selecting a site, given the possible variation in stream flows during the year and from river to river. Costs vary significantly. depending on the site and the terrain. In Nepal. for example, some 25 percent of total costs for a micro-hydro project can be for transportation of equipment and materials alone, but are muchi lower in less mountainous regions. However, if all elements of the project cycle adopt- ed a low-cost approach, even in Nepal the costs of extending the grid to small consumers could be as low as US$150 per consumer. One important aspect of such approaches is the participation of the local community, which reduces costs, enhances consumer satisfaction, and helps to provide a financially viable investment. Box 4.3 provides an example of an innovative and suc- cessful micro-hydro development effort in Peru. at a village center distant from the national electricity grid. The development of micro-grids, whatever their primary source of ener- gy. requires a significant level of community consensus and support regard- ing such factors as billing, service, and organization. Local participation is a key ingredient in the design of such isolated systems, in their inmplemnentation, and in their day-to-day operation. This is self-evident in the case of small local systems that are the result of local self-help or private initiative. However, even isolated systems put in place by a national program are more efficient if they enlist the cooperation of local consumers. Central grid systems also ben- efit from local participation in rural distributioll. In Bangladesh, for example. locally managed rural electric cooperatives are responsible for distributing power that they purchase from the grid or generate locally. Their record of billing, collection, losses. and maintenanice is significantly better than that of the main power utility in charge of urban distributioni. Developments in new renewable energy technologies have greatly expanded the options for supplying electricity in rural areas (see chapter 5 for a more complete discussion). Consider developments in solar PV technolo- gies, for example. Barely twenty years ago. at the time of writing the Bank's 52 R U R A L E N E R G Y A N D D E V E L O P M E N T BOX 4.3 A CREDIT PROGRAM son-days of labor). The ITDG super- FOR MICRO-HYDRO IN PERU: vised construction and provided tech- A PROMISING BEGINNING nical and management training. The capacity of the Chalan micro- One constraint to the wider use of hydro scheme is 25 kW and the capi- small-energy technology is access to tal cost was US$82,700. Direct con- capital. Many financiers see micro- nections have been provided to hydro schemes as high risk, and are eighty families to date, with most therefore reluctant to make funds remaining households expected to available. This reluctance stems from connect in the future. A further 617 the frequent lack of adequate support families from surrounding communi- for training, operation, and manage- ties are indirect beneficiaries of the ment of schemes by borrowers. scheme as they now have access to In Peru in 1993, the Intermediate improved health and education facili- Technology Development Group (ITDG) ties, battery charging, agro-process- and the Inter-American Development ing, and workshop and communica- Bank established a rotating fund to tion services and also benefit from make loans of up to US$30,000 avail- public lighting in the village. able to rural communities and rural An elected committee that is inde- enterprises for micro-hydro schemes. pendent of the village council man- The interest rate is 8 percent per year, ages the schemes and is responsible and the payback period is five years. for overseeing operation and mainte- The first scheme to be installed nance; collecting tariffs; promoting under the program supplies electricity safe and efficient use of the energy; to the village of Chalan, fifty miles and managing the introduction of new from the national grid. Chalan has a uses, such as workshop equipment. population of 540 people in 120 fami- At present, the power from the plant lies. In addition, nineteen other small is priced at USc9 per kWh, although settlements or hamlets in the area this will increase to cover the costs of use Chalan as their main center for generation and to provide a reserve services. fund for repairs and replacements. The initiative for the micro-hydro Conversely, as demand increases, scheme came from the village council the costs of generation will fall, pro- and a local NGO that was working viding the opportunity to reduce tar- with local farmers. The funding for the iffs. scheme came from four sources: a The credit program aims to install loan from the credit program, a grant around twelve schemes. The pro- from the local NGO, the resources of gram has already attracted consider- the village council, and the labor of able interest among local communi- every family in Chalan, who agreed to ties and enterprises, government contribute a set amount of time to bodies, and international agencies. construct the channel, install the pipe, and build the powerhouse (the com- Source: Courtesy of Stephen Fisher, munity provided a total of 4,318 per- ITDG, personal communication (1996). O PT IO NS F OR R UR AL EL EC TR IF ICA T IO N 53 previous policy paper on rural electrification, costs were several hundred thousand dollars per kilowatt, conversion efficiencies were low, and the oniv applications were in aerospace and in specialized switching circuits. Today PV systems are providing electricity economically to rural areas of develop- ing countries for domilestic lighting and appliances. vaccine refrigerationi for health clinics. village water pumps. telephones. inrigation pumping. street lighting. and schools. In areas far from the grid or where delivering conven- tional fuels is difficult. renewables are often the least-cost options. Significant markets for PVs are thriving in rural areas of developing countries-in Kenya. for example (box 4.4). Cost-efficient use of renewable energy technologies in rural areas can be elIcouraged by educating people about the possibilities and by providing train- ing in how to install and maintain the techinologies. Electricity distribution companiies also need to be encouraged to consider the costs and benefits of the solar alternative to grid supplies to meet small loads and provide supplemen- tary power on the longer distributioni networks. The World Bank Group's Solar Initiative, launched in 1994. is identifyinig such oppoilunities (see chapter 7). Regulatory and Price Reforms, Unbundling, and Privatization Extendinig electricity supplies to large numbers of people an(d growing mar- kets in developing countries will require financially sound and efficiently maniaged electricity supply industries. However, despite a sixfold increase in electricity generating capacity and output since 1970 and the raipid increase in the number of customiiers served. by the late 1980s the financial positions and efficiency of electricity supply industries had deteriorated badly in many countries. Several of these countries are now attemptinig to reform their elec- tricity sLipply industry and its ownership and regulationl so that it can be oper- ated on commercial principles and under independent managemiient (see World Bank 1993b. 1994d). Box 4.5 summarizes the key aspects ol' the reform process and provides some examples. hnplication.s for Ruiral Electrification Countries that permit more conmiercial policies and independent ("arm's length") regulation in the electricity supply industry will see the industry's finhancial positioii greatly strengthened, and thereby its ability to extend and offer reliable service to rural areas. Average real tariffls in the electricity supply industry in developing countries had(l eclined from lUSe5.2 per kWh (already 54 R U R A L E N E R G Y A N D D E V E L O P M E N T below the costs of supply) in 1979 to less than USą3.8 per kWh by the early 1990s, whereas supply costs were around USe I0 per kWh. Given electricity sales of approximnately 2.2 trillion kWh per year, annual revenue shortfalls were thus in excess of US$130 billion per year, more than was needed to BOX 4.4 PV MARKET GROWTH areas is now competitive with the cost OUTPACES GRID CONNECTIONS IN of grid electricity. RURAL KENYA Rural electricians are joining forces with urban entrepreneurs and solar About 20,000 households in rural electric suppliers based primarily in Kenya have purchased PV systems Nairobi. Currently, at least eight com- for electric lighting, despite the panies supply the PV market and absence of any government program compete through agents in rural and in the face of import taxes of 30 areas to market, install, and maintain percent or higher on PV equipment. PV systems. Such growth in PV use The private sector has accounted for illustrates the high value people in about two-thirds of the total installed remote locations place on electric capacity. By contrast, the country's lighting. They view PVs as an attrac- highly subsidized rural electrification tive alternative to waiting indefinitely (RE) program has reached only for the grid system, which may never 17,000 households (see the figure reach them. below). The cost of PV-supplied elec- tricity for low levels of lighting in rural Source: ESMAP (1994a). Comparing Grid and PV UJse. 1987-92 Connectiojis (Ihoml,ands) 20 15 5 1987 1988 1989 1991) 1991 1992 Grid RE HOUs,ehold PV O PT IO NS F OR R UR AL EL EC TR IF ICA T IO N 55 BOX 4.5 INSTITUTIONAL REFORM ing much greater attention to the per- IN THE ELECTRIC POWER SECTOR formance of the distribution enterpris- es to ensure that they can act as reli- Many countries are introducing reforms able power purchasers and be to increase the efficiency of their elec- depended upon to collect electricity tricity industries and attract more pri- bills on time and minimize losses. vate sector financing and manage- Asian utilities are tending toward incor- ment. Reforms include poration, as companies and govern- * Cutting governments' role by separat- ments are not focusing on unbundling ing policymaking from regulation and to date except in Pakistan and the operation of utilities Philippines, but are paying particular * Establishing transparent regulatory attention to regulation. China has systems with predictable price-set- passed a new electricity law that is ting rules and procedures expected to pave the way for separat- * Putting power sectors on a commer- ing the regulatory functions of the cial basis, preferably as companies provincial power enterprises. Likewise, under a commercial code in India, many of state boards are con- * Raising capital on local and foreign sidering radical reform based on the markets "Orissa model." The government * Encouraging direct investment and com- passed the first major regulatory law in petition in generation and distribution. Orissa, and this is likely to be adopted Since 1992, the pace and depth of in the other states. energy sector reform have intensified In Eastern Europe and the former in developing countries, particularly in Soviet Union, progress has been much Latin America and Asia. In Argentina. slower. Many are beset by old, poorly Bolivia, and Peru, governments are maintained plant; excess capacity; and unbundling the sector and selling facili- environmental and safety problems. ties to the private sector. Competition is These issues are limiting the focus on evident in generation and the acquisi- reform, and little new investment is now tion of distribution facilities. While each being sought. The Czech Republic and country is developing its own approach Hungary have made good progress to reform, many of the reforms have toward privatization. however, and been modeled on those in Chile, reforms in Poland and Ukraine are because it completed its restructuring promising. Still governments seem and privatization long before other reluctant to relinquish their regulatory countries in the region. Most recently, roles. Brazil and the smaller countries of In Africa, governments are increas- Central America have begun to follow ingly recognizing the need for reform the trend. as they come to understand that reform In most Asian countries, progress is in other regions could leave them intensifying. As private power projects behind and discourage private involve- develop, reforms to encourage compe- ment unless they, too, commit to tition and increase the efficiency of changes. existing producers and distributors Source: World Bank (1 993a,b, pick up speed. Governments are pay- 1994d) and internal staff reports. 56 R U R A L E N E R G Y A N D D E V E L O P M E N T expand capacity, and ten times the amount needed for even ambitious rural electrification programs. As box 3.1 pointed out, the bulk of the implied subsi- dies were for commercial, industrial. andl household consumers who were both able and (in the economist's sense of the teriml) willing to pay tor service. The main question about the reforms now being sought, therefore, is not about their desirability, but whether they will bring about sufficient commer- cial interest in rural electrification. Ulrban mar-kets are not only easier to serve and more lucrative, but also have big supply backlogs and are growing rapid- ly. All the countries that have so far succeeded in providing universal service to rural areas and towns (the high-income industrial countries) did so with public leadership and financial support. Governments maiidated service expansion, often coupled with regulatory permission to recover the extra costs from higher prices in urban areas. provided tax incentives, or both. Whether the private sector in developing countries will respond to the challenges involved is unknown. So far, private investors have been mainly interested in genieration. not distribution, something that adds to the problems of public authorities that wish to encourage private investment and service extension on the one hand. but find limited private response on the other. In Asia, private ownership of installed capacity almost wholly in generation ranges from less than 20 per- cent in China, India. and Thailand. to 20 to 50 percent in Indoniesia, Pakistan. and the Philippines. to more than 60 percent in Malaysia. The figures are higher, oni average, for the Latin Americanl countries, where privatization is proceeding more rapidly. but as in Asia, private investors' intelest in distlibii- tion is still quite low. Yet another complicatioln is that with rare exceptions. countries cannot accomiiplish the reforimi in a short period, so for the next sev- eral years. policymackers and the industry will have to continue to supply and expanid service while reforms are under way. Approaches In these circumstances, rural electrification will require public involvement for some time to come, and as with the general institutional reformis now tak- ing place. ilo uniquely best approachi is available: much will depend on indi- vidual country's circumstances. The following are some possibilities: * A continuance of public involvement in distribution. with regulations requiring companies to expand service in ways consistent with achieving satisfactory financial rates of return to investment. * Joint public-private investments in distribution. with the same regulatory requiremelnt as the first option. O P T IO NS F OR R UR AL EL EC TR IF ICA T IO N 57 * When the distribution company is privately owned, a regulatory require- mient to expand service. coupled with permnissioni to meet the financing requiremlients out of rural-specific tariffs or (a second-best measure) a sur- charge oni general tariffs. * Full price liberalization on private investimients in distribution, with regu- lators acting as monlitors of prices anid the efficiency of service, the latter being defined to inlclude the expansion of service where the costs and returns justify it. * Encouiragement of the formation of electricity cooperatives, the private development of micro-grids, or both. This list is not comprehensive, and other options are possible. Barring an economic surprise. however, public leadership does seem to be necessary to expand service to rural areas and towns: it will need to be open to alternative anrangements and to a variety of ways of encouraging private investment. C H A P T E R F I V E Innovations in Renewable Energy Recent developmenits in renewable energy technologies have greatly added to the options available for improving rLiral ener-gy supplies. The main technologies suite(d to rural areas are micro-hydro. biogas. wind generators. wincd pumps. solar heaters for hot water, and sustainable ways to provide wood supplies. All these are important souL-ces of energy andi can be developed furthel,- as illustrated by the examples of China (box 5.1 ) and Piura in India (box 3.2 ). A mor-e receni development has been the use of photo- voltaic (PV) systems to provide electricity supplies for such small-scale appli- cationis as electric lights and dotimestic appliances, refr-iger-ationl for clinics, vil- lage water puimps, street lightiig, and healthi clinics and schools. For small- scale applications in rulal areas, PVs are often less expensive and more reli- able thani grid supplies or diesel motors. The encouraging feature of the Kenya example discussed in box 4.4 was that it was financed on a purely private basis (van der Plas 1994). Solar thermal electric systems U Silng parabolic dish- es at-e also showing much promilise for small-scale supplies (Ahmed 1993). Aside fiomi their environimenital appeal, new' renewable energy technologies are attracting priofessional interest for several reasons, namnely: the abundanice of the solar resource, from which most formis of reniewable energy are derived: technical progress and cost reductiolIs: and the modularity of the technologies. The rest of this chapter will focLis on techliological progress and on the sup- porting policies needed if renewable energy is to be widely used in rural areas. Technical Progress in Using the Solar Resource Each vear. the earth receives energy from the SLuln equal to 10,000 times the world s commercial energy consLImption and mor-e than 1(00 times the wor-ld's 58 INN O VAT IO NS IN RE NEW AB LE ENE R G Y 59 proven coal. gas. and oil reserves. Modern solar electric schemes, such as PV systems and solar-thermial power stations, can currently convert 7 to 15 per- cent of the incidenit energy into a form useful for consumption. and in theory would need less than I percent of the world's land area to meet all its energy needs. Solar eneryv is an abundanit and infinitely renewable resource. Insolationis are about 2,000 to 2,500 kilowatt hours (kWh) per square meter per year in m1aniy areas of cleveloping countries, which means that a PV scheme of I square meter can supply 100 to 300 kWh. dependinlg on the type of cell used, which is sufficient for lighting. radio. television. and ironing, while a 5 square meter panel set is sufficient to meet the water pumping needs of a village or to provide for irrigation on a small farin. Technical dexelopments have been impressive, and reductions in the costs of all major- solar energy technologies, incIlidinig derived forms of solar ener- gy such as wind. have been substantial (Alimned 1993: Johansson and others 1993). As figure 5.1 shows, in the early 1970s PV modules cost several hun- (ired thousanid dollars per peak kilowatt and applications were largely con- litied to aerospace and other specialized uses. By 1990 costs had fallenl to USS6,000 per peak kilowatt and PVs had become commercially Xviable for a wide range of small-scale uses: costs have (leclined by another 20 to 30 per- cenit since. An estimated 1(0,0)0 to 2'00.000 systems are installed in (level- oping countries. includinig 40,00() in Mexico, 0,(000 in Kenya. 16,0)00 in Indonesia. 1 5,000 in China. I 0,00() in Brazil. and 4.000 in Sri Lanka. BOX 5.1 RENEWABLE ENERGY China's Solar Energy Research IN CHINA Institute. This is equivalent to the heat that several hundred megawatts of China has long promoted renewable electricity generating capacity could energy technologies for its large rural deliver. Demand has grown by 50 per- population. Nearly 800 of China's cent in each of the past two years, 2,166 counties depend on small-scale stimulated by a rise in farm incomes. hydro for electricity, and some 5.5 mil- Until recently, the PV program had lion households use biogas systems reached a modest 4.500 households. that process animal manure, kitchen However, an active research program wastes, and night soil into biogas for is under way, and in Qinghai Province cooking. Many small-scale wind alone, where insolations are good, machines are in household use- plans call for PV electrification of 120,000 in Inner Mongolia alone. The 100,000 households in the next twen- Ministry of Agriculture estimates that ty-five years. private artisans have assembled more than 4 million square meters of solar Source: Terrado and Cabraal, staff heaters using devices designed by memorandum (1996). 60 R U R A L E N E R G Y A N D D E V E L O P M E N T Engineering and econiomic data show that further progress is likely on two fronts: * Scale and productioni economies. World output grew from I megawatt per year fifteen years ago to around 70 megawatts today. a growth rate of more than 30 percent per year. Markets are still small, but the technolo- gies are modular, and economies of scale and the technical possibilities for batch production have barely been exploited. * Cell, module, and systems design, along with improvements in conver- sion efficiencies. Improved materials. multijunction devices and novel cell designs to capture more of the solar spectrum. and concentrator (Fresnel) lenses to focus sunlight onto high-efficiency cells are further areas of rapid developimient. Figure 5.1 Actual (1970-92) and Projected (1993-2015) Costs of PV Module U.S. dollars per Aatt-peak 1 0,000 0 1,000~ 100- 00 0, 0 00 o o 0 00 0 Cost ranges required for 10 - 0 80 0r commercial application: . - S ~ ~ ~ _ F Small-scale - - 8 o 8 _ s applications l _ _________________________________________________ _ oPower generation_ ._ (decentralized) at Power peak loao generation (decentralized) 0.1- ....................... ~~~~~~~~~at base load 0. 