ENERGY 48017 AND POVERTY Special Report March 2009 Bangladesh' Restoring Balance: sRural Bangladesh's Rural Energy Energy Realities Realities M. Asaduzzaman Special Douglas F. Barnes Shahidur R. Khandker Repor t006/09 March Energy Sector Management 2009 Assistance Program Copyright © 2009 The International Bank for Reconstruction and Development/THE WORLD BANK GROUP 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing March 2009 ESMAP Reports are published to communicate the results of ESMAP's work to the development community with the least possible delay. Some sources cited in this paper may be informal documents that are not readily available. Thefindings,interpretations,andconclusionsexpressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, or its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The Boundaries, colors, denominations, other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the ESMAP Manager at the address shown in the copyright notice above. ESMAP encourages dissemination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. ENERGY AND POVERTY Special Report 006/09 March 2009 Restoring Balance: Bangladesh's Rural Energy Realities M. Asaduzzaman Douglas F. Barnes Shahidur R. Khandker Energy Sector Management Assistance Program 5580-CHFM.pdf i 5580-CHFM.pdf 3/11/09 10:57:31 AM 3/11/09 10:57:31 AM 5580-CHFM.pdf iiii 5580-CHFM.pdf 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM Contents Acknowledgments ix Abbreviations and Acronyms xi Units of Measure xiii Currency Equivalents xv Energy Conversion Factors xvii Executive Summary xix 1 Introduction 1 What Is the Knowledge Base? 1 Study Goal and Objectives 2 Principles for a Rural Energy Strategy 2 Study Method 4 Survey Instruments 4 Structure of This Report 7 2 Household Energy Use 9 General Consumption Patterns 9 Energy for Cooking 11 Energy for Lighting 13 Summing Up 14 3 Household Energy Demand 15 Expenditure Patterns 15 Cooking Energy: Biomass and Its Opportunity Cost 16 Lighting Energy 17 Asset Ownership Patterns 21 Demand for Other Energy: Quantitative Analysis 24 Summing Up 26 4 Household Gains from Energy Use 27 Biomass Cooking: From Traditional to Modern Energy 27 From Kerosene to Electric Lighting 32 From Manual to Mechanized Farming 35 Overall Income Gains: Moving Up the Energy Ladder 37 Summing Up 40 iii 5580-CHFM.pdf iii 5580-CHFM.pdf iii 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities 5 Energy Consumption and Rural Production 41 Enterprise Types and Distribution 41 Growth-Center Microenterprises 42 Home and Village Enterprises 47 Enterprise Energy Demand and Profitability: A Quantitative Analysis 51 Energy for Farming 53 Summing Up 55 6 Institutional Framework for Rural Energy Supply 57 Institutional Challenges and Potential 57 Building an Integrated Approach 63 7 Policy Recommendations 65 Problems and Potential of Biomass Energy 66 Rural Electrification and Rural Development 71 Toward Policies for Household Petroleum Fuels 73 Enhancing Rural Energy Projects, Policies, and Strategies 75 The Way Forward 76 Annexes 1 Selected Tables from the Household Survey 77 2 Statistical Models 103 3 Consumer's Surplus as a Measure of Welfare 111 References 113 Glossary 115 List of Special Reports 117 Boxes 1.1 What Are the Benefits of Rural Energy? 4 2.1 Modern Energy Benefits for Rural Families 13 Figures ES.1 Energy Use and Expenditures in Rural Bangladesh, 2004 xx 1.1 Types and Sources of Energy Demand 3 1.2 Map of Survey Areas in Bangladesh 5 2.1 Rural Household Consumption by Source (percent energy consumption) 11 3.1 Energy Expenditure Distribution by Energy Source (percent expenditure) 15 3.2 Household Biomass Collection Time by Income Quintile (thousands of Tk per year) 18 3.3 Regional Variation in Electrification 19 4.1 PM10 Concentration by Time of Day and Household Location 29 4.2 PM10 Concentration by Age Group and Gender 29 4.3 Pollution Factors and PM10 Concentrations 31 5.1 Comparison of Energy Use Structure by Enterprise Location 50 6.1 Annual Load Shedding, 1993­2004 59 6.2 Bangladesh Power Sector Organization 60 6.3 Oil and Gas Sector Organization 61 A3.1 Consumer Demand Curve 111 A3.2 Consumer's Surplus 112 iv 5580-CHFM.pdf iv 5580-CHFM.pdf iv 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM Contents Tables Cooking Fuels and Energy Efficiencies xvii Household Lighting Energy in Developing Countries xviii Battery Electricity xviii 1.1 Number of Entities in the Census and Sample by Type and Division 6 1.2 Coverage of Growth Centers by Division 7 2.1 Household Distribution of Energy Sources by Division (percent) 9 2.2 Annual Household Consumption (physical quantity) 10 2.3 Energy Sources for Cooking and Parboiling of Rice by Division (annual kgoe per household) 11 2.4 Distribution of Biomass Supply Sources by Energy Type 12 2.5 Energy Sources for Lighting (annual kgoe per household) 13 2.6 Electrification of Villages and Households by Division 14 3.1 Annual Energy Expenditure by Income and Division (Tk per household) 16 3.2 Biomass Collection Time by Income and Division (annual hours per household) 17 3.3 Household Electrification Connection Rate and Consumption by Income Quintile 18 3.4 Comparison of Kerosene and Electric Lighting Services 20 3.5 Annual Price of Lighting by Income Quintile and Division 20 3.6 Price of Electric Lighting by Income, Connection Type, and Division 21 3.7 Household Ownership of Cooking Stoves by Type (number per 100 households) 22 3.8 Average Effective Cooking Hours by Income Quintile 22 3.9 Household Ownership of Lighting Appliances by Electrification Status (number per 100 households) 23 3.10 Electric Lighting by Income and Division 23 3.11 Ownership of Non-lighting Electric Appliances by Income Quintile 24 3.12 Summary Statistics of Outcomes and Important Explanatory Variables of Household Energy Demand Regressions 25 3.13 Estimates of Household Energy Demand 25 4.1 Mean Daily Hours in Household Location by Age Group 28 4.2 Annual Energy Use by Number of Students in Household 33 4.3 Study Time by Household Electrification Status (average number of hours) 33 4.4 Households by Lighting System and Income 34 4.5 Household Expenditure by Electrification Status on Kerosene and Other Energy Sources, by Income Group (Tk per year) 34 4.6 Welfare Gains from Switching Lighting Systems 35 4.7 Farm Household Distribution by Tillage/Irrigation Technologies 36 4.8 Gross Productivity of Paddy by Tillage/Irrigation (Tk/decimal) 36 4.9 Paddy Productivity in Boro Season by Land Ownership (gross value in Tk/decimal) 37 4.10 Household Income Differentials by Electrification Status (Tk per year) 38 4.11 Summary Statistics of Outcomes and Energy Use Variables of Household Welfare Regressions 39 4.12 Estimates of Household Welfare (Tk per month) 40 5.1 Microenterprise Distribution by Location and Activity Type 41 5.2 Distribution of Growth-Center Microenterprises by Region 42 5.3 Distribution of Energy Sources by Activity Type 44 5.4 Major Energy Sources by Use Type 44 5.5 Energy as a Percentage of Total Costs by Activity Type 45 5.6 Price per Unit of Energy for Non-lighting Business Use 46 5.7 Perceived Advantages and Disadvantages of Electricity Use 46 5.8 Perceived Advantages and Disadvantages of Fuelwood Use 47 5.9 Perceived Advantages and Disadvantages of Kerosene Use 48 5.10 Microenterprise Distribution by Location and Length of Operation 48 5.11 Microenterprise Distribution by Number of Employees and Location 49 5.12 Comparison of Energy Output and Cost by Use and Location 50 v 5580-CHFM.pdf v 5580-CHFM.pdf 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities 5.13 Summary Statistics of Outcomes and Explanatory Variables of Enterprise Energy Demand Regressions 51 5.14 Estimates of Enterprise Energy Demand (N = 2,290) 52 5.15 Summary Statistics of Outcomes and Energy Use Variables of Enterprise Profitability Regressions (N = 2,290) 52 5.16 Effects of 1 Tk Increased Energy Use on Enterprise Outcome (annual Tk per enterprise) (N = 2,290) 52 5.17 Comparison of Tillage Methods by Division 54 5.18 Irrigation Method and Energy Type by Division 54 5.19 Farmer Irrigation Method by Division 55 5.20 Energy Consumption in Farming (average owner-user households per year) 55 A1.1 Survey Sample (number) 77 A1.2 Household Distribution by Ownership of Agricultural Assets (percent households) 77 A1.3 Household Distribution by Various Characteristics (percent distribution) 78 A1.4 Annual Household Income by Source (average Tk per household) 79 A1.5 Household Distribution by Income Source (percent) 79 A1.6 Household Consumption Expenditure by Budget Item (Tk per year) 80 A1.7 Household Distribution by Type of Energy Consumption (percent) 80 A1.8 Annual Household Energy Consumption: All Divisions (average per household) 81 A1.8.1 Annual Household Energy Consumption: Dhaka (average per household) 81 A1.8.2 Annual Household Energy Consumption: Rajshahi (average per household) 82 A1.8.3 Annual Household Energy Consumption: Chittagong (average per household) 82 A1.8.4 Annual Household Energy Consumption: Khulna (average per household) 83 A1.9 Average Energy Consumption for Agriculture and Transportation (owner-users only) 83 A1.10 Energy Consumption by Business Type 84 A1.11 Household Distribution of Cooking Stove Ownership (percent) 84 A1.12 Household Ownership of Cooking Stoves by Type (number per 100 households) 85 A1.13 Household Distribution of Number of Meals and Cooking Times (percent distribution) 85 A1.14 Hours of Operating Cooking Stoves by Type (daily) 85 A1.15 Household Ownership of Lighting Appliances (number per 100 households) 86 A1.16 Hours of Running Lighting Appliances (cases per day) 86 A1.17 Household Ownership of Plug-in Electric Appliances (number per 100 households) 87 A1.18 Hours of Running the Plug-in Type of Electric Appliances (daily) 87 A1.19 Ownership of Irrigation Pumps and Power Tillers/Tractors (percent households) 87 A1.20 Hours of Running Pumps and Power Tillers/Tractors (annual hours per unit) 87 A1.21 Cost of Household Energy Consumption: All Divisions (Tk per household per year) 88 A1.21.1 Cost of Household Energy Consumption: Dhaka (Tk per household per year) 88 A1.21.2 Cost of Household Energy Consumption: Rajshahi (Tk per household per year) 89 A1.21.3 Cost of Household Energy Consumption: Chittagong (Tk per household per year) 89 A1.21.4 Cost of Household Energy Consumption: Khulna (Tk per household per year) 90 A1.22 Household Biomass Energy Behavior: All Divisions (percent distribution) 90 A1.22.1 Household Biomass Energy Behavior: Dhaka (percent distribution) 91 A1.22.2 Household Biomass Energy Behavior: Rajshahi (percent distribution) 91 A1.22.3 Household Biomass Energy Behavior: Chittagong (percent distribution) 91 A1.22.4 Household Biomass Energy Behavior: Khulna (percent distribution) 92 A1.23 Household Consumption of Selected Biomass Energies: All Divisions (per year) 92 A1.23.1 Household Consumption of Selected Biomass Energies: Dhaka (per year) 92 A1.23.2 Household Consumption of Selected Biomass Energies: Rajshahi (per year) 93 A1.23.3 Household Consumption of Selected Biomass Energies: Chittagong (per year) 93 A1.23.4 Household Consumption of Selected Biomass Energies: Khulna (per year) 93 A1.24 Operational Land and Household Ownership of Gardens 94 A1.25 Household Ownership of Trees and Bamboo 94 vi 5580-CHFM.pdf vi 5580-CHFM.pdf vi 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM Contents A1.26 Household Ownership of Cattle 94 A1.27 Household Distribution by Duration of Electricity Connection (percent distribution) 95 A1.28 Electricity Sources (percent of connected households) 95 A1.29 Basis of Charging for Electricity (percent of connected households) 95 A1.30 Household Distribution by Electricity Supply Problem (percent of connected households) 96 A1.31 Cost of Kerosene--Using Devices 96 A1.32 Cost of Electric Bulbs and Tube Lights 96 A1.33 Women's Time Use for Various Activities (daily hours per person) 97 A1.34 Time Spent Making Clay Stoves (hours per stove) 97 A1.35 Time Spent on Clay Stove Maintenance 97 A1.36 Respondents' Attitude toward Electricity (percent distribution) 98 A1.37 Respondents' Attitude toward Fuelwood (percent distribution) 98 A1.38 Respondents' Attitude toward Kerosene (percent distribution) 99 A1.39 Respondents' Attitude toward LPG/LNG (percent distribution) 99 A1.40 Respondents' Knowledge of Price of Kerosene (percent distribution) 99 A1.41 Respondents' Reasons for Costliness of Kerosene (percent of those who reported kerosene as costly) 99 A1.42 Respondents' Knowledge of Price of Electricity (percent distribution) 100 A1.43 Respondents' Reasons for Costliness of Electricity (percent of those who reported electricity as costly) 100 A1.44 Respondents' Knowledge of Renewable Energy (percent distribution) 100 A1.45 Respondents' Reasons for Interest in Purchasing Renewable Energy Devices (percent of those interested in purchasing) 101 A1.46 Respondents' Reasons for Disinterest in Purchasing Renewable Energy Devices (percent of those not interested in purchasing) 101 A2.1 Tobit Estimate of Household Energy Demand (N = 2,388) 103 A2.2 Effects of Energy Use on Household Outcomes by Use Type (N = 2,388) 105 A2.3 Effects of Energy Use on Household Outcomes by Energy Source (N = 2,388) 106 A2.4 Tobit Estimate of Enterprise Energy Demand (N = 2,290) 107 A2.5 Effects of Enterprise's Total Energy Use on Profitability (N = 2,290) 109 A2.6 Effects of Enterprise's Energy Use on Profitability by Source (N = 2,290) 109 vii 5580-CHFM.pdf vii 5580-CHFM.pdf vii 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM 5580-CHFM.pdf viii 5580-CHFM.pdf viii 3/11/09 10:57:33 AM 3/11/09 10:57:33 AM Acknowledgments This study was conducted with technical and financial support from the Energy Sector Management Assistance Program (ESMAP), the collaborative efforts of the Bangladesh Institute of Development Studies (BIDS) in Dhaka and the World Bank. In addition, a detailed background survey of microenterprises was completed by Data International, Dhaka. The survey team completed its work in 2005 and the consultants submitted a preliminary report to the World Bank in 2006. The World Bank team completed the revised report by 2007 and after extensive peer review by the South Asia Regional Energy Program the report was finalized and submitted to ESMAP for publication in late 2008. At that time, the report was put into production under the guidance of ESMAP. It is the intention of the authors that you find this report interesting and insightful. The principal project leaders were: Douglas F. Barnes and Shahidur R. Khandker, who supervised the international World Bank team comprised of: M. Asaduzzaman and Abdul Latif who led the local BIDS team, and Najmul Hossain who was the Data International project manager. Special thanks go to Keith Openshaw and Rashid Faruqee, Hussain Samad, and Grayson Heffner of the World Bank team for having made extremely valuable contributions to the project, including the drafting of sections of this report. Thanks also go to M. Iqbal, who coordinated all work in Dhaka as part of the World Bank team. The work on institutional issues was originally prepared by M. Khalequzzaman, who also commented extensively on the final draft of the report. An early review of rural energy issues in Bangladesh, prepared by M. Nurul Islam, was quite useful in setting the overall context of rural energy in Bangladesh. Preparation of this