1 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 Note: The range of costs ntarkedon the graph show the P'V isodeilc costs reqLoired to compete wiltl sTflall-,cale applications and Wiill) deceletraliZ,d power generation assutinag seipplv Costis of USCX to USse ti per kWh idi hase load i and US, 16.5 per kWh itt peak loadi. The spread in the points reflects the spread in costs ot diffel- enttechnotogies. Sorce: Ahmed ( 1993 INN O VAT IO NS IN RE NEW AB LE ENE R G Y 61 The technologies are now at the point where they are competitive for off- grid supplies, and are therefore of special interest to rural areas. Box 5.2 on Indonesia's experience with PVs provides a good example of the respective economics of grid and PVs for supplying rural areas. At high load densities the grid is clearly preferable, at lower load densities PVs are a more cost- effective option. Good progress has also been made with small- and large-scale thermal solar schenmes and with derived forms of solar energy. such as wind and bio- mass resources for power generation. In China and Middle Eastern countries solar-thermal collectors are a popular heat source for domestic hot water, and promising experiments with solar cookers are afoot in Asia and Africa. Another promising solar-thermal technology is the parabolic dish for small- scale power generation. and when scaled up. for grid supplies. Studies by the U.S. Department of Energy have indicated that the costs of 25-kilowatt mod- ular units vary from USą12 to USą2() per kWh IAhmed 1993). In the case of wind for power generation on a larger scale, costs have declined from around USą 15 to USą25 per kWh to USO4 to USc8 per kWh in favorable locations. For small-scale applications the costs are USą2n per kWh. but can be comii- petitive for off-grid supplies. Of all reniewable energy sources. biomass (ligneous and herbaceous crops and agricultural and mullicipal wastes) is the largest, most diverse, and most readily exploitable. Biomass residues are often available in large quantities as agro-industrial wastes. Recuperation, more efficient production, and more rational use of bioinass residues and forest resources could make many agro- industries energy self-sufficient as well as provide additional energy to the economy in general. This requires the conversion of biomass into cleaner and more convenient fuels (gases, electricity. bri-quettes). Developing-couLitry agro-industries (saw mills, sugar mills, and palm oil mills) already use biomass residues to generate power and heat for the indus- try's own use. On-site utilization is currently limited to raising process heat and power. but its use could be expanded to heat for drying and product treat- ment. Also. cogeneration of electricity for a minii-grid can be economically beneficial. Off-site utilization of residues includes direct utilization of residues in industrial oil-, wood-, and coal-fired combustion systems. Biomass conversion technology may find application in situations where petroleum fuels are either unavailable or where the cost of power from engines fueled by producer gas is lower than from diesel or gasoline engines. Gasification combined with the use of gas in an internal combustion engine or tulbine is an efficient way to convert solid fuels into shaft power or electrici- ty on a small scale, and more advanced processes for larger scale gasification are under developmeit. Heat gasifiers are technically reliable anid economi- 62 R U R A L E N E R G Y A N D D E V E L O P M E N T cally attractive compared with conventional alternatives. Apart from use in the rural agro-industry (for example, for tea drying), non-agro-industrial applications are also viable (for instance, brick and ceramic kilns). Consequenitly. this technology is al-eady in large use in developing countries. Economists believe that the costs of new renewable energy technologies will decline further because of scale economies and the stimulus that market growth will give to further research and development and innovation. Japanese and American studies of the learining curves for PV technologies have found that for each doubling of the cumulative volumIIe of production during the past fifteen years. costs have declined by 2(0 percent (Ahmed 1993: Anderson and Williams 1994). Renewable eneigy technologies are fertile ground for innovation; the possibilities for further developments and cost reductions are far from being exhausted. Policies toward New Renewable Energy Sources in Rural Areas New reniewable energy technologies still accounlt ftor less than 2 percent of the primar-y energy supplies of developing countries. but in light of their promise. with good economic and environimenital policies and with the devel- opment of the necessar-y support systems for installation and maintenance. their market shares should expand. Investments will also be required to acquainit energy engineers and managers with the technologies and to edu- cate and train eneineers and skilled workers. As with all new and innovative technologies. developino the best approaches will take a good deal ot' effort (and some trial and en-or). A recenit review of' PV programs in the Pacific Islands (Liebenthal. Mathur, and Wade 1994), for example. found that many PV systems failed after installation. and it was only when supporting services were introduced that the programs beganl to succeed. These services included training techni- cianis. ensuring timely mainiteianice, collecting fees on a regular basis, pro- viding proper oversight to prevenlt the diversioni oft revenues to other projects, anid obtaining promilpt feedback on needs from local user communllities and passing the iiforimiation on to the supplying utility. Similarly. a program in India introduced PVs in several states for domestic and street lighting, coIll- munity televisions. water pumnping, and other puiposes dul-ilIg 1986-92. Yet out of more than 5,000 street lights, mor-e than half in some states all were not working a short while later, and the other applications exhibited similar' fail- ure rates. The findings were particularly distul-binig because PVs are durable, are relatively simple to install, and require little maintenianice. As in the Pacific Islands. the prohlem turned out to be the lack of supporting services. I N N O V A T I O N S I N R E N E W A B L E E N E R G Y 63 BOX 5.2 SOLAR PV HOME decentralized rural electrification in SYSTEMS IN RURAL INDONESIA two niche" markets: (a) where the grid may be nearby, but consumers Solar home systems using PV tech- are dispersed and the load density is nologies have the potential to provide low, making grid extension to individ- electricity to a large number of rural ual consumers expensive; and (b) households. In Indonesia 3,000 solar where the grid is more distant and is home units were installed during not expected to be extended soon 1988-92. A recent evaluation found (see the figure). In these two markets that the units are working as planned. combined, the economic potential for Today, more than 16.000 units have solar home systems in Indonesia is been installed through public pro- estimated at more than 5 million of the grams and by commercial dealers. approximately 30 million rural house- In a typical 50-watt-peak system, a holds that lack grid supply. solar PV panel is installed on the roof Solar home systems also offer (a) of a rural home or shop. The panel superior quality and quantity of light, charges an automobile-type battery compared with kerosene lamps, along that is used at night to run up to four with absence of indoor fumes, pollu- energy-efficient light bulbs and a tion. and fire hazard associated with black-and-white TV or a radio for four such lamps; (b) no need to haul bat- to five hours a day-approximately tery to a central charging facility; and equivalent to consumption of 0.5 kWh (c) environmental benefits. One envi- of grid electricity per day. Systems ronmental problem that must be dealt ranging from 20-watt-peak (for lighting with in all solar home projects involves alone) to 200-watt-peak (for schools, ensuring proper disposal or recycling meeting halls, or higher-consumption of batteries. households) have also been used successfully. Source: Arun Sanghvi, internal com- Solar home system units offer a fast munication. and least-cost means of providing Grid Extension and Solar PV Switchover Values Outside Java Grid 3 Kilometer Medium-Voltage Extension Number of' HouLeholds Served 25(0 - Inn . . Grid extenisioni is cheaper (1) . . - , -..... . ;- (1 4 8 2 1 20 24 28 Grid Load Densityv kilowatts per squalre kilomieter) 64 R U R A L E N E R G Y A N D D E V E L O P M E N T Successftul programs recluire two m.ain ingredients: (a) paying propel attention to program development. for example. initiating surveys of renew- able energy resources, carrying out project identification and preparation, and investing in education and training: and (b) creating good enabling conditions through economically efficient pricing. credit. and tax policies. Program Development Establishing a program involves significant effort. The first task is to survey solar and wind resources. Such surveys have long been carried out for hvdcro programs, as geological and engineering investigations have usually been car- ried out for many potential sites, and data on river flows have been collected for several decades. but they are rarely available for solar or wind energy. In addition. a program of field tests of equipmenit with a fairly substantial nluill- ber of consumers (often several thousandl households) will be necessary not only to justify the investment in the equipmiienit. but to establish supporting maintenance services and to monitor progress. As with any new area of investment, issues arise in connection with risks and uncertainties. In the field of renewable energy. sonme of the questions raised ar'e at a quite elementary level. For example. sonie projects designers may not even have assessed the level of solar. wind, and biomass resources. while potential consumers are often not up-to-date on technical developments. costs, and how similar projects elsewhere have performed. The predisposition of institutions to resist change is also a factor that widely inpedes new invest- ment and initiative. Another major task is to ftamiliarize professionals in the electricity indus- try enginieers, managers. financiers, regulatorswitih the new possibilities. Expanded education and training, including visits to operating projects. may help to change negative perceptions and aid the development of investment programs. Beyond this, the facilities and curriculums of universities and technical colleges may need to be developed to provide appropriate educa- tion and training. The fiianicial requirements needed to develop programs. idenitify and pre- pare investments, and provide education and trainiir are generally small in relation to the costs and benefits of the investments that eventually emerge. As with the developmenit of programs using more traditional renewable energy forms, such as micro-hydro schemes. biogasifiers, and sustainable ways of usinig woodfuels. the participation of nongovernmental organizations in pr'o- ject development can be beneficial. Bilateral aid organizations and non- governmiental organizations, often working in collaboration. have also been influential in establishing pilot schemes and offering education and training to INN O VAT IO NS IN RE NEW AB LE ENE R G Y 65 engineers and technicians from developing countries. The many applications of PVs in developing counitries owe much to such efforts. Prices As discussed in chapter 4. for rural areas both peak and average costs are sometimes twice the marginal costs of electricity supplies in a typical urban situation. If new renewable energy technologies are to succeed as an eco- nomic alternative to conventional power plants and to grid electrification in rural areas, then as chapter 4 emphasized. the electricity industry must adopt cost-reflecting price policies. Such pricing policies include time-of-day and seasonal, as well as regional. variations in prices. Given the high costs oft meeting peak demand and the declining cost of PVs, several European countries, Japan. and the United States are now con- ducting trials on the use of PVs to supplement peak loads. These trials include inet metering," arrangements such that small users can sell surplus power to the grid. Aside from the cost advantages. the relevance for rural areas and towns is that decentralized generation reduces line losses and voltage drops and provides a backup source of supplies in the event of line failures. Credit As with some other forms of rural energy. the initial cost of acquiring the equipment constitutes a significanlt barrier to the widespr-ead adoptioni of renewable energy sour-ces. The development of innovative financing schemes. including supplier credits and leasing arrangements. is thus a critical element of any renewable energy program. Those who have studied the credit problem closely have concluded that subsidies are not needed, at least for small-scale applications, for which renewables are competitive (Cabraal an(d Cosgrove- Davies 1995). Taxes and Subsidies In recognition of their positive externalities (of innovation) and their enviroll- mental advantages, one can make a good case for exempting renewable ener- gy technologies from or at least substanitially lowering taxes and duties. while at the same time taxing convenitional energy industries' supplies in accordance with standard principles of tax policy. Experts have also long argued in favor of imposing corporate and sates taxes on electricity on the grounds that it is a fair-ly price inelastic product. In practice. however. governmients have typical- ly pursued the opposite policy, namilely. of not only exempting electricity from 66 R U R A L E N E R G Y A N D D E V E L O P M E N T taxes, but often subsidizing it. and imposing significant taxes on renewable energy equipment. (See Cabraal arid Cosgrove-Davies 1995. who report that in Sri Lanka. import duties added some 30 percent to the costs of PVs. Kenya also has duties and taxes on PVs while subsidizing electricity to encourage local manuifacture and assembly.) There is. furthennore, an economic case for providing public financial sup- port for renewable energy technologies in program development, demonstra- tion, education, training. and monitorinig, again, in recognition of their positive externalities and environmenital advantages. This is the rationale behind the Global Environment Facility's (GEF's) financinig of renewable energy: during 1991-94, the GEF financed renewable energy projects in eight countries that included PVs. wind power, and micro-hydio in India and biomass for power generation in Brazil (the GEF quarterly operations reports provide more details). Twenty-five renewable energy projects are cunently under review or in vanous stage of preparation and appraisal in twenty couLIntries. At the national level, finanicial support could come either fiom public rev- enues or from user char-es or surcharges on the use of fossil fuels. The latter has the advantage of making the programs less dependent on public financing. A good example is the United Kingdom's "noffo." or non-fossil-fuel obligatioll programil, which applies the revenues from a small surcharge on electricity to sUpport the winners of competitive bids to supply electricity from reniewable energy sources. Several similar schemes exist in Europe. Japan, and the United States, all of which have active public programs to develop renewables. C H A P T E R S I X Cooking Fuels: Toward More Sustainable Supply and Use As incomes rise, rural and urbani households gradually shift from using duLng, crop wastes, and wood for cooking to mnodein fuels, such as liq- ulid petroleum gas (LPG) and kerosene. Households in rural areas and lower- income urban households rarely use electricity for cooking. Given that people will be using biofuels for many years to come, policies need to focus on (a) improving the efficiency with which biofuels are used. (b) promoting more sustainable ways to supply biofuels, and (c) facilitating the transition to the use of modern fuels for cooking. Improving End-Use Efficiency with Biomass Stoves The modern bioniass stove is an importan1t development for the millions of people who have ready access to low-cost biomass. but who caninot afford more expensive modern fuels. The fuel savings, often as much as 30 percent, reduce cash outlays: diminish the time spent collecting, fuelwood; decrease smoke by ilproviing combustion and the use of flues, thereby reduci ng the worst health effects of biofuel use (see the figures on Brazil in table 2. 1): and reduce pressures on scarce wood resources. Developing-country governmients. donors, and nongovernmental organi- zations (NGOs) supportedl programs implemented in the late 1970s and early 1980s. and commonly assumed that their benefits were self-evident. They believed that people would adopt the improvedl stoves quickly, and that the initial promotional programs would leacd rapidly to self-sustaining markets in the new products. Hence. most early efforts focused only on dissemination. and were oblivious to local customs, the economic setting. and the availabili- 67 68 R U R A L E N E R G Y A N D D E V E L O P M E N T ty and prices of local biofuels. Based on laboratory experimenits. early pro- grains anticipated three- to fourfold increases in energy efficiency and a 75 percent decrease in wood consumption. With hinidisight. we can see that many stoves did not perform as well as anticipated in the field, and the experiments both overestimated the energy efficiency of improved stoves and underesti- mated the energy efficiency of traditional stoves. Today, a 25 percent reduc- tion in wood consumptioni is considered more realistic. However, development practitioners have learned a great deal from these early efforts, and the notionis that improved stoves could improve energy effi- ciency substantially. albeit less than originally thought. and reduce the dlamage to health caused by smoke. are still validl. A recenit evaluation (Barnes and others 1994a) of the programs fouLid that the best of themii had the following features: * Identification of appropriate markets. Through local inquiry, the best pro- grams first identified the families most likely to a(lopt and benefit from the improved stoves. These were always low-incomile households (but generally not the poorest) who usually had some cash inconme, a large por- tion of which they spent on food and cooking t'uel. * Participation in stove design and mar-ket testing. The best progr-amils involved muci initer-actioni between designers, producers, and users and( included stove testing by representative households. This process kept pro- ducers and designer-s focused on meeting- the nee(ds of prospective users. which increased the likelihood that more people would purchase the stoves. a Provision of public funding. The stoves were not heavily subsidized. Public funds were used to support marketin,-. design. and extension. * Standardizationi. Standardization of stove parts andl techniques facilitated widespread manufacture and reduced costs. Properly managed. improved stove programils. such as the one in China (see box 6.1). caii result in significanit adoption rates. Successful programlis have focused on the user groups most likely to benefit from improved stoves. including those paying significant amounts for fuel or those who must walk long distances to collect fuelwood. Subsidies have not proven particularly effective in promoting stoves because they create an artificial demand that hinges on the subsidy that is not sustainable when the subsidy is withdrawil. However'. external granit support can be valuable if used for SuChI activities as laboratory testing. carrying out field work. obtaining users' views on alterna- tive approaches. and demonistratinig the stoves. An examinationi of successful stove projects in Africa leads to similar con- clusions. A project in Madagascar financed by the International Development C O O K I N G F U E L S 69 Association promoted ihproved charcoal stoves (Government of Madagascar 1994). After two years of preparatory activities (identifyingc potential users. testing stoves. selecting appropriate stove models, obtaining feedback from users and producers, and testing marketing channels), an active commercial- ization phase consistitn of promotional activities resulted in the use of 45,000 improved stoves in the capital. Antananarivo. which amounts to about 20 per- cent of all stoves in use. More than ten different models of improved stoves are available (all metal, all clay, and mixed models). and several different types of production units exist, both in thie informal and in the fonnal sectors. The private sector car-ied out all production. marketinig, and sales. Today. the sup- port for the project has been eliminated. but activities continULe. and estimates BOX 6.1 LESSONS FROM THE stoves are affordable (about US$9), WORLD'S LARGEST STOVE with government contributions aver- PROGRAMS aging only US$0.84 per stove. Indian stoves have a minimum 50 The two largest stove programs in percent government subsidy (about the world are the Chinese National US$4.30 per stove). While dissemi- Improved Stove Program, which has nation has been impressive, follow- installed 120 million stoves in rural up surveys indicate that only half the households, and the Indian National improved stoves are still in use. India Programme on Improved Chulhas, has since revised much of its pro- which has provided 8 million stoves. gram to use existing commercial dis- Chinese stoves are mainly biomass tribution and marketing channels. types for cooking. but include dual- use stoves for cooking and heating in Source: Barnes and others (1994a); the northern states, where winter Ramakrishna (1991); Smith and oth- temperatures are low. Improved ers (1993). Grid Extension and Solar PV Switchover Values Outside Java Grid 3 Kilometer Medium-Voltage Extension ([Chillm, /Idia * Programii concenitrated on areas with - Programil disperses efforts. including greatest fuelwood shortages. areas with no fuelwood shortages. * Direct contacts between goveriinmenit * Culimbersome, top-down administration and cotLities bypass much bureaucracy. dilutes the program's effectiveness. * Stove adopters pay the ftull cost of * Governmenit pays producers half the cost materialis and labor. of every stove sold. * Governimient sUppoNts producers with * Governmenit support is mainily financial. training in stove building and promotion. 70 R U R A L E N E R G Y A N D D E V E L O P M E N T indicate that more than 100.000 improved stoves are in use. The goveinment intends to extend the activities to all urban areas where char-coal is the major fuel. Another project funded by the International Development Association in Tanzania had a similar outcome. The four-year project (1988-92) established a production capacity. mostly among private sector artisans, of 5.000 improved stoves per month. Distribution and sales are through a network of private sec- tor retail outlets. By the end of the project. more thani 60,000 improved stoves had been sold, while production and sales of improved stoves have continued on a self-sustained market-driven basis since (Goveriment of Tanzania 1992). The social, economic. and environmental benefits of improved stoves can be large. and the successes have demonstrated the utility of well-maliaged programs. The economic returns of successful programs are gooc. In urban areas, where most people purchase woodfuels rather than gather woodfuel themselves, the pavback time is sometimes as little as a few months. In rural areas of resource scarcity, the savings in terms of the time and energy people spend gathering fuelwood are significanit. Improving Charcoal Efficiency For many families. an interimiediate-level fuel between woodfuels and kerosene and LPG is charcoal. Not only is it relatively inexpensive. but it gen- erates much employment. Like fuelwoodl. charcoal can be purchased in pre- ferred quanitities; but unlike fuelwood, it burns without smoke, does not cre- ate dangerous tlames around cookinig vessels, and requires a simple stove whose heat output is relatively easy to control. However, dispersed local char- coal industries often use inefficient charcoaling kilns that use more wood resources than necessary. The potential for increasing biomass efficiency with improved charcoal kilns has been recognized since the early 1980s. Kiln models based on tradi- tional designs, but with higher heat-tranisfer efficienicies, have been developed in Rwanda, Senegal. and Tanzania in collaboration with end-users. However. this has showil that such technical innovations should be complemented by easily enforced standards, systematic training, and demonstrationls to convince traditional charcoalers to adopt the improvements. Such programs. particular- ly in wood-scarce areas, work best when complemilenited by natural forest man- agement efforts that have raised the price of wood stocks. thereby creating an incentive for charcoalers to use more efficient technologies. Recent experience in East Africa has demonstrated that a comprehensive approach to introducing new kiln technologies to the small-scale charcoaling industry can succeed. More efficient charcoal production is beginning to result in the use of fewer wood resources in peri-urbani areas and is bringing down the cost of the fuel. C OO K I N G F U E L S 71 Developing More Sustainable Ways to Supply Biomass In the late 1970s and early 1980s, biofuels developimient tended to focus on plantations and commuLnity or private woodlots. Such progr-ams proved expensive and had limited success, as discussed in chapter 2. During the past ten years or so, the possibilities for increasing biomass production through agro-forestry and farm forestry have become more widely recognized, as have more sustainable forms of forest management involving local participation. Agro-forestry. ftarm forestry. and natural torest inaagenielit projects yield a host of benefits in addition to an increase in fuelwood production-including fodder, medicines, fruits, and poles for constr-uctioni. Agro-Forestry aind Farmti Forestrv Agro-forestry entails planting trees, shrubs (and sometimes grasses, such as vetiver grasses), on farmlands in alternating patterns with more traditional crops. In addition to enhancing agricultural production by providinlg shade and complementary use of soils, agro-forestry plays an importanlt role in alle- viating fuelwood shortages and resource custody. Such intercroppilng reflects old and once well-known practices that have been neglected in recent years because of persistent poverty, population pressures. and failure to address tenurial or proper-ty rights. However, interest has been rekiindled. Small far-m- ers often practice agro-torestry and farm forestry independently in response to economic needs and ecological problems. The spread of farm forestry in Kenya is an often-cited example of relatively autonomous development of effective practices. but many others can be found in China. India, Pakistan, the Philippines, and elsewhere. In the Philippines. the island of Cebu lost most of its forests in the early part of this century. The resulting fuelwood scarcity and rising prices caused farmers to begin planting trees to supply fuelwood to urban markets, whicil has helped ameliorate at least some of the environimental damage caused by the earlier deforestation (see box 6.2). In India. the first social forestry pro- grams were aimed at establishinig community woodlots. but few communi- ties wvould engage in tree planting for urban markets. However, private faml- ers did respond to the incentives and began planting trees on their marginal lands. Recent programs in West Bengal have granted landless laborers tenure to manage trees on public wastelands. Agro-forestry and farm forestry can be encouraged through farmer edu- cation and extension programs and supported through agricultural and forest-y research. Such efforts are both environmentally desirable and eco- nomically beneficial foir the following reasons: 72 R U R A L E N E R G Y A N D D E V E L O P M E N T * Farmner-s outnumber foresters by several thousand to one. so involving farmers in planting trees and shrubs can dramatically accelerate afforestation. * By their nature, agro-forestry and faariml forestry investments are more closely related to faimiers' needs, as they supply fodder, building materi- als, green mulch. fruit. and other by-products that are sometimes more valuable than the firewood itself (technically also a by-product). BOX 6.2 WOODFUEL MARKETS inducing more intensive tree planting PROMOTE TREE PLANTING ON THE and management activities among ISLAND OF CEBU rural farmers and landowners. Most commercially traded woodfuels origi- Cebu, the island province in the south nate from intensively managed agri- central Philippines, is one of the worst cultural lands and consist mainly of cases of environmental degradation in fast-growing, multipurpose tree Southeast Asia. With virtually no for- species like Gliricidia and Leucaena. est cover, but with steep terrain, pop- The growing, harvesting, and trading ulation densities of 520 people per of woodfuels is a substantial source of square kilometer, and widespread cul- income and employment. Woodfuels tivation of annual crops like corn meet significant urban energy throughout its rugged interior, environ- demands and represent a renewable mentalists consider Cebu to be on the and locally produced energy source brink of ecological collapse. that annually saves the economy mil- Despite increased use of kerosene lions of dollars in foreign exchange. and LPG, most of the island's 2.7 Rather than promoting the cultiva- million inhabitants continue to tion of tree species that require exten- depend on wood, particularly house- sive inputs, which require more labor, holds and businesses in the metrop- capital, and time to harvest, rural olis of Cebu City. This heavy depen- development programs should also dence on woodfuel is often cited as recognize the benefits of the low-input a major cause of the province's envi- alternatives that Cebuano farmers ronmental woes. As a result, many have practiced for nearly a century. government and NGO officials They should view commercial wood- believe that woodfuel use should be fuel markets as an opportunity to pro- discouraged in urban areas and that mote more widespread tree planting commercial trade in woodfuel should and management practices through- be more tightly regulated, if not alto- out rural areas of the province rather gether banned. than as a problem to be controlled by A recent study found, however, that restrictive legislation. commercial markets for fuelwood and charcoal in Cebu City may actually be Source: Bensel (1994). C O O K I N G F U E L S 73 Agro-forestry and tann forestry practices reduce run-off, soil erosion, and surface evaporation and iilpr-ove micro-climiiates and soil water retention. Farmers can use the foliage of the trees they grow to provide nitrogen-rich manure or mulch for their fields and to improve soil structure. Such attrib- utes contribute to sustainable farming systems by raisinig the produLIctivity of farm soils (Gregerson. Draper, and Elz 1989). Fieldl studies of agyro-forestry practices in developinig countries over the past twenty-five years have consistently found that agro-forestry has favor- able effects on farm yields and incomes, just as similar studies have found in the high-inicome countries for the last celntUry (Doolette and Magrath 1990: Gregersen. Draper, and Elz 1989). This is encouraging. because it demon- strates that biofuels cani be supplied in ways that will not only help make agri- cultural practices sustainable but will also improve agricultural productivity and incomes. A recent stu(dy of twenty-onie agro-forestry projects in six Central Amer-icall countlies (Currenit. Lutz. and Schen- 1995) found that agro-forestry practices were profitable under a broad rangie of conditions (table 6.1 sumiliarizes the results). These findings echo those of several other studies. An earlier survey of field results (Doolette and Magrath 1990) of niore thani a dozen studies ill Brazil, Colombia, India. Indonesia. Malawi. Niger, the Philippines, the Republic of Korea. and Sudan did not find a single case in wlhich the returns were less than 15 perceit. Table 6.1 Returns to Agro-Forestry Practices in Six Central American Countries NMimber of Bemfi&t/-'co.U IVl Pralio 60) k /rio'd A grofs.jore,vn SVsw"Im .steink shiied oti 20).', discuOI' (O ear.) Trees with crops 5 1.8 3.4 Alley cropping 9 '.1 1.9 Contour planting 4 i 1.6 2.) Pereninials with trees 4 1.8 4(0 Hoime Larden 4 2.2 na Taungyah 8 2.5 4.9 Woodlot It) 1 (i 9.2 a The high dnstratett corinmitmelnt tlo power wettor reform,l anid it, projc Papaper . Waslinigton, D.C.: World Bank, I ridustlv aitld Ener'gy Depart menit. vail cIer Plas. Robert. aiid Will eim Flioor. ()96. 'Market-Drisen Appo tac h Can Ill tilllate Lightirig Optiotns tor Rit-;il Areas. Power Engineeriti liticrnatitnial tMarclh/Alpil. Von Piscibke. J.D.. Dale W. Adaims. anld Goirdin Donald. i 983. Rural FinatincialI Markets in Developi ng Coultit -ies. BaItimoitre. Marylavid: The i Johns Hopkins tniAnersity Press. Wade. Herbert. 19()1. PhotolsttItaic Rural Electrification Projects in the Pacific Region atnd Lessons Learned.'' Repiti prepari-ed for t(lie World Baitk, Pacific Regionala Energy Assessment. Talliti. FreticiI PolViiesia: Son th Pai 'ifiC IilstiitLate for Renewtieble Fnergy. Whittiiigton. Dale. Donald T. LaLina. anid Ximning Mu. i991. 'avine totr Urlban Services: A Study ot' Water Veidilig alid WillitILgess to Pay for Water il Onitshia. Nigeria.'' World Developimtett 19(2 - ) 179) 98 WVorld Batnk. 1975. Rural Electrificatiott. Policy Paper. Washingtott, D.C. 1978. Forestry. Polic\ Papei. WVashington, D.C. - .1987. Tropical Forest Action Plait. Wisltiiigtoit. I).C. - 9. i9 )() World Developirtenit Reporit 1991): Poverty. News York: Osford University Press. - . 1991. The Forest Sector. Policy Paperl Washinigton, D.C. -. 199'2 World Development Report 1992: DCselopimenti aitd the Enviritinent. New York: Oxfocd U niversitv Press. - .I 993a. Enetrgy Efficiency aiid Conservationi in the Des eloping World. Policy Paper. Washington, D.C. 19. I993b. The Wiotld Batik's Role inl the Electric Powver Sector. Policv Paper. Washi netotl, D.C. - 1993e. World Bank Atlas 1994. Washinttoin. I).C.: The Wotrld Bank. - .1994;1. 'RUral Electrificaitiori in Asia: A Review otf Bank Experietice." OperaLtionis Esalitationu DepaLtment. Report No. I3P91. Woirld Bank. Washington, D.C. June. 1. 994b. World Develoipment Report 1994: Infrastructulre for Development. New' York: Oxtoird ULniveisity Press. 118 R U R A L E N E R G Y A N D D E V E L O P M E N T - 1994c. "Review of Implemilentationi of the Forest Sector Policy." Washin-gton, D.C. - 1994d. Poweer and Energy Efticienicy: Status Report oil the Bank's Policy and IFC' Activities.' Joint Wol-d Ban k/Intern ational Fiianice Corporation semi uiar. presented to the Executive Directors. July 7. Washington, D.C.: Eneigy and Industry Departilmelit. . 1995a. "Lending for Electric Power in Sub-Saharan Africa." Washington, D.C.: Operat ions Evaluation Departmnent. 1 I995h. World Bank Participation Sourcebook. Eniviroriument Department. Social Policy and Resettlemilenit Division. Washington, D.C.: The World Bank. 1 995c. Global Econlomic Prospects tid the Developing Cotintries. Washington, D.C. World Bankl/United Nationis Developmenit Programime. Forthconmirig. India RUral Energy Study. Washington, D.C.: Worldk Bank. }, . " ,>1f, '0 \\ S |, - !0| B. I .3 - - -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~