report would not have been possible without the dedicated efforts of the BIDS staff and consultants. We greatly appreciate the comments of the World Bank peer review team, which vastly improved the final version of the report. Peer reviewers included: Priti Kumar, Dana Rysankova, Chrisantha Ratnayake, Raihan Elahi, Sudeshna Bannerji, Sameer Akbar, and Marlon Lezama. Salman Zaheer, Manager of the South Asia Energy Unit, guided the review meeting for the report and also made many valuable comments on the report. We also thank the staff of the Ministry of Power, Energy, and Mineral Resources, whose constructive comments over the course of the project and on the final results were useful in completing this report. Finally, we thank former and current staff members of the Energy Sector Management Assistance Program, for their help in making this report possible. Ede Jorge Ijjasz-Vasquez provided support for the completion of this work. Also, Dominique Lallement provided both earlier assistance and direction in completing this report. Marjorie Araya from ESMAP and Thaisa Tiglao from ESMAP (now EXTOP) played a crucial role in assisting throughout all stages of the publication process, including production of the study in its final form. Finally, the project owes a debt of gratitude to Norma Adams, who edited the final version of this report and played a major role in enhancing its quality. ix 5580-CHFM.pdf ix 5580-CHFM.pdf ix 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM 5580-CHFM.pdf x 5580-CHFM.pdf 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Abbreviations and Acronyms ARI acute respiratory infection BBS Bangladesh Bureau of Statistics BCSIR Bangladesh Council for Scientific and Industrial Research BIDS Bangladesh Institute of Development Studies BPC Bangladesh Petroleum Corporation BPDB Bangladesh Power Development Board CNG compressed natural gas ERC Energy Regulatory Commission ESMAP Energy Sector Management Assistance Program EXTOP World Bank Office of the Publisher FAO Food and Agricultural Organization GOB Government of Bangladesh HIES Household Income and Expenditure Survey IAP indoor air pollution IDCOL Infrastructure Development Company Limited IEA International Energy Agency IFRD Institute of Fuel Research and Development LNG liquefied natural gas LPG liquefied petroleum gas NGO Non-governmental organization PBS Rural Electric Cooperative (Palli Bidyut Samity) PGCB Power Grid Company of Bangladesh PM10 particles with an aerodynamic diameter less than 10 microns PM2.5 particles with an aerodynamic diameter less than 2.5 microns PV photovoltaic REB Rural Electrification Board SEDA Sustainable Energy Development Agency SHS solar home systems xi 5580-CHFM.pdf xi 5580-CHFM.pdf xi 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM 5580-CHFM.pdf xii 5580-CHFM.pdf xii 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Units of Measure GJ gigajoule GWh gigawatt hour kg kilogram kgoe kilograms of oil equivalent klm-hr kilolumen-hour kWh kilowatt-hour loe liters of oil equivalent MCF thousand cubic feet MJ megajoule MW megawatt µg/m3 micrograms per cubic meter W watt xiii 5580-CHFM.pdf xiii 5580-CHFM.pdf xiii 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM 5580-CHFM.pdf xiv 5580-CHFM.pdf xiv 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Currency Equivalents 2004 US $1 = 60 Bangladesh takas 2005 US $1 = 64 Bangladesh takas 2006 US $1 = 68 Bangladesh takas 2007 US $1 = 69 Bangladesh takas 2008 US $1 = 68 Bangladesh takas 2009 US $1 = 69 Bangladesh takas xv 5580-CHFM.pdf xv 5580-CHFM.pdf xv 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM 5580-CHFM.pdf xvi 5580-CHFM.pdf xvi 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Energy Conversion Factors Cooking Fuels and Energy Efficiencies Energy Content Cooking Stove Efficient Cooking Fuel Source (MJ per unit) Efficiency (%) Energy (MJ per unit) Electricity (kWh) 3.6 70 2.5 LPG (kg) 45.5 60 27.3 Natural gas (m3) 38.0 60 22.8 Biogas (60% methane) (m3) 23.0 60 13.8 Kerosene, pressure (kg) 43.0 55 23.7 Kerosene, wick (kg) 43.0 35 15.1 Charcoal, efficient (kg) 30.0 30 9.0 Charcoal, traditional (kg) 22.5 20 6.0 Bituminous coal (kg) 22.5 25 5.6 Fuelwood, efficient (15% moisture) (kg) 16.0 25 4.0 Fuelwood, traditional (15% moisture) (kg) 16.0 15 2.4 Crop residues (5% moisture) (kg) 13.5 12 1.6 Dung (15% moisture) (kg) 14.5 12 1.7 Leaves and grass (kg) 13.5 12 1.6 Sources: O'Sullivan and Barnes (2006) and various World Bank reports. Note: Energy content = total energy by energy source; efficient cooking energy = energy absorbed into cooking pans or devices. xvii 5580-CHFM.pdf xvii 5580-CHFM.pdf xvii 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Household Lighting Energy in Developing Countries Output Based on Lumens per Watt Lighting Type Light Output (lumens) klm-hr per kgoe klm-hr per kWh Non-electric Paraffin candle 11.8 2.33 0.20 Kerosene, wick 11.4 1.15 0.10 Kerosene, hurricane 32 1.92 0.16 Kerosene, pressure 2,040 17.53 1.48 Incandescent (watts) 25 230 109.12 9.20 40 430 127.50 10.75 50 580 137.58 11.60 60 730 144.30 12.17 100 1,280 151.80 12.80 Florescent (watts) 10 600 711.63 60.00 20 1,200 711.63 60.00 40 1,613 478.27 40.33 Compact florescent Philips lamp (15-W) 894 706.88 59.60 Philips lamp (9-W) 369 486.28 41.00 Osram sol lamp (6.14-W) 240 463.60 39.09 Source: O'Sullivan and Barnes (2006). Battery Electricity Storage Capacity Usable Storage Capacity Lighting Type Power Rating (amps) (watt hrs.) (watt hrs.) D Cell (high-quality battery) 0.5 6 4.8 Chinese D Cell 0.2 2 1.6 Car battery (60 amp) 60.0 720 576.0 Source: O'Sullivan and Barnes (2006). Note: Battery amp hours * 12 = watt hours capacity. 20 percent is lost in charging. A battery can only discharge to 80 percent of capacity (amp hours*12*.8). xviii 5580-CHFM.pdf xviii 5580-CHFM.pdf xviii 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Executive Summary Sincethe1990s,Bangladesh--oneoftheworld'spoorest institutions that implement them, have not been countries--has taken large strides toward achieving assigned a high priority. the Millennium Development Goals (MDGs), having With the exception of rural electrification, most outperformed most low-income countries on a range intervention programs have emphasized commercial of social indicators. The country is on target to achieve energy, consumed mainly in urban areas. At the same a two-thirds reduction in infant and child mortality time, the rural electrification program has continued by 2015. Food security has improved significantly, to expand, a forestry master plan has been approved and gender disparity in primary and secondary for implementation, and both the government and enrollment has been eliminated. Yet in 2005, nearly donors are supporting efforts to popularize improved 40 percent of the nation's 140 million residents and biomass stoves. Yet the effectiveness of these 44 percent of rural residents were below the poverty initiatives, in terms of people's overall energy-using line (World Bank 2006).1 behavior, remains under the radar screen of many Bangladesh is one of the world's lowest energy development researchers. producers. Electricity, critical to economic growth, has reached only about one-third of households, despite Study Goal the country's successful rural electrification program. With the exception of kerosene, commercial fuels Lack of comprehensive data and analysis and are beyond reach for many. Biomass fuels, collected knowledge gaps regarding Bangladesh's current rural mainly from the local environment only two decades energy reality have precluded the development of a ago, are fast becoming a marketed commodity as strategy for modern and efficient energy use. Against access to local biomass becomes ever more difficult. this backdrop, the World Bank Group, through a The stark reality is that many rural residents are consultative process, initiated two surveys--one dependent on such fuels as agricultural residues, focused on rural households and the other on village dung, and even leaves and grass for cooking. microenterprises and rural growth centers--to Knowledge of rural energy in Bangladesh lags elicit information on energy-using behavior and recent developments in energy technologies and characteristics. Subsequently, the Bank commissioned policies. Until recently, the potential for renewable studies on the market structure for energy and the energy and the adverse health effects of indoor air macro-level dimensions of biomass supply and pollution (IAP) on women and children were ignored. demand. Research on the structure and efficiency of rural Thisstudy--thefirsttoconcentrateonBangladesh's energy markets has also been overlooked. Policy energy systems and their effects on the lives of rural analysis and recommendations for rural energy people--drew on these background studies, as well development, along with their implications for the as other World Bank­financed research on IAP 1See Bangladesh Country Brief (www.worldbank.org). xix 5580-CHFM.pdf xix 5580-CHFM.pdf xix 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities and rural infrastructure, to present a rural energy on urban growth--representative of only about strategy for the country. The study's broad aim was to 20 percent of the country's 140 million people--would identify ways to improve the living standard in rural lead to inequitable social and economic development. Bangladesh through better and more efficient use of Making the benefits of modern energy services energy, while creating an environment conducive to available in rural areas can promote decentralized growth and poverty reduction. development and growth and help rural residents For any developing country, the crux of a rural become more productive, thereby mitigating urban­ energy strategy is to have more and better choices rural disparities. for meeting rural demand for energy through market mechanisms and sound policy. This goes Cooking Energy hand in hand with the development of competent BiomasscontinuestoplayacriticalroleinBangladesh's implementing institutions, which are critical to the rural energy balance; today, it is just as important, if process. Also important are new supply- and demand- not more so, than 25 years ago. Unlike some other side technologies that can be used to raise rural South Asian countries, where liquefied petroleum gas people's welfare and improve productivity to increase (LPG) and other modern fuels have begun to enter the growth prospects. Accordingly, the rural energy marketplace, rural Bangladesh continues to depend strategy advocated by this study aims to satisfy the heavily on biomass fuels (fuelwood, cow dung, crop types of demand that increase household welfare residues, and tree leaves and grass) to meet household and raise rural growth prospects as energy becomes cooking needs (see Figure 1). a direct input into the production process. Some95percentofBangladeshihouseholdscollect or purchase biomass energy with which to cook all or Benefits of Moving up the Energy part of their meals, mainly using fixed clay stoves. The Ladder inherent inefficiency of such stoves, combined with the high moisture content of biomass cooking fuels, For Bangladesh to grow and prosper, its rural results in incomplete combustion, producing IAP. In economy must not be ignored. Focusing solely rural Bangladesh, age-sex composition of households Figure ES.1 Figure ES.1 Energy Use and Expenditures in Rural Bangladesh, 2004 % Average Energy Use % Average Energy Expenditures Electricity (Including Imputed Collection Values) Kerosene 1% Others Others Non-grid 2% 0.3% 1% electricity Grid 3% electricity 10% Dung 17% Firewood Firewood Kerosene 38% 44% 12% Crop residue Dung 21% 14% Crop Tree Tree residue leaves leaves 13% 9% 15% Source: BIDS Survey, 2004. xx 5580-CHFM.pdf xx 5580-CHFM.pdf xx 3/11/09 10:57:34 AM 3/11/09 10:57:34 AM Executive Summary indicates that generally, women and their children also shows a distinct trend in lighting energy use; as suffer greatest exposure. Women also bear the brunt of incomes rise, households increasingly switch from time spent collecting biomass fuels, accounting for an kerosene- to electricity-based lighting. The potential average of 150 out of 200 household hours per year. gain from this transition, as measured by consumer's Switching to improved biomass cookstoves and surplus, is enormous--40 to 45 percent of household modern cooking fuels can enable rural families, income. A separate quantitative analysis shows net especially women, to improve their quality of life. positive gains in household income from lighting Because improved stoves require about 25 percent less energy use of similar levels, confirming that lighting fuel than traditional ones, using them can similarly has a high value for rural households. reduce the amount of time spent collecting biomass fuels or the money used to purchase them. Cooking Farming Energy with modern petroleum fuels, such as LPG, entails A third critical energy use for rural Bangladeshi similar welfare gains. In addition to time and cost households is farming. The country's potential for savings, cooking with cleaner fuels avoids the health increased irrigation, though not as developed as risks associated with IAP. that of other South Asian countries, is significant. The most obvious energy inputs are diesel engines Lighting Energy and electric motors for pumping irrigation water. In In rural Bangladesh, some 70 percent of lighting addition, various small machines, including power energy is derived from kerosene, while electricity tillers, are used increasingly in land preparation, accounts for most of the other 30 percent. Households alleviating much of the drudgery associated with such with electricity prefer electric lighting to kerosene, but work. This study concluded that farm households that unreliable supply drives them to use kerosene lamps move up the energy ladder enjoy a substantial gain as a backup source in case of power failure. in productivity. For households that use mechanized In terms of cost and kilolumen output, electricity instead of manual irrigation, farm productivity is 15 is far superior to kerosene. Electricity's output per unit to 23 percent higher. Econometric analysis shows of energy is 100 times higher than that of kerosene. that farm income increases as much as 6.3 percent When one compares households that use kerosene from diesel use and 7.9 percent from electricity use lamps or electric lighting, two trends are obvious. in irrigation. First, the level of lighting increases with income for The study's village survey found that 68 percent both kerosene and electricity. Second, the quality of of villages used mechanized irrigation methods; of lighting service obtained from electricity is an order these, 70 percent used diesel power, 10 percent used of magnitude higher than that from kerosene for all electricity, and 3 percent used both. One possible income groups across all regions. For the average reason for farmers not having adopted electric household, the number of kilolumens is nearly 80 pumps to the extent that they have diesel ones is times higher for those who use mainly electricity, that 34 percent of villages remain without electricity. versus kerosene. Moreover, even in villages with electricity, the The high-quality lighting afforded by electricity potential gain of electricity (7.9 percent) over diesel has important consequences for Bangladeshi (6.3 percent) may not suffice to warrant large-scale household welfare. A household's electricity substitution; that is, both diesel and electric pumps consumption for lighting appears to rise steadily as result in similar levels of improved farm productivity. the number of students in the household increases. But electricity can be used in tube wells, a common The study's findings show that total hours of study practice in other South Asian countries; while diesel time are greater in households that use electric engines are generally limited to surface wells. Thus, in lighting than in those that use kerosene lamps, rural Bangladesh, the unrealized potential application confirming electricity's contribution to human-capital of electricity is widespread, presumably for reasons formation. Apart from education, household income of policy. xxi 5580-CHFM.pdf xxi 5580-CHFM.pdf xxi 3/11/09 10:57:35 AM 3/11/09 10:57:35 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Microenterprise Energy and Productivity adopted electricity, although 80 percent of villages had Switching to modern energy and the modern use of some type of electricity service available. biomass also contributes to the productivity of rural If electricity is available, household access microenterprises,representingasubstantialincreasein is determined by income: the higher the income rural residents' income. The three types of enterprises (approximated by land and non-land assets), the examined in this study--growth-center, village, and greater the demand. With higher incomes, rural home-based--use a wide array of energy sources. All Bangladeshis were found to pursue improved sources rural businesses in the growth centers use electricity, of energy and invest in assets that require more either from the grid or local generators; and most have energy. Higher-income rural households typically electric lighting. By contrast, not all free-standing diversify their energy portfolios, moving toward village and home-based enterprises have access to modern energy, which is not only more efficient but electricity; those that do may choose not to use it. also allows household members to invest time and Such enterprises depend heavily on biomass energy, money in more productive activities than biomass mainly for heating and manufacturing needs. Though energy alone would make possible. inconvenient, biomass is the least expensive heating For any energy source, pricing also determines fuel; also, most rural areas lack access to LPG. demand. Thus, energy access is not equitable because Increasingly, Bangladeshi entrepreneurs perceive the source is either priced too high or carries a high the health problems associated with traditional opportunity cost (e.g., biomass collection). Inequitable fuelwood use, as well as the growing scarcity of local access and level of use may lead to further inequities biomass. They also view traditional lighting sources (e.g., low-income households may pay far higher as inefficient and of poor quality, but unreliability prices for equivalent lighting services). of electricity supply forces businesses to depend on kerosene and diesel as backup lighting sources. Analysis of the relationship between modern Paying More Attention to Rural Energy energy sources, as defined by the energy ladder Services concept, indicates that rural businesses that use Rural energy must be viewed as a basic input to modern energy generally have more revenue and the rural economy, in line with its role in rural are more profitable than those that rely on traditional productivity and income generation. Achieving a energy sources. Thus, a viable rural energy strategy supply-and-demand balance is critical at all levels. for Bangladesh must consider the best ways to Given rural residents' heavy reliance on traditional promote modern energy, as well as the modern use biomass supply, biomass must be used in more of biomass. efficient ways that mitigate damage to human and global environmental health. Complementing these Energy Access efforts, more diversified modern energy sources must be made available to fulfill unmet rural household and In rural Bangladesh, access to energy is governed business demand. To this end, appropriate pricing by its availability, pricing (both monetary and policies are vital: they must be market-based to ensure non-monetary), household income, and other that suppliers can sustain the higher cost of rural characteristics. One major reason for the prominence of kerosene for lighting is that electricity is not operations, yet remain affordable to the poor. available in all areas. Even where it is available, consumer density is low. Thus, while 66 percent of the Biomass: Increasing Use Efficiency and villages sampled in this study had a grid connection, Supply only 29 percent of households were connected. Even in Mitigating the ill effects of biomass burning requires the capital division of Dhaka, where one would expect both demand- and supply-side solutions adapted a higher density, less than 50 percent of households to local realities. Biomass demand can be reduced xxii 5580-CHFM.pdf xxii 5580-CHFM.pdf xxii 3/11/09 10:57:35 AM 3/11/09 10:57:35 AM Executive Summary by adopting improved cooking stoves that rural to reach the entire rural population because, as the people want and harnessing biogas technology electricity grid expands, it becomes increasingly through village-based production and distribution expensive to serve remote households. Thus, networks. Biomass supply can be increased by the government and policy makers, along with planting trees around homes, maintaining and internationaldonorsupport,shouldcontinueandeven improving local natural resources, and increasing increase the successful expansion program financed agricultural productivity. Promoting a sustainable through the REB. Concurrently, off-grid systems can supply-demand balance implies the need for policy complement this program to meet the enormous initiatives that encourage intermediate and end-use unmet need for electricity. In 2002, the Infrastructure efficiency, especially at the household level; increased Development Company Limited (IDCOL) was made productivity to keep pace with growing demand; and the country's focal agency for coordinating the off- fuel substitution, where possible. grid program, the first phase of which has focused on promoting solar photovoltaic (PV) systems. To Rural Electrification: Toward a Pro-poor date, most work has been done by non-governmental Approach organizations (NGOs) specialized in microfinance Both grid and off-grid electrification programs are and microenterprise development. Working with 16 critical to the socioeconomic development of rural partner organizations, including Grameen Shakti Bangladesh. As this study shows, electrification and other NGOs involved in microcredit and solar translates into substantial gains in household welfare home systems (SHS), the program has succeeded in and a higher quality of life. As electricity demand installing more than 80,000 systems over a three-year is projected to double over the next nine years, the period. It is also important to diversify product lines investment required to generate sufficient capacity to to micro-grids, which could be connected directly to accommodate future demand growth is immense. the national grid system once the PBSs reach more If electricity is available, households will pursue remote communities. its use. But nationwide access to grid electricity is only 40 percent; in rural areas, it is just 30 percent. Household Petroleum Fuels: Toward The Rural Electrification Board (REB), responsible Equitable Use for electrifying rural Bangladesh, is based on the U.S. For the past 30 years, the Bangladesh Petroleum model of consumer-owned rural electric cooperatives. Corporation (BPC) and its subsidiaries have controlled Today, rural Bangladesh is divided into 70 cooperatives most aspects of petroleum supply, including its or Palli Bidyut Samities (PBSs), benefiting an estimated pricing system. Kerosene is priced uniformly across 7 million households--a remarkable achievement the country, but transport-cost adjustments are made in a country of some 22 million rural households for market distances greater than 40 kilometers from and about 110 million rural residents. Although the a supply depot. The price difference between rural PBSs are generally well managed, the program has and urban areas is not great, and the system for a nationwide electricity reliability problem. Some containing kerosene within a price range works well 80percentofruralhouseholdsreportdailyoutages,while throughout the country. By contrast, LPG, which is 60percentreportsignificantpowerfluctuations.Survey widely available in urban and peri-urban markets, results bring into question the REB's selected coverage has largely failed to reach rural households. By approach, suggesting a too-stringent connection policy switching from biomass to LPG to meet a portion of or other barriers that prohibit many poor households their cooking needs, rural family members, especially from receiving electricity directly. In addition, most of women, could realize large time savings and better the PBSs are not yet financially viable. health. Finally, piped natural gas (methane) is used Given that only 3 percent of Bangladesh's mainly by householders in Dhaka and other large 22 million rural households are gaining access to the towns. But under current pricing policy, households national grid each year, it could take several decades are charged a (subsidized) flat monthly rate, xxiii 5580-CHFM.pdf xxiii 5580-CHFM.pdf xxiii 3/11/09 10:57:35 AM 3/11/09 10:57:35 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities irrespective of the amount of gas consumed. Such a gas, it is recommended that the government reverse its policy leads to abuse and waste and discriminates flat-rate tariff so that households are charged per unit against households without connections in smaller of consumption and that subsidies, which invariably towns and rural areas. benefit wealthier households, be removed. Importance of Sound Pricing and Institutional Challenge: The Way Subsidy Policies Forward Promoting more equitable rural access and use of Rural energy is a complex issue, encompassing a broad modern energy implies the need for sound pricing and diverse spectrum of resources--from household policies to increase supply and reliability. It should be petroleum fuels to biomass and renewable energy-- noted that the urban areas of Bangladesh are served spanningmultiplesectors,includingforestry,electricity, by a state utility, which is generally considered to be less efficient than the rural cooperatives electricity and health. To date, the many diverse institutions companies. Even though performing better, the rural addressing rural energy issues in Bangladesh have electricity policies and service still can be improved. been poorly coordinated. One major recommendation For rural grid electricity, a careful review of the resulting from this study is to develop the long-term policies of uniform subsidies and similar tariffs for institutional capacity to tackle rural energy issues in all PBSs is recommended. One potential solution all their complexity. Such an institution would promote might include mechanisms that give preferential rural energy solutions through its technical assistance, treatment to companies serving poorer regions, such advice to government, and facilitation of grants and as permitting higher tariffs or providing electricity at loans for worthy projects. lower bulk prices. To date, off-grid electricity schemes Rural energy's importance to the Bangladesh have succeeded by providing subsidies through economy cannot be underestimated, given the individual projects, allowing prices to float based on world's focus on globalization and market reform. the cost of service after subsidies.2 The problems rural people face in obtaining safe, As the previous discussion suggests, pricing clean, and reliable energy supplies are not minor policies for household petroleum fuels also require inconveniences. On the contrary, they represent a careful review. Kerosene, used by most rural significant barrier to rural economic development and households for lighting, is accessible to most improved social well being. A multifaceted approach consumers at world-market rates, and is available to solving Bangladesh's rural energy problems is an in quantities sufficient for cooking. Thus, its pricing essential building block to propel the country into the policies require no substantial revision. For LPG, it twenty-first century. is recommended that the upfront costs of needed stoves and cylinders be partially subsidized or paid The rural energy strategy identified is this: for in installments, which would put LPG within Bangladesh has a comprehensive need for better reach of households that otherwise could not afford institutional coordination and attainment of a critical it. Developing and making available smaller LPG mass of technology and market development. With cylinders is occurring in Bangladesh, and should be effective institutional coordination, combined with encouraged even further. This would help extend the market development, appropriate subsidy, pricing reach of LPG into less wealthy markets. In addition, policies, and government and donor support, a level playing field should be promoted for all current and proposed programs can succeed beyond public and private companies that market LPG to expectations. The call for action is urgent, not only for encourage innovation and expansion into more rural development, but for the country's equitable and remote, underserved areas. With regard to natural overall economic growth. 2Through the World Bank­financed Rural Electrification and Renewable Energy Development Project (RERED), IDCOL has worked with selected local partners to provide rural consumers a 20 percent subsidy for SHS using a microfinancing scheme. xxiv 5580-CHFM.pdf xxiv 5580-CHFM.pdf xxiv 3/11/09 10:57:35 AM 3/11/09 10:57:35 AM 1 Introduction Bangladesh is one of the world's poorest countries.3 electricity for lighting, agricultural processing, and Nearly 80 percent of the nation's 140 million people other productive activities. reside in rural areas; of these, 20 percent live in Gradually, the types of energy used in rural areas extreme poverty. Geographically, many low-lying are improving. Diesel-powered irrigation pumps and areas are vulnerable to severe flooding, while other agricultural tillers are more common than in the past, regions are prone to drought, erosion, and soil when manual and animal powers were the traditional salinity. Such an unfavorable agricultural landscape, energysourcesforcultivation.Grid-basedelectrification combined with mismanagement of natural resources isreachingmoreruralhouseholdsatafasterpace.Recent and increasing population pressure, is pushing many projectsinrenewabletechnologies,includingsolarhome of the rural poor to the brink. systems (SHS), reflect slow, yet significant, progress. In Because Bangladesh is such a poor country, it sum, while the traditional biomass system continues to also is one of the world's lowest energy producers. predominate, rural Bangladesh is moving slowly, yet Total annual energy supply is only about 150 liters inexorably, toward modern forms of energy and more of oil equivalent per capita (IEA 2003); in rural efficient use of biomass as options for energy supply. areas, conditions are even worse. Compared to other developing countries, Bangladesh uses little modern energy. Despite its successful rural electrification What Is the Knowledge Base? program, close to two-thirds of households remain In the mid-1970s, Bangladesh conducted its first large- without electricity and, with the exception of scaleenergystudy,whichwasfollowedintheearly1980s kerosene, commercial fuels are beyond reach for by a more comprehensive study that included a rural many. Moreover, biomass fuels are becoming energy component (GOB 1987). Now, more than two increasingly scarce. Collected mainly from the local decadeslater,littlenewinformation,withtheexception environment as recently as two decades ago, biofuels of small rural energy surveys, is available. Indeed, are fast becoming a marketed commodity as access to knowledge of rural energy in Bangladesh lags recent local biomass continues to shrink.4 developments in energy technologies and policies. Today, people in rural Bangladesh use a mosaic Studies on rural electrification and development of energy sources to meet their various domestic and have been conducted over the past decade; but many productivity needs. Biomass is still used extensively new issues remain unexplored. Until recently, the for cooking, while kerosene is the main lighting potential for renewable energy and the adverse health source. Electrified households prefer electricity effects of indoor air pollution (IAP) on women and for both lighting and appliance use. Many small children were ignored. Research on the structure businesses, including home-based enterprises, use and efficiency of rural energy markets has also been 3The World Bank estimates that, in 2005, 28 percent of the urban population and 44 percent of rural residents were below the poverty line, with a nationwide poverty rate of 40 percent (World Bank 2006). 4In 2005, the FAO Forestry Department reported that only 6.69 percent of Bangladesh's total land area (13.02 million hectares) remained under forest cover. 1 5580-CH01.pdf 1 5580-CH01.pdf 3/11/09 10:54:02 AM 3/11/09 10:54:02 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities overlooked. Policy analysis and recommendations surveys--onefocusedonruralhouseholdsandtheother for rural energy development, along with their onvillagemicroenterprisesandruralgrowthcenters--to implications for the institutions that implement them, elicit information on energy-using behavior and have not been assigned a high priority. characteristics. Subsequently, the Bank commissioned With the exception of rural electrification, most studies on the market structure for energy and the intervention programs have emphasized commercial macro-leveldimensionsofbiomasssupplyanddemand energy, consumed mainly in urban areas. At the same (Openshaw 2004; Khalequzzaman 2005). time, the rural electrification program has continued Thisstudy--thefirsttoconcentrateonBangladesh's to expand.5 A forestry master plan has been approved energy systems and their effects on the lives of rural for implementation, and various community and people--drewonthesebackgroundstudies,aswellas agroforestry projects have been implemented. The other World Bank­financed research on IAP and rural country has renewed its emphasis on tree planting infrastructure,topresentaruralenergystrategyforthe and management as part of activities to mitigate country. Much of this study's analytical underpinning climate change. To curb the health hazards of IAP, was based on several background studies. This study both the government and donors are supporting also reanalyzed data from earlier research to better efforts to popularize improved biomass stoves. Yet the understandthebenefitsofmodernenergyuseforrural effectiveness of these initiatives, in terms of people's households, farm activities, and small businesses. overall energy-using behavior, remains under the Thisstudy'sbroadaimwastoidentifywaystoraise radar screen of many development researchers. the living standard in rural Bangladesh via better and One of the country's major economic policies is to moreefficientenergyuse,whilecreatinganenvironment promote agriculture and rural development. Indeed, conducivetogrowthandpovertyreduction.Tothisend, food production and consumption are inextricably the study sought to do the following: linked to rural energy. But this reality--recognizing · Describetheruralenergysituationforhouseholds the energy needs for various rural economic and and small businesses. development activities--appears as an afterthought in · Determinethedemandforandbenefitsofmoving many government documents. Such policy documents up the energy ladder. equateenergywithelectricity,gas,andhydrocarbons.6But · Describe the energy-supply situation and identify the main energy uses in rural areas involve fuelwood, bottlenecks to improved supply. cow dung, and crop residue; rural Bangladeshis use · Recommend policies to help rural regions move these fuels in large quantities to cook their food, heat toward improved productivity and higher quality their homes (in cooler regions), produce bricks in kilns, of life. and make molasses (gur) from sugar cane. Based on these analyses, the study put forward Study Goal and Objectives a set of recommendations as part of a rural energy strategy for Bangladesh. Lackofcomprehensivedataandanalysisandknowledge gapsregardingBangladesh'scurrentruralenergystatus haveprecludedthedevelopmentofastrategyformodern Principles for a Rural Energy Strategy and efficient energy use. Against this backdrop, the The rural energy strategy of any developing country WorldBank,throughaconsultativeprocess,initiatedtwo must be guided by basic principles. First, because the 5A major study has been conducted on the effects of rural electrification on development (Barkat et al. 2002); at the time of this writing, a monitoring and evaluation study, involving a large survey, was under way. 6In various government documents, the energy chapter discusses electric power, gas, coal, and liquefied petroleum gas (LPG); while rural energy is relegated to the forestry section of the agriculture chapter. The National Energy Policy (GOB 1996), a decade-old document, reflects little on the importance of rural energy. Similarly, the government's recent poverty reduction strategy makes little mention of rural energy (GOB 2005). 2 5580-CH01.pdf 2 5580-CH01.pdf 3/11/09 10:54:03 AM 3/11/09 10:54:03 AM 1 Introduction types of energy services that people demand may be for non-residential process heat. Both households obtained from more than one source, issues of cost- and rural enterprises require lighting, space cooling effectiveness and satisfying people's preferences arise; and heating, and motive power (among productive notallservicesmaybeequallydesirablebecauseoftheir activities, crop cultivation generates demand for priceoraestheticorpracticalvalue.Second,becauseany motive power for irrigation pumps, tillers, and particular fuel may have a variety of uses, issues of fuel transport) as illustrated in Figure 1.1. substitution and complementarity also are important. The crux of a rural energy strategy is to have more Household demand for energy primarily involves andbetterchoicesformeetingruraldemandforenergy cooking, lighting, heating, productive uses, and other throughthemarketmechanismandsoundpolicy.This end uses. Rural enterprises have a significant demand goeshandinhandwiththedevelopmentofcompetent Figure 1.1 Figure 1.1 Types and Sources of Energy Demand Home, village, Tree biomass conventional forest Domestic cooking, parboiling Animal dung Livestock Agricultural Agriculture Residues Bio-gas Lighting Domestic mines ENERGY Kerosene DEMAND Nonresidential, Import process heat Coal Space heating or Gas, LPG, CNG Domestic gas fields cooling Import Petroleum fuels Motive power/traction/ transport/pump Grid, PVs, batteries Electricity generator Amusement/leisure Draught power Source: Bangladesh Institute of Development Studies (BIDS) (2005). 3 5580-CH01.pdf 3 5580-CH01.pdf 3/11/09 10:54:03 AM 3/11/09 10:54:03 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities implementing institutions, which are critical to the forestry situation, relevant institutions and policies, process.Alsoimportantarenewsupply- anddemand- and delivery mechanisms for several fuel types in sidetechnologiesthatcanbeusedtoraiseruralpeople's rural areas. Various other secondary materials were welfare and improve productivity to increase growth used, including an earlier World Bank report on IAP prospects. Accordingly, the rural energy strategy in Bangladesh (Dasgupta et al. 2004). advocated by this study aims to satisfy the types of demand that increase household welfare and raise rural growth prospects as energy becomes a direct Survey Instruments input into the production process (see Box 1.1). This study was based on information from a comprehensive survey, whose objectives were to Study Method determine the overall energy-use patterns, potential market for modern energy in rural areas, willingness This study was based mainly on the results of to pay for electricity services, and barriers to adopting two large-scale primary surveys, one on rural renewable energy.8 To meet these objectives, and households and the other on rural growth centers and thereby assist in the development of a rural energy microenterprises defined in the Bangladesh context strategy, these surveys were conducted: as large marketplaces. Because of their importance, these surveys, along with background information, · Household survey. This study determined the are described in greater detail in this chapter.7 socioeconomic characteristics, energy demand Furthermore, selected information from the surveys andavailability,consumerabilityandwillingness was reanalyzed to address the welfare and growth to pay, attitudes toward various energy sources, effects of modern energy use. and perceived benefits of energy. The study also drew on reports prepared · Growth center and microenterprise survey. This specifically to assess Bangladesh's biomass and study developed village-level profiles, including characteristics and potential energy demand, and Box 1.1 Box 1.1 What Are the Benefits of Rural Energy? For Bangladesh to grow and prosper, its rural economy must not be ignored. Focusing solely on urban growth-- representative of only about 20 percent of the country's 140 million people--would lead to inequitable social and economic development. Making modern energy services and jobs available in rural areas can help rural residents become more productive, thereby mitigating urban­rural disparities and reducing rural-to-urban migration. The benefits of rural energy range from increased time savings and farm productivity to improved education, communication, and overall quality of life. For example, electric lights emit more than 100 times the amount of lighting provided by a kerosene lantern or candle, permitting household members--both adults and children--to read and study during evening hours. Electric irrigation pumps make it possible to grow and market crops year-round and raise rural incomes. Electric lighting and motive power can increase the productivity of women-owned microenterprises. Availability of radios and televisions can improve communication and information sharing. Use of well-designed stoves and efficient cooking fuels can reduce IAP and the number of hours that families, especially women and children, spend collecting biomass. Taken together, these benefits can lead to greater rural productivity and a higher quality of life for many of the country's poorest citizens. Source: Barnes and Floor (1996). 7For the basic technical analysis of these surveys, see Asaduzzaman and Latif (2005). 8A portable document format (PDF) file of survey modules and questionnaires is available from ESMAP upon request. 4 5580-CH01.pdf 4 5580-CH01.pdf 3/11/09 10:54:03 AM 3/11/09 10:54:03 AM 1 Introduction assessed the energy used by small- and medium- Household Survey sized businesses located in village marketplaces, The household survey, conducted in 2004, was based known as growth centers. on a cluster sampling strategy. Nationwide, a total · Because 25 years had passed since Bangladesh of 40 rural subdistricts (known as upazilas or thanas) had conducted such a major rural energy survey, were randomly selected from four older divisions the present one used representative samples of (proportionate to population). This was followed rural households and commercial businesses (see by the random selection of three villages from each Table A1.1, Annex 1). subdistrict (see Figure 1.2). Figure 1.2 Figure 1.2 Map of Survey Areas in Bangladesh Source: This map was prepared by the World Bank Map Design Unit (2008). Shaded areas are sample upazilas (political units) from the Bangladesh Rural Energy Survey (BIDS, 2005). 5 5580-CH01.pdf 5 5580-CH01.pdf 3/11/09 10:54:03 AM 3/11/09 10:54:03 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Inordertodevelopavillagesampleframe,acensus Theseruralenergysurveyresultscomparefavorably was conducted in each village. This census covered with those of the Bangladesh Household Income and all households; village enterprises based outside the Expenditure Survey (HIES), conducted in 2005. For home but not located at the growth center; and all both surveys, the percentage of rural households with educational,religious,andhealthinstitutions.Samples electricity was close to 30 percent. There were slight weretakenofhouseholds,enterprises,andinstitutions differencesinfuelexpenditures.Accordingtothisrural outside the home for further in-depth data collection energysurvey,theaverageannualexpenditureforgrid (see Table 1.1). In addition, a community survey was electricity was Tk 488 and Tk 608 for kerosene. Results conducted to identify village characteristics. The from the 2005 HIES showed that rural households' resulting sample closely resembled Bangladesh's average annual expenditure for grid electricity was overall rural population. The survey revealed that Tk 522 and Tk 566 for kerosene. The results are close, about88percentofhouseholdsareheadedbymen26to given the different methods used to measure these 45yearsofage,nearly50percentofwhomareilliterate. expenditures. These findings indicate that rural Theaveragehouseholdsizeis5.Nearly55percentown householdshaveincreasedtheirelectricityconsumption littleornoland(0.5acresmaximum);only6to7percent while lowering their dependency on kerosene. ownmorethan5acres.About30percentofhouseholds own little more than their homestead. In 2004, annual Growth Center and householdincomewasroughlyUS$1,000(Tk62,000)or Microenterprise Survey US$200perperson.Some43percentofthisamountwas Because most of the country's rural microenterprises derivedfromagriculturalsources(e.g.,cropcultivation, are located in growth centers, a separate survey non-crop agriculture, and agricultural wages), while of microenterprises in rural growth centers was theremainderwasfromnon-agriculturalsources(e.g., conducted to obtain a representative sample of processingandtradingactivitiesandremittances).All enterprises located outside the home. ofthesefigurescloselycorrespondtoequivalentfigures Given that no prior information existed on for rural Bangladesh. enterprise distribution in the growth centers, a two- In the case of village-based enterprises, several phase random sampling technique was adopted. had more than one distinct business activity because The first phase involved the random selection of theywereownedoroperatedbyhouseholdsthat,when 40 subdistricts (from a list of 460 subdistricts, covering interviewed for the household survey, were found to six divisions), from which three growth centers (per have home-based enterprises. These 137 enterprises subdistrict) were chosen. The second phase involved a were treated separately. Table 1.1 able 1.1 Number of Entities in the Census and Sample by Type and Division Census Type Chittagong Dhaka Khulna Rajshahi Total Household 5,178 5,928 5,508 5,611 22,225 Business enterprise 342 487 659 394 1,882 Institution 201 223 249 184 857 Total 5,721 6,638 6,416 6,189 24,964 Survey Type Household 640 603 548 600 2,391 Home enterprise 31 50 29 27 137 Other enterprise 63 85 136 58 342 Institution 36 38 50 32 156 Source: Asaduzzaman and Latif (2005). 6 5580-CH01.pdf 6 5580-CH01.pdf 3/11/09 10:54:06 AM 3/11/09 10:54:06 AM 1 Introduction sample survey of business establishments. From each Structure of This Report growth center, 15 or more enterprises were randomly selected. The first step in the selection process entailed The structure of this report reflects the directional a census of the enterprises, which provided the sample organization of the study. Chapter 2 presents an frame for enterprise selection in each growth center. overview of the rural energy situation of Bangladeshi Becauseofthesmallnumberofgrowthcentersinsome households, moving up the energy ladder--from subdistricts,thenumberofgrowthcenterssampledwas biomass cooking energy to kerosene and electric 115insteadofthe120targetedforthestudy.Asaresult, lighting. Chapter 3 offers an in-depth analysis of forseverallocations,morethantheminimumrequired household energy demand, examining expenditure sample was covered. The total number of enterprises patterns for cooking, lighting, and appliance studied from the 115 growth centers was 1,801. Dhaka, ownership. It presents the results of an econometric Chittagong, and Rajshahi accounted for about 75 analysis to predict how household energy demand percent of regional distribution; while Khulna, Sylhet, is affected by changes in such factors as household and Barisal constituted the remainder (see Table 1.2). income and energy pricing. Chapter 4 provides a descriptive analysis of the household welfare gains that can result from transitioning up the energy Implications for a Rural Energy Strategy ladder, including better health from use of improved The two surveys described here represent the biomass stoves and improved education from electric first comprehensive attempt to assess the rural lighting.Forfarmhouseholds,makingtheswitchfrom energy needs of and benefits for Bangladesh's rural manual to diesel- and electric-powered equipment households and microenterprises. Along with a is examined in terms of gains in agricultural variety of other materials, the surveys provide the productivity. Chapter 5 explores the effects of modern analytical underpinnings of this study--the first energy on rural production, focusing on the energy- in 25 years to examine Bangladesh's complex rural usingcharacteristicsofmicroenterprisesandfarming. energy situation. By covering both households Chapter6turnstothecurrentinstitutionalframework and microenterprises, the study has allowed us to for rural energy supply and implications for meeting examine rural energy-use patterns and the adequacy rural energy demand. It focuses mainly on delivery of the country's current rural energy policies and mechanisms of the biomass, power, oil and gas, and institutional framework. The study moves toward renewable energy (non-biomass) sectors. Finally, broad recommendations that can support a higher Chapter 7 suggests a broad institutional strategy to quality of life for rural residents and, thus, more meet the complex challenges of Bangladesh's current equitable economic growth for the country. and future rural energy needs. Table 1.2 able 1.2 Coverage of Growth Centers by Division Growth Centers Shops Division Total No. Sampled Total in Growth Centers No. Sampled Dhaka 531 29 101,256 477 Chittagong 373 25 64,022 403 Rajshahi 543 31 91,592 456 Khulna 274 15 41,685 226 Sylhet 157 3 20,986 48 Barisal 180 12 34,635 191 Bangladesh 2,058 115 354,175 1,801 Source: Data International (2004). 7 5580-CH01.pdf 7 5580-CH01.pdf 3/11/09 10:54:06 AM 3/11/09 10:54:06 AM 5580-CH01.pdf 8 5580-CH01.pdf 3/11/09 10:54:06 AM 3/11/09 10:54:06 AM 2 Household Energy Use Unlike South Asian regions where liquefied petroleum General Consumption Patterns gas (LPG) and other modern fuels have entered the marketplace, rural Bangladesh still depends heavily Nearly all households use both biomass and non- on fuelwood--as well as dung, crop residue, and even biomass energy (see Table 2.1). A vast majority use tree leaves--for cooking fuel. Indeed, in many areas of biomass sources, including fuelwood, tree leaves, the country, biomass energy is in short supply. At the and crop residue. While kerosene is the predominant same time, the importance of kerosene, electricity, and non-biomass energy source, many households also LPG is growing. This chapter provides a snapshot of use grid electricity and dry-cell batteries. rural Bangladeshi households' energy-use situation, Biomass is used almost exclusively for cooking focusing on general consumption patterns and those (see Table 2.2). Fuelwood constitutes 41 percent of for the key energy uses of cooking and lighting. total biomass cooking energy. Surprisingly, 39 percent Table 2.1 able 2.1 Household Distribution of Energy Sources by Division (percent) Division Energy Source Chittagong Dhaka Khulna Rajshahi All Divisions Biomass 98.6 99.8 100.0 99.5 99.5 Fuelwood 95.8 85.1 88.7 67.2 84.3 Tree leaves 61.6 81.4 91.8 72.0 76.1 Crop residue 53.6 81.4 75.4 93.2 75.5 Dung cake/stick 29.5 56.9 64.8 72.3 55.2 Sawdust 0.3 0.7 1.6 0.8 0.8 Non-biomass 100.0 100.0 100.0 96.5 99.1 Kerosene 98.4 99.0 100.0 91.5 97.2 Grid electricity 38.9 43.8 10.8 20.3 29.0 Dry-cell battery 42.3 39.1 61.3 50.7 48.0 Candle 10.6 2.5 4.4 0.8 4.7 LPG/LNG 1.1 -- -- 0.2 0.3 Natural gas 0.9 -- -- -- 0.3 Storage cell 0.8 1.3 0.9 -- 0.8 Solar PV -- 0.3 1.5 -- 0.4 Source: BIDS Survey (2004). Note: In addition to the main authors, M. Abdul Latif was a contributing coauthor of this chapter. 9 5580-CH02.pdf 9 5580-CH02.pdf 3/11/09 10:54:24 AM 3/11/09 10:54:24 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 2.2 able 2.2 Annual Household Consumption (physical quantity) Energy Use Heating Energy Source Cooking Parboiling Other Cooling Lighting Amusement All Uses Biomass (kg) Fuelwood 1,064.84 28.60 92.77 -- -- -- 1,186.21 Tree leaves 470.67 29.99 0.85 -- -- -- 501.51 Crop residue 538.86 164.41 2.72 -- -- -- 708.18 Dung cake/ stick 503.68 16.07 4.16 -- -- -- 523.90 Sawdust 8.36 0.02 0.02 -- -- -- 8.40 Non-biomass Kerosene (liter) 1.76 -- 0.07 -- 27.16 -- 28.98 Grid electricity (kWh) 0.25 -- 4.00 49.50 80.74 9.34 143.83 Dry-cell battery (piece) -- -- -- -- -- -- 15.01 Candle (piece) -- -- -- -- 15.86 -- 15.86 LPG/LNG (liter) 0.05 -- -- -- -- -- 0.05 Natural gas (Tk) 9.59 -- -- -- -- -- 9.59 Storage cell (kWh) -- -- -- -- 0.14 0.41 0.55 Source: BIDS Survey (2004). is represented by agricultural residue or leaves and grid system. Nearly 30 percent of rural households grass, indicating a shortage of fuelwood energy.9 use electricity provided by the PBSs, and the annual Kerosene is used primarily for lighting; grid percentage continues to climb. This figure compares electricity is another important lighting source. well with that of the Bangladesh Household Income Monthly average consumption of kerosene is 2.25 and Expenditure Survey (HIES), conducted during liters, which barely covers basic lighting services, the same period, which was 30 percent. even considering that households with access to grid Fuelwood is also the single most important electricity use little kerosene for lighting. Batteries, an rural energy source in terms of energy unit (kgoe), expensive but extensively used source of electricity, accounting for some 44 percent of total consumption are convenient for powering flashlights, radios, and (see Figure 2.1). Including tree leaves and twigs, the other consumer electronics. Grid electricity is mostly share of tree-based biomass is nearly 60 percent of derived from the cooperatives or PBSs (Palli Bidyut total household energy. Crop residue (e.g., bagasse, Samities), which serve rural areas through the national jute sticks, rice hulls, bran; as well as various types of 9Such fuels as unprocessed agricultural residue and cow dung are considered inferior to wood for cooking and generally produce higher levels of IAP when burned in traditional stoves. 10 5580-CH02.pdf 10 5580-CH02.pdf 10 3/11/09 10:54:24 AM 3/11/09 10:54:24 AM 2 Household Energy Use straw and uprooted plant remains) and animal residue and the inefficiency of most rural stoves, the useful (e.g., cow dung made into round cakes or sticks and or delivered energy is lower than the percentages dried before burning) constitute other major sources. presented; however, they highlight the importance Not generally used for cooking, modern fuels account of biomass for cooking. for only 3 percent of the energy balance. Because of Estimated annual energy consumption by rural the significant amount of energy required by cooking households is 1,049 kgoe or 8.9 gigajoules (GJ) per person--a vast increase from the 5 GJ per person Figure 2.1 Figure 2.1 consumed 25 years ago. Over the same period, the Rural Household Consumption by Source annual growth rate was more than 2.6 percent, higher (percent energy consumption) than the average growth rate in per capita income. Indeed, when only biofuels are considered, growth Kerosene Electricity Other appears even more remarkable. Leach (1987), using 2% 1% 0% figures based on Islam (1980, 1986), reported that rural households used an estimated 4.2 GJ of biofuels. The Dung 17% corresponding figure for biomass in the current study's Firewood household survey is 8.6 GJ per person; this figure 44% translates into an average annual growth rate of 3.2 percent, outstripping even population growth rate.10 Crop This finding highlights the critical role that biomass residue continues to play in the rural energy balance; today it 21% is just as important, if not more so, than 25 years ago. Tree leaves 15% Energy for Cooking Source: BIDS Survey (2004). The energy required for basic household cooking-- Note: The "Electricity" category includes only grid-based includingtheparboilingofrice--mirrors,inlargepart,the electricity. Other types of electricity (solar PV, storage cells, and dry-cell batteries) were either nearly absent or could not country'soverallruralenergysituation;yetoneobserves be quantified; whatever could be quantified is included in the major regional differences (see Table 2.3). For example, "Other" category. in Khulna and Chittagong, regions home to major Table 2.3 able 2.3 Energy Sources for Cooking and Parboiling of Rice by Division (annual kgoe per household) Chittagong Dhaka Khulna Rajshahi All Divisions Energy Source C P C P C P C P C P Fuelwood 644.8 7.0 365.5 16.7 404.6 9.7 180.5 9.70 400.4 10.7 Tree leaves 120.3 2.6 157.6 12.1 218.0 19.3 116.2 5.40 149.7 9.5 Crop residue 117.4 27.0 159.4 36.9 183.1 114.7 269.4 37.70 171.3 52.3 Dung 79.4 0.8 155.8 7.6 220.6 4.5 251.4 9.20 171.7 5.5 Sawdust 1.4 0 1.0 0 5.8 0 2.8 0.02 2.6 0.01 Kerosene 1.8 0 2.8 0 0.6 0 0.6 0.00 1.4 0 Total 965.0 37.4 842.1 73.4 1,032.7 148.2 820.9 62.00 897.3 78.1 Source: BIDS Survey (2004). Note: C = cooking; P = parboiling. 10The increase may involve advances in survey design instruments with which to measure rural energy use (e.g., previous surveys had more basic questions). 11 5580-CH02.pdf 11 5580-CH02.pdf 11 3/11/09 10:54:24 AM 3/11/09 10:54:24 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities forests, consumption levels are higher. Indeed, nearly may be somewhat less in Chittagong. The little use 67 percent of Chittagong's cooking energy is derived made of improved biomass stoves that vent smoke to from fuelwood, indicating that plentiful availability the outdoors and modern fuels are important related leads to greater use. Not surprisingly, Rajshahi, the issues, which are discussed in Chapters 3 and 4. division least endowed with forest resources, depends least on tree-based biomass (fuelwood and leaves) for Commercialization of Biomass Energy household cooking. In Khulna, where one might expect A major finding of this study, as mentioned previously, higher consumption levels, strict forest regulations is the increasing commercialization of biomass energy and restricted access keep consumption lower than in in rural areas. Just two decades ago, most biomass fuels Chittagong,wherelocalresidentscanmoreeasilyaccess were collected from the local environment. Fuelwood, scattered forest patches. conventionally derived from rural residents' own Significant use of biomass energy--especially tree production or local collection, is today more likely to leaves, crop residue, and dung--means that indoor be purchased from local markets. Indeed, about 40 air pollution (IAP) and local biomass shortages are percent of all fuelwood is now purchased from local potential problems in rural areas of all four divisions. markets; in Dhaka, the proportion is nearly 67 percent Since fuelwood is a comparatively more efficient fuel (see Table 2.4). with a higher energy content, the potential for IAP Table 2.4 able 2.4 Distribution of Biomass Supply Sources by Energy Type Energy Type (%) Supply Source Fuelwood Tree Leaves Crop Residue Dung Cake/Stick Chittagong Own production 14.0 57.8 69.4 82.8 Gathered 61.9 41.2 28.8 3.6 Purchased 24.1 1.1 1.8 13.6 Dhaka Own production 6.3 55.5 72.2 23.2 Gathered 26.9 43.9 15.0 9.2 Purchased 66.8 0.6 12.8 67.6 Khulna Own production 7.6 55.5 66.6 92.5 Gathered 32.4 40.3 25.7 1.3 Purchased 60.0 4.2 7.8 6.2 Rajshahi Own production 7.3 41.0 67.5 25.0 Gathered 55.4 58.9 28.3 17.1 Purchased 37.3 0.1 4.2 57.9 All divisions Own production 11.0 52.2 68.5 72.2 Gathered 49.8 46.3 24.7 5.3 Purchased 39.3 1.4 6.8 22.5 Source: BIDS Survey (2004). 12 5580-CH02.pdf 12 5580-CH02.pdf 12 3/11/09 10:54:24 AM 3/11/09 10:54:24 AM 2 Household Energy Use Given that fuelwood constitutes the major portion A major reason for the prominence of kerosene ofenergyconsumption,itsincreasingcommoditization is that electricity is not available in all areas; even has major implications for energy costs, an issue where available, consumer density is low. Thus, addressed in Chapter 3. The marketplace is also while 66 percent of the villages sampled had a grid a major source of other biomass energy supplies, connection, only 29 percent of households were including dung cakes or sticks, particularly in Dhaka connected. Even in Dhaka, where one would expect and Rajshahi. As Table 2.4 indicates, more than half of a higher density, less than 50 percent of households tree leaves are still gathered, while most crop residue adopted electricity, although 80 percent of villages is derived from residents' own production. had some type of electricity service available (see Recent development of rural markets for biomass Table 2.6). energy reflects an increase in overall supply; but perhaps more important, a decrease in local supply. The implications for cooking fuels and their increasing scarcity in the local environment suggest the need for Box 2.1 Box 2.1 policy dialogue on the evolving status of cooking fuels Modern Energy Benefits for Rural Families in rural Bangladesh. Making the switch to electric lighting and appliances, petroleum fuels, and improved cooking stoves can Energy for Lighting enable rural families to raise their incomes and improve their quality of life. In a hot tropical climate such as In rural Bangladesh, some 70 percent of energy Bangladesh, the addition of a simple electric fan can consumed for lighting is derived from kerosene significantly improve a rural household's indoor comfort and most of the other 30 percent from electricity level and ward off insects. Electric lighting offers 100 (Box 2.1). Nearly 100 percent of households claim to use times more light than traditional kupi or kerosene lamps kerosene lamps. Although households with electricity commonly used in households without electricity. The prefer electric lighting to kerosene, unreliable supply higher quality of lighting makes it possible for families drives them to use kerosene lamps as a backup in case to pursue reading and other educational activities of power failure. While overall energy consumption during evening hours. Cooking with kerosene or for lighting is low, regional differences are substantial. LPG, or using improved biomass stoves--still a rare occurrence in rural Bangladesh--can result in fewer For example, rural households in Chittagong use hours spent collecting biofuels, less cooking time, more than twice as much energy as those in Rajshahi and reduced IAP. Chapter 4 considers these benefits (because of differences in kerosene and electricity in more detail. consumption). In Khulna, nearly 92 percent of lighting Source: Barnes and Floor (1996). is derived from kerosene (see Table 2.5). Table 2.5 able 2.5 Energy Sources for Lighting (annual kgoe per household) Energy Type Chittagong Dhaka Khulna Rajshahi All Divisions Kerosene 27.86 18.64 25.30 17.62 22.38 Grid electricity 11.82 9.70 1.86 3.29 6.86 Candle 2.60 0.09 0.30 0.02 0.79 Solar PV 0 0.04 0.13 0 0.04 Storage cell 0 0.05 0 0 0.01 All 42.28 28.52 27.59 20.94 30.09 Source: BIDS Survey (2004). 13 5580-CH02.pdf 13 5580-CH02.pdf 13 3/11/09 10:54:25 AM 3/11/09 10:54:25 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 2.6 able 2.6 Electrification of Villages and Households by Division Grid Connection (%) Division Villages Households Chittagong 78 39 Dhaka 80 44 Khulna 50 11 Rajshahi 53 20 All 66 29 Source: BIDS Survey (2004). Summing Up electric lighting if it were available. Electricity is not used as extensively as it might be, perhaps because of This chapter has raised an important question: Are restrictive household connection policies. rural Bangladeshis moving up or down the energy The energy-use situation of rural households ladder? In terms of household energy use, biomass implies the need for policy initiatives that promote in all its various forms is of paramount importance. more efficient use of biomass energy, and its increased For cooking, fuelwood is the predominant form of supply and more efficient use and better pricing of biomass used; but it is becoming scarcer than in the modern energy to increase supply and reliability. To past, and a significant portion is being purchased. gain a clearer picture of whether rural Bangladeshis At the same time, many rural residents are turning are moving up or down the energy ladder, it is critical to residues, including grass and leaves, to meet their to examine patterns of energy demand (expenditure daily cooking needs. and asset ownership), which are the subject of the In terms of lighting, kerosene is the main energy next chapter. source, although most rural households would prefer 14 5580-CH02.pdf 14 5580-CH02.pdf 14 3/11/09 10:54:25 AM 3/11/09 10:54:25 AM 3 Household Energy Demand In Bangladesh, access to modern energy is a Expenditure Patterns contentious issue, revolving around the role of energy markets. Although financially motivated Household energy expenditure depends on many rural shopkeepers should want to offer all types of factors, from the availability and pricing of goods and commercial fuels, government policies determine, services to household income. The following sections in part, which modern fuels are marketed in rural summarize the share of total rural energy expenditure areas. A major question is how markets can function by energy source and household income and region. to serve rural people. To determine the extent to which modern energy factors in market development, this By Energy Source chapter examines patterns of rural household energy As Figure 3.1 illustrates, rural residents still depend on expenditure and asset ownership. traditionalbiomass,particularlyfuelwood,tomeetmost Figure 3.1 Figure 3.1 Energy Expenditure Distribution by Energy Source (percent expenditure) Non-grid electricity 3% Other 1% Grid electricity 10% Kerosene Fuelwood 12% 38% Dung 14% Tree leaves Crop residue 9% 13% Source: BIDS Survey (2004). Note: Costs are actual or imputed for each energy source; consumer households must also bear the costs of procurement and supply. Note: In addition to the main authors, Hussain Samad was a contributing coauthor of this chapter. 15 5580-CH03.pdf 15 5580-CH03.pdf 15 3/11/09 10:54:43 AM 3/11/09 10:54:43 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 3.1 able 3.1 Annual Energy Expenditure by Income and Division (Tk per household) Income Category Crop Grid Non-grid (thousands of Tk) Fuelwood Tree Leaves Residue Cow Dung Kerosene Electricity Electricity Other < 25 928 419 538 562 440 167 90 18 25­50 1,688 498 582 745 570 304 146 15 50­75 2,455 469 692 711 667 553 190 23 75­100 2,877 500 811 730 706 664 254 42 > 100 3,016 454 779 875 822 1,288 293 194 Division Chittagong 3,763 300 352 479 732 761 185 124 Dhaka 1,611 713 823 654 560 710 136 10 Khulna 1,644 612 772 837 661 146 231 20 Rajshahi 685 279 648 922 474 284 140 14 All 1,962 470 641 716 608 488 172 44 Source: BIDS Survey (2004). Note: The annual energy expenditures include an imputed value for the collection time involved in gathering biomass fuels. The figures for expenditures on electricity and kerosene are reasonably close to those of the Bangladesh Household Income and Expenditure Survey (HIES). According to that 2005 national survey, the annual rural expenditure for kerosene was Tk 566 and Tk 522 for electricity. The results are close, given the different methods used to measure these variables. oftheirenergyneeds.Buttheimportanceofcommercial greater percentage of their income. The lowest-income fuels far exceeds their basic energy content. households spend slightly more than Tk 3,000 or 15 Fuelwoodaccountsfor44percentoftotalhousehold percentoftheirannualincomeonenergy.Higher-income energy use and 38 percent of cost. With regard to the households spend twice as much, but their income is commercial fuels purchased in rural areas, kerosene morethanfourtimeshigher.Thispatterniscommonin accounts for only 2 percent of total household energy other developing countries (World Bank 2002b). use but 12 percent of cost, while electricity accounts for just 1 percent in terms of energy content, but 13 percent of cost (10 percent for grid-based electricity Cooking Energy: Biomass and Its and 3 percent for solar PV, storage cells, and dry-cell Opportunity Cost batteries). The implied differences in fuel and energy Rural Bangladeshis not only spend a high proportion prices are discussed later in this chapter. of their cash income on energy; they also collect substantial amounts of biomass from fields and local By Income and Region forests. The opportunity cost of biomass collection Inmostcases,incomelevelandfuelexpenditureexhibit must be considered to understand the nature of this a positive monotonic relationship (see Table 3.1). Even energy cost. In this context, regional differences are with regard to tree leaves and cow dung, where the likelybecauseofvariationsinamountoftreecover(e.g., pattern is not completely monotonic, higher-income Chittagong and Khulna have major forest lands). households generally spend more than lower-income On average, households spend about 200 hours ones on energy. Only for the "other" category, which is per year collecting biomass fuels (see Table 3.2). numerically unimportant, is there no clear pattern. Valued at the average agricultural wage derived Poorer households spend less than wealthier from the survey for both men and women, the households on energy, but the amount represents a annual value of this work equals about Tk 1,625 per 16 5580-CH03.pdf 16 5580-CH03.pdf 16 3/11/09 10:54:43 AM 3/11/09 10:54:43 AM 3 Household Energy Demand Table 3.2 able 3.2 Biomass Collection Time by Income and Division (annual hours per household) Income Category Biomass Source (thousands of Tk) Fuelwood Cow Dung Tree Leaves Crop Residue < 25 44.4 38.8 112.2 2.1 25­50 74.5 26.0 120.4 1.3 50­75 75.5 13.9 119.9 1.0 75­100 51.2 15.9 106.5 0.8 > 100 31.9 19.7 98.1 0.7 Division Chittagong 152.9 20.7 94.9 1.0 Dhaka 28.4 14.9 116.1 0.8 Khulna 32.2 28.8 143.9 1.6 Rajshahi 20.4 34.6 106.3 1.8 All 60.6 24.6 114.3 1.3 Source: BIDS Survey (2004). family. When added to the energy expenses of the Lighting Energy poorest households, expenditures increase about 50 percent. Most rural Bangladeshis use kerosene as their Collectiontimeisevenlydividedbetweengathering primary lighting source. Some 70 percent of lighting tree leaves and collecting fuelwood and cow dung. energy is derived from kerosene, even though the Crop residue involves little time overall, since it is quality of lighting service from electricity is an order generally collected as part of farm work. By contrast, of magnitude higher across all income groups and residue collection is seasonal and thus periodically regions. In terms of pricing, kerosene varies little timeintensive.Foralltypesofbiomass,therelationship across income quintiles or regions. But for electricity, between collection time and income is generally from which about 30 percent of lighting energy is negative. One also observes substantial regional derived, prices fall steadily as household incomes rise. variations.Forexample,inforestedareasofChittagong, Other factors that influence electricity pricing include households spend much time collecting fuelwood; but connection type and regional variations. in tree-deficient Rajshahi, more time is spent gathering tree leaves and cow dung. Electricity Connection and Consumption Women devote an average of 150 hours per Electricity connection and consumption rates are year collecting biomass, disproportionately more influenced by a variety of factors, among which time than men or children spend (see Figure 3.2).11 income level and regional differences (discussed next) Children play a small role (probably collecting figure prominently. A third factor (not discussed in biomass from nearby the house). Previous studies this section) is proximity to the electricity supply appear to have overestimated the burden of fuel line. Although one might expect that consumption of collection by children, which findings from other significant supplies may be beyond the reach of poor recent studies on energy and time use confirm households, it may be that poor consumers living (World Bank 2002a). nearby the grid are being excluded from the potential 11Much of women's time is spent collecting tree leaves and grass; in terms of fuelwood collection, men's time is at least as important as that of women. 17 5580-CH03.pdf 17 5580-CH03.pdf 17 3/11/09 10:54:43 AM 3/11/09 10:54:43 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Figure 3.2 Figure 3.2 Household Biomass Collection Time by Income Quintile (thousands of Tk per year) Source: BIDS Survey (2004). Table 3.3 able 3.3 Household Electrification Connection Rate and Consumption by Income Quintile Annual Income (thousands of Tk) Connection Rate (%) Household Consumption (kWh)* < 25 15.0 32 25­50 23.2 76 50­75 33.7 167 75­100 38.2 180 > 100 54.2 455 All quintiles 29.0 144 Source: BIDS Survey (2004). Note: The survey showed that the overall monthly electricity use for households with electricity is 41 kWh, which is close to the 38-kWh figure from the records of the Rural Electrification Board. * Household use = annual consumption per household for all survey respondents. development benefits of basic electricity services. (Tk 25,000­50,000) is 50 percent more likely to have Thus, the question is whether certain policies exclude an electricity connection (see Table 3.3). Thus, for poor households from connecting to the grid. poorer households, the high cost of obtaining a grid connection may be prohibitive. When connection Income Disparities and consumption disparities by income quintile When other variables are held constant, the are compared, one sees that disparity by income is probability of having a household grid connection higher for consumption because households without has been found to increase monotonically as a connection are included in the consumption income levels rise (Asaduzzaman and Latif 2005). calculation. For example, compared to the lowest income When this study considered the independent quintile (less than Tk 25,000), the next higher group influence of income, results showed that the highest 18 5580-CH03.pdf 18 5580-CH03.pdf 18 3/11/09 10:54:43 AM 3/11/09 10:54:43 AM 3 Household Energy Demand Figure 3.3 Figure 3.3 Regional Variation in Electrification Source: BIDS Survey (2004). income quintile (more than Tk 100,000) was four times obtainedfromelectricityisanorderofmagnitudehigher morelikelythanthelowest(lessthanTk25,000)tohave than that from kerosene for all income groups across electricity, and the likelihood increased with income. all regions. For the average household, the number of kilolumens is nearly 80 times higher for those who use Regional Variations mainly electricity, versus kerosene (see Table 3.4). Electricity connection and consumption rates also The survey findings show that welfare gains from vary significantly by region. As Figure 3.3 illustrates, access to electric lighting are extremely high. Chapter 4 Dhaka and Chittagong have far higher percentages analyzes these benefits in more detail. of households with electricity connections and, consequently, kilowatt-hours of consumption. Khulna Price Differences exhibits the lowest percentage of connections and Prices paid for kerosene and electric lighting services kilowatt-hours used. differ markedly. For kerosene, prices for quantities used are fairly uniform, at Tk 20 to 21 per liter, with Kerosene versus Electricity little variation across income quintiles or regions (see Although 70 percent of lighting energy is derived from Table 3.5). When prices paid by level of lighting use kerosene and only 30 percent from electricity, in terms are considered, households in the middle quintile of cost and kilolumen output, electricity is far superior (Tk 50,000­75,000) pay the lowest price per unit and to kerosene. Electricity's output per unit of energy is poorest households (less than Tk 25,000) pay the 100 times higher than that of kerosene.12 highest price. But for electricity, prices fall steadily as household incomes rise. The price paid by the Trends highest income quintile is less than half that paid When one compares households that use kerosene by the lowest. By region, households in Dhaka and lamps or electric lighting, two trends are obvious. First, Chittagong pay the least amount. leveloflightingincreaseswithincomeforbothkerosene Because households with electricity enjoy a much and electricity. Second, the quality of lighting service higher quality of lighting than those that use kerosene 12Efficiency of appliances (kerosene kupis and hurricane lamps or electric bulbs and tubes) determines the ultimate lighting use rate, which, in turn, determines the quantity of fuel used. 19 5580-CH03.pdf 19 5580-CH03.pdf 19 3/11/09 10:54:44 AM 3/11/09 10:54:44 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 3.4 able 3.4 Comparison of Kerosene and Electric Lighting Services Income Quintile (thousands of Tk/year) Kerosene lamps (klm-hr/month) Electric lights (klm-hr/month) < 25 2.99 130.30 25­50 4.00 218.22 50­75 4.71 353.11 75­100 4.97 329.73 > 100 5.86 509.32 Division Chittagong 4.34 441.31 Dhaka 3.44 297.67 Khulna 4.46 209.57 Rajshahi 4.81 202.40 All 4.26 324.97 Source: BIDS Survey (2004). Table 3.5 able 3.5 Annual Price of Lighting by Income Quintile and Division Income Quintile Relative Price of (thousands of Tk) Kerosene (Tk/liter) Kerosene (Tk/klm-hr) Electricity (Tk/klm-hr) klm-hr (K:E) < 25 21.2 46.1 0.71 64 25­50 21.1 39.1 0.54 72 50­75 20.9 27.7 0.43 64 75­100 20.9 35.5 0.40 88 > 100 20.9 30.5 0.33 92 Division Chittagong 20.3 22.7 0.41 55 Dhaka 21.5 46.2 0.36 128 Khulna 20.9 43.0 0.61 70 Rajshahi 21.5 30.1 0.75 40 All 21.0 34.9 0.47 74 Source: BIDS Survey (2004). lamps, kerosene-using households would need to pay about 21 percent are from neighbors connected to 74timestheamountpaidbyhouseholdswithelectricity the PBSs, while the remaining 4 percent are mainly toattainthesameillumination(seeTable3.5).This ratio from the Power Development Board (see Table A1.28, rises as households move up the income ladder; that Annex 1). The average prices for kilowatt-hours and is, higher-income households pay a lower price for kilolumen-hours are therefore estimated by the type electric lighting than do poorer ones. of connection and level of income (see Table 3.6). For PBS connections, the same monotonic Price Variations of Electric Lighting and negative relationship with income is clearly In rural Bangladesh, about 75 percent of household discernible for price by kilowatt-hour and kilolumen- connections are obtained directly from the PBSs; hour. The same is generally true for connections 20 5580-CH03.pdf 20 5580-CH03.pdf 20 3/11/09 10:54:45 AM 3/11/09 10:54:45 AM 3 Household Energy Demand from neighbors. (Because the "other" category of Asset Ownership Patterns connection sources has few observations, the results are probably not meaningful.) More interestingly, Cooking,lighting,andnon-lightingelectricappliances PBS-connected households pay the highest price all have important implications for rural well being. for each income class, but the difference narrows Biomass cookstoves, for example, are closely tied to substantiallyasincomesrise.Whendivisionalpricing levels of indoor air pollution (IAP) and negative effects is considered, Khulna and Rajshahi households are on human health. The superior lighting quality of found to pay higher prices in nearly all cases. The electric lamps is closely tied to evening study and increase in takas paid per kilowatt-hour is probably educational improvements. Electric fans provide caused by a large first block for households in the indoor space cooling, while radios and televisions PBSs, combined with fixed charges on electricity offer communication and entertainment. Thus, it is bills; as the amount of electricity increases, the price important to understand typical rural household per kilowatt-hour decreases because fixed charges ownership patterns of such appliances. are averages across total kilowatt-hours used. Also, the marginal cost of a kilowatt-hour is less than the Cookstoves and Health Implications price charged to neighbors. But the profits taken by Across all regional divisions, fixed clay cookstoves are those who sell electricity to their neighbors appear used for burning all types of biomass.14 In the more negative.13 Such findings call for more in-depth highly developed division of Dhaka, nearly 7 percent analysis of formal and informal pricing of PBS of households use kerosene stoves. On average, connection arrangements. households own two clay stoves (see Table 3.7). In Table 3.6 able 3.6 Price of Electric Lighting by Income, Connection Type, and Division Income Electricity Price Quintile Electricity Price (Tk/kWh) (Tk/klm-hr) Average Price (thousands of Tk) PBS PBS-N Other PBS PBS-N Other Tk/kWh Tk/klm-hr < 25 9.2 5.8 4.1 0.87 0.48 0.34 7.7 0.71 25­50 7.2 4.7 2.9 0.63 0.40 0.23 6.3 0.54 50­75 5.4 4.1 4.6 0.47 0.36 0.28 5.0 0.43 75­100 5.1 5.4 1.7 0.40 0.46 0.14 5.1 0.40 > 100 4.3 3.4 3.2 0.34 0.28 0.23 4.2 0.33 Division Chittagong 5.4 4.4 3.8 0.44 0.36 0.26 5.1 0.41 Dhaka 4.7 4.5 2.8 0.35 0.38 0.21 4.6 0.36 Khulna 7.2 4.3 3.5 0.69 0.41 0.28 6.4 0.61 Rajshahi 8.1 5.4 -- 0.79 0.50 -- 7.7 0.75 All 5.9 4.6 3.7 0.50 0.39 0.26 5.5 0.47 Source: BIDS Survey (2004). Note: PBS-N = indirect connection to the PBS via neighbor (N). The PBS-Ns are estimates based on appliance ownership, as no bills are available and electricity is often based on fixed rates. 13Households that sell electricity to their neighbors cannot accurately determine the amount of such use. The neighbor buyers, who do not have meters, usually pay a fixed monthly amount based on a rough estimate of load, which they underestimate or under-report to the household sellers. The actual load is determined by the wattage and hours of use of all electric appliances and lightbulbs, which is difficult for the sellers to compute. 14In such locations as Sylhet District, sun-drying, rather than parboiling, is used to process paddy before milling. 21 5580-CH03.pdf 21 5580-CH03.pdf 21 3/11/09 10:54:45 AM 3/11/09 10:54:45 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 3.7 able 3.7 Household Ownership of Cooking Stoves by Type (number per 100 households) Division Stove Type Chittagong Dhaka Khulna Rajshahi Total Clay (fixed) 182.8 200.0 174.8 172.3 182.7 Clay (portable) 3.9 33.2 12.6 32.8 20.5 Kerosene 0.8 7.1 1.3 1.0 2.5 Gas 1.3 0.1 -- -- 0.4 Electric heater 1.7 0.3 -- 0.2 0.2 Total 190.5 240.7 188.7 206.3 206.3 Source: BIDS Survey (2004). per day. In short, the net effect of IAP on human health Table 3.8 able 3.8 and nutrition is difficult to predict. Average Effective Cooking Hours by Income Quintile Income Quintile Lighting Appliances (thousands of Tk) Total Cooking Hours Kerosene Lamps < 25 4.2 Bangladeshi households use two major types of 25­50 5.9 kerosene appliances: kupis and hurricane lamps. 50­75 7.0 Typically, kupis are uncovered lamps with a single 75­100 8.0 handmade wick made of discarded fabric. Hurricane > 100 9.2 lamps, which have a glass chimney and thicker (often Average 6.4 purchased) wicks, provide brighter light than kupis. Source: BIDS Survey (2004). In addition to these types, pressurized kerosene lamps, called petromaxes or hachaks, provide even brighter light. Dhaka and Rajshahi, about 33 percent of households Kupi ownership is virtually universal, while own portable stoves, which may be used for cooking hurricane lamps are owned by far fewer households outdoors. A few households also use kerosene stoves, (see Table 3.9). Poor residents tend to rely more on while gas and electric stoves are rare. Average cooking kupis, which emit little light. The proportion of time is three to four hours per day using fixed stoves households who own hurricane lamps is similar and about two hours a day using portable ones. across divisions, probably indicating their function as The average clay stove, whether fixed or portable, backup lighting for households with electricity. is used two to three hours daily. However, when total effective hours (number of stoves times number Electric Lamps of hours of operation) are considered, wealthier Virtually all households with electricity have electric households are found to cook for longer periods of lamps (using 60-W incandescent bulbs), fluorescent time than poorer ones (see Table 3.8). tubes, and compact fluorescent lamps. The number of As discussed in Chapter 4, IAP is a major health householdsusingmoreefficientfluorescentlampsisless risk for women and children. It may present an even thanthoseusingincandescentbulbs.Infact,ownership more serious hazard for wealthier households, given of fluorescent lamps is limited almost exclusively to their longer cooking hours. At the same time, poorer the two highest income quintiles (see Table 3.10). Rural households, who cook only once or twice a day, Bangladeshhasvirtuallynocompactfluorescentlamps. often end up eating cold meals more frequently than These patterns are understandable, given the large wealthier households, who cook two to three meals price differentials between an incandescent bulb and 22 5580-CH03.pdf 22 5580-CH03.pdf 22 3/11/09 10:54:45 AM 3/11/09 10:54:45 AM 3 Household Energy Demand Table 3.9 able 3.9 Household Ownership of Lighting Appliances by Electrification Status (number per 100 households) Household Appliance Type Status by Hurricane/ Charger Division Kupi/cherag Lantern Petromax Light Bulb Tube Light (with torch) Households with electricity Chittagong 212.4 95.6 -- 414.5 77.9 19.3 Dhaka 153.0 80.3 1.1 310.6 36.7 -- Khulna 161.0 81.4 -- 308.5 20.3 1.7 Rajshahi 162.3 77.0 1.6 315.6 18.9 0.8 All 176.7 85.3 0.7 348.6 47.0 7.2 Households without electricity Chittagong 210.2 98.2 -- -- -- 0.5 Dhaka 162.5 89.7 0.6 -- 2.9 -- Khulna 196.7 84.7 0.4 -- 6.7 -- Rajshahi 160.5 90.2 0.8 -- -- -- All 182.8 90.3 0.5 -- 2.5 0.1 All households Chittagong 211.1 97.2 -- 161.3 30.3 7.8 Dhaka 158.4 85.6 0.8 136.0 17.7 -- Khulna 192.9 84.3 0.4 33.2 8.2 0.2 Rajshahi 160.8 87.5 1.0 64.2 3.8 0.2 All 181.0 88.9 0.5 101.2 15.4 2.2 Source: BIDS Survey (2004). Table 3.10 able 3.10 Electric Lighting by Income and Division Income No. Compact Quintile Households Total % Using Compact Fluorescent (thousands with Electric Wattage/ % Using No. Bulbs/ Bulb Fluorescent Fluorescent Lamp of Tk) Lighting (%) Household Bulbs Household Wattage Lamps Lamps Wattage < 25 15.0 16 14.8 0.3 16 0.6 0.01 0.5 25­50 23.2 35 23.1 0.6 33 2.7 0.04 1.4 50­75 33.7 69 33.7 1.1 62 9.2 0.16 6.1 75­100 38.2 79 38.7 1.4 69 12.4 0.31 10.0 > 100 54.2 156 53.9 2.7 137 24.3 0.55 20.0 Division Chittagong 38.9 94 38.8 1.6 84 13.8 0.29 10.6 Dhaka 43.8 86 43.4 1.4 79 9.6 0.17 6.4 Khulna 10.8 17 10.9 0.3 15 2.9 0.08 2.0 Rajshahi 20.3 27 20.3 0.6 25 2.0 0.04 1.4 Average 29.0 58 28.9 1.0 52 7.3 0.15 5.2 Source: BIDS Survey (2004). 23 5580-CH03.pdf 23 5580-CH03.pdf 23 3/11/09 10:54:46 AM 3/11/09 10:54:46 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities fluorescent tube or compact fluorescent lamps using landholding determine the patterns and types of similar wattage. energy used by rural households. Additional variables Ownership of electric lamps thus depends on both that may influence household energy demand include therateofruralelectrificationandhouseholdincome.The community infrastructure and consumer prices. number of lighting watts per household also increases Together, such factors affect energy demand for both by income level. The wealthiest quintile has electric household consumption and production. For example, lights with a total of 156 watts per household, nearly 10 both income and prices influence farmers' selection of times more than the poorest quintile. In divisions with fuels and irrigation pumps. From a policy perspective, low rates of electricity use, the average number of watts it is critical to determine the relative importance per household is typically low, generally reflecting the of these factors, given the competing demand for respective region's rate of rural electrification. alternative sources of energy and its quality. The question for researchers is this: What are the direction Non-lighting Electric Appliances and magnitude of these factors' effects on demand for Given Bangladesh's hot tropical climate, it is not various types of energy? surprising that the most frequently observed non- To answer this question, we conducted an lighting appliance in households with electricity is the econometric analysis to predict how household fan (see Table 3.11). Two out of three households own at energy demand is affected by changes in the various leastoneelectricfan.Oneoutofthreehouseholdsowns factors. A tobit regression was run to account for atelevisionset.Ownershipoffans,radios,andtelevision households having zero values for one or more types sets is highly correlated with income level. With few of energy demand. The variables used as influencing exceptions, fans are found only in households with factors were gender of household head, age, maximum electricity.Somehouseholdswithoutelectricityownand level of education in the household, household income operate battery-powered radios and television sets. (proxied by land and non-land assets), community prices for major energy sources and consumer goods, Demand for Other Energy: and community infrastructure variables.15 Table 3.12 Quantitative Analysis provides summary statistics of energy demand and selected influencing variables of policy relevance, The cross-sectional analysis from the previous while Table 3.13 presents the effects of selected sections suggests that factors such as income and variables on energy demand. Table 3.11 able 3.11 Ownership of Non-lighting Electric Appliances by Income Quintile Income Quintile Electric Fans Television Sets Radios/Tape Recorders (thousands Mean Number Mean Number Mean Number of Tk) % Connected % Ownership Owned % Ownership Owned % Ownership Owned < 25 15.0 40.0 0.53 16.0 0.16 17.3 0.17 25­50 23.2 52.4 0.75 23.6 0.24 26.0 0.26 50­75 33.7 61.0 1.07 36.4 0.36 37.0 0.38 75­100 38.2 71.1 1.42 49.4 0.49 43.4 0.45 > 100 54.2 73.0 1.87 59.2 0.62 50.0 0.55 All quintiles 29.0 60.4 1.16 37.6 0.38 35.6 0.37 Source: BIDS Survey (2004). 15Only those influencing variables considered most important are reported here; see Table A2.1 (Annex 2) for the complete regression output. 24 5580-CH03.pdf 24 5580-CH03.pdf 24 3/11/09 10:54:46 AM 3/11/09 10:54:46 AM 3 Household Energy Demand Table 3.12 able 3.12 Summary Statistics of Outcomes and Important Explanatory Variables of Household Energy Demand Regressions Variable Type Mean Standard Deviation Household energy demand Fuelwood (kg/month) 98.86 104.57 Kerosene (liter/month) 2.41 1.94 Diesel (liter/month) 1.80 11.18 Electricity (kWh/month) 25.72 252.29 Explanatory Maximum education of household adults (years) 5.02 4.17 Household assets Land (acres) 1.25 2.22 Non-land (Tk 10,000) 1.68 5.55 Village price Fuelwood (Tk/kg) 1.65 0.68 Kerosene (Tk/liter) 22.94 2.73 Diesel (Tk/liter) 23.19 2.61 Electricity (Tk/kWh) 2.85 0.40 If village has electricity 0.66 0.47 Source: BIDS Survey (2004). Table 3.13 able 3.13 Estimates of Household Energy Demand Household Demand (monthly) Fuelwood Kerosene Diesel Electricity Explanatory Variable (kg/month) (liter/month) (liter/month) (kWh/month) Household Maximum education of adult males (years) 1.46** 0.020* 0.001 1.49 Maximum education of adult females (years) 0.76 0.020 ­0.001 2.38 Land assets (acres) 1.91** 0.080** 0.020** 3.71 Non-land assets (Tk 10,000) 1.55** 0.020** 0.020** 2.65** Village price Fuelwood (Tk/kg) -34.73** 0.320** -0.003 ­0.12 Kerosene (Tk/liter) 5.95** -0.009 0.002 0.53 Diesel (Tk/liter) -8.81** -0.110** 0.001 ­0.41 If village has electricity 14.21** -0.600** -0.020 32.78** Source: BIDS Survey (2004). * = significance level of 10 percent; ** = significance level of 5 percent or stronger. Figures represent changes in energy demand for fuelwood, kerosene, diesel, and electricity caused by unit changes in the explanatory variables. 25 5580-CH03.pdf 25 5580-CH03.pdf 25 3/11/09 10:54:46 AM 3/11/09 10:54:46 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities The study results support the assumption that an the price of kerosene rises. These phenomena clearly increase in income increases the demand for energy. suggest that, for rural households, fuelwood and Indeed, with higher incomes, rural Bangladeshis were kerosene are substitutes. found to pursue improved sources of energy and Thepresenceofelectricityalsoincreasesfuelwood invest in assets that require more energy. Household demand. That fuelwood is considered a good fuel income, apportioned by landholding and non-land for cooking in Bangladesh supports the notion assets, influences the demand for fuelwood and that villages with electricity are more affluent than diesel, but non-land assets matter most in demand those without electricity. These measures have an for electricity. It should be noted that income flow overall village-level income effect on the demand for can be influenced by electricity consumption, where alternative energy sources. In addition, availability electricity use is viewed as an input in production. of electricity results in reducing villagewide energy By contrast, assets are stocks that do not change over prices, which affords households more to spend on the short term but only over the long run. Therefore, energy generally, including fuelwood. But demand for we use assets rather than income to estimate the kerosene declines with village electrification, which effect of household income proxies on energy use. signals the substitution effect of users switching from For example, an increase of 1 acre in landholding kerosene to electricity (mostly for lighting). increases household consumption of fuelwood by 1.91 kg per month and kerosene consumption by 0.08 liters per month. An additional Tk 10,000 of non-land Summing Up assets increases household electricity consumption In rural Bangladesh, access to energy is governed 2.65 kWh per month. not only by its availability, but also by its pricing As incomes increase, rural households typically (monetary and non-monetary), household income, diversify their energy portfolios, moving toward and other characteristics. Household demand for modern energy, which is not only more efficient but energy is sensitive to both price and income. If also allows household members to invest time and electricity is available, household access is determined money in more productive activities than biomass by income; the higher the income (proxied by land energy alone would make possible. The study and non-land assets), the greater the demand. For any findings confirm that education reflects rural people's energy source, pricing also determines demand. Thus, preference for improved sources of energy. For energy access is not equitable because the source is example,anadditionalyearofeducationforhousehold either priced too high or carries a high opportunity adult males increases electricity consumption by cost (e.g., biomass collection). Inequitable access nearly 1.46 kWh per month. Chapter 4 discusses in and level of use may lead to further inequities (e.g., more detail the benefits of the transition to modern low-income households may pay far higher prices energy services (e.g., from biomass to kerosene or for equivalent lighting services). By exploiting price kerosene to electricity) on overall household welfare sensitivity,policymakerscanplayaroleininfluencing and productivity. demand (e.g., by withdrawing the price subsidy for The price of an energy source has a direct effect an energy source whose demand is insensitive to on its own consumption and a cross-effect on the price or, conversely, introducing a price subsidy for a demand for alternative energy sources, depending source whose demand is negatively sensitive to price). on whether they are substitutes. For example, a price The next chapter demonstrates the loss to household increase in fuelwood leads to decreased demand for welfare that can result from such inequities and the it (direct effect) and increased demand for kerosene important gains that moving up the energy ladder (positive cross-effect). A similar positive cross-effect makes possible. is observed with regard to fuelwood demand when 26 5580-CH03.pdf 26 5580-CH03.pdf 26 3/11/09 10:54:46 AM 3/11/09 10:54:46 AM 4 Household Gains from Energy Use Across the developing world, rural residents have Although there are many ways to measure exposure benefited from switching to electricity and other to smoke, the most common method involves the forms of modern energy. Productivity gains in measurement of small particulates. Small particulates agriculture and business have increased household are a heath concern because they have an impact welfare directly, while skills development and deep in the lungs and are a major cause of respiratory improved education have contributed indirectly to a illness. The most general measurements of particulate higher quality of life. matter are called PM, with a number afterwards that This chapter first considers the gains in represents the aerodynamic diameter of the particle. Bangladeshi household welfare and farm productivity For example, PM10 refers to fine particulate matter with that result from moving up the energy ladder. Given an aerodynamic diameter smaller than 10 micrograms Bangladesh's heavy reliance on biomass cooking (µg). PM2.5 has an aerodynamic diameter smaller energy, the health risk of indoor air pollution (IAP) than 2.5 µg. The exposure levels recommended by is analyzed, and potential mitigation measures are health and environmental agencies generally refer to considered. Next, the direct and indirect benefits of the number of particulates of a certain diameter per energy use on household income and consumption cubic meter of air. The most recent standards of the are presented within the context of the energy ladder World Health Organization for PM10 are 20 µg/m3 of typology. air annual mean exposure levels and 50 µg/m3 24- hour mean exposure levels (WHO 2006). For PM2.5, the recommended levels are 10 µg/m3 annual mean Biomass Cooking: From Traditional exposure levels and 25 µg/m3 24-hour mean exposure to Modern Energy levels. For Bangladesh, the standard for annual Some 95 percent of Bangladeshi households collect recommended average for PM10 is 50 µg/m3. Worldwide evidence indicates that women and or purchase biomass energy to cook all or part of their children bear the brunt of indoor air pollution their meals, mainly using fixed clay stoves. Although (IAP). Moreover, poorer households, compared to biomass is freely available or inexpensive, the inherent wealthier ones, bear the heavier burden. In rural inefficiencyofstoves,combinedwiththehighmoisture Bangladesh, age­sex composition of households content of biomass cooking fuels, results in incomplete indicates that generally women and children suffer combustion, which produces excessive smoke. greatest exposure. Adult men are the least exposed because they spend more time outside the home. Indoor Air Pollution from Traditional Stoves: Empirical Evidence Solid particulates, especially those with diameters smaller than 10 microns (PM10) are released into the Long-term exposure to smoke and small particulates air; and, if inhaled for prolonged periods, can lead to haslongbeenconsideredanenvironmentalhealthrisk. various diseases, loss of health, and early morbidity. Note: In addition to the main authors, Hussain Samad and Susmita Dasgupta were contributing coauthors of this chapter. 27 5580-CH04.pdf 27 5580-CH04.pdf 27 3/11/09 10:55:03 AM 3/11/09 10:55:03 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 4.1 able 4.1 Mean Daily Hours in Household Location by Age Group* Female Male Female­Male Age Cooking Living Outdoors Cooking Living Outdoors Cooking Living Outdoors Group Area Area Area Area Area Area 0­1 1.12 20.04 3.01 1.11 19.29 3.57 -0.01 -0.75 0.56 2­5 1.08 18.44 4.52 0.93 18.13 4.97 -0.15 -0.31 0.45 6­8 1.01 16.40 6.61 0.48 16.41 7.17 -0.52 0.01 0.56 9­12 1.32 15.55 7.19 0.31 15.61 8.06 -1.01 0.06 0.87 13­19 2.38 15.71 5.97 0.28 14.33 9.41 -2.10 -1.38 3.44 20­60 3.75 16.05 4.27 0.19 13.07 10.79 -3.56 -2.98 6.52 > 60 1.48 19.76 2.93 0.17 16.56 7.37 -1.31 -3.20 4.45 Source: Dasgupta et al. (2004). * For seven regions in Bangladesh. Finer particles (PM2.5) are even more damaging. This As Dasgupta et al. (2004) have estimated, both section discusses various aspects of IAP caused by males and females are exposed to solid particulate incomplete biomass combustion. pollution far above accepted standards for all Both male and female children spend many hours ages, except for males 20 to 60 years of age, whose indoors in the cooking space or adjacent living area exposure level is little above the standard pollution until 6 to 8 years of age, when gender-based patterns concentration (see Figure 4.2); but they, too, are at begin to diverge (see Table 4.1). Adolescent males risk for respiratory and other diseases. Exposure begin to spend more time outdoors, while adolescent levels and their consequences may be tempered by females move in the opposite direction, spending various socioeconomic factors, including education more time in indoor cooking and living areas. Women and income, which influence awareness, fuel 20 to 60 years of age spend nearly 4 hours per day choice, mitigation measures, and general living cooking, nearly 16 hours in the household living area, environment. and the remaining hours outdoors. For those above 60 Among Bangladeshi children, acute respiratory years of age, the patterns are somewhat reversed. infection (ARI) remains a major cause of morbidity Evidence suggests that pollution levels in living and mortality. A 2004 survey reported that 21 percent areas adjacent to the kitchen are broadly similar. of rural children 1 to 5 years of age suffered from Except for a few hours in the early morning and early ARI.16 This study's household survey indicates a afternoon, PM10 concentrations in kitchen and living broadly similar picture. However, the survey data areas remain far above 100 µg/m3,the Indian standard are imprecise. Because they only reported whether for air pollution load. Ambient concentration, for the working days were lost in the prior month due to most part, remains below the Indian standard, except specific diseases, the reported incidence of children's during early evening hours (see Figure 4.1). diseases may have been lower. In addition, the specific 16The 2004 survey was conducted by the National Institute of Population Research and Training (NIPORT), Mitra and Associates and Measure/DSH+. 28 5580-CH04.pdf 28 5580-CH04.pdf 28 3/11/09 10:55:03 AM 3/11/09 10:55:03 AM 4 Household Gains from Energy Use Figure 4.1 Figure 4.1 PM10 Concentration by Time of Day and Household Location Source: Dasgupta et al. (2004). Figure 4.2 Figure 4.2 PM10 Concentration by Age Group and Gender Source: Dasgupta et al. (2004). 29 5580-CH04.pdf 29 5580-CH04.pdf 29 3/11/09 10:55:03 AM 3/11/09 10:55:03 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities symptoms or diseases reported by household heads from indoor PM10 concentration. Figure 4.3 presents were subject to bias. Because symptoms were not the interactive effect of critical pollution factors by always well clarified, it is difficult to know whether computing mean PM10 concentrations for groups that respiratory problems were present.17 distinguish between clean (kerosene and natural Despite these information gaps, the findings are gas) and biomass fuels, inside and outside (detached instructive.Overall,10to11percentofmenandwomen or open-air) cooking, and mud-wall and other experienced respiratory problems and diseases. About construction materials. For statistical comparison 24 percent of male children and 20 percent of female of means, the benchmark is the mean living-space children in the 0 to 5 age group suffered from such PM10 concentration for households with biomass diseases. For children and adolescents 10 to 19 years fuels, inside cooking, and non­mudwall construction of age, about 8 percent of both genders were affected. (acronym BIOL). The mean concentration for these For adults 20 to 60 years of age, the proportions were households is 223 µg/m3. 9 percent for men and 10 percent for women. Thus, Pitt, Rosenzweig, and Hassan (2005), working there was little difference by sex, but young children with Bangladesh data, show that such exposures may suffered more than adolescents and adults. not be random. Their findings show that cooking is Evidence from rural Bangladesh indicates done mostly by women. But those women in poorer that IAP varies by fuel type, cooking location, health are more involved in cooking, possibly because kitchen ventilation characteristics, and other factors; they have little chance to switch to cleaner fuels furthermore, non-fuel factors may be as important as or because the task falls onto those who otherwise fuel characteristics (Dasgupta et al. 2004). The extent cannot contribute to more productive activities. and duration of particulates in the kitchen, as well as Among mothers, those with young children are the amount of smoke that leaks into other living spaces less involved in cooking than others, indicating that or the outdoors, may depend on kitchen location and cooks exposed to IAP are conscious of its hazards ventilation, as well as the porous quality of materials and avoid them as much as possible. Yet, because used to construct kitchen roofing and walls. This fuel-substitution possibilities are extremely limited survey revealed wide variation in household- or costly, the adjustments are insufficient. specific ventilation characteristics. Regression of This study has detected dangerously high 24-hour PM10 concentration on fuel use and a large levels of pollution in rural Bangladeshi households. set of variables that describe household cooking and Concentrations of respirable airborne particulates ventilation practices suggest that ventilation factors (PM10) 300 µg/m3 or greater are common, implying play a larger role than fuel choice. Two construction widespread exposure to a serious health hazard. materials, mud walls and thatch roofs, significantly Within households, individuals' exposure is related to affect PM10 concentration.18 If the cooking location is concentrations of pollution in indoor locations during indoors, the sealing effect of mud walls increases PM10 the daily round of activities. The estimates reveal concentration significantly. high levels of exposure for children and adolescents After controlling for stove locations, construction of both sexes, with children under five years of age materials, and opening of doors and windows after at increased risk. Among prime-aged adults, men cooking, the PM10 effects of biomass fuels (dung, have half the exposure of women (whose exposure is fuelwood, twigs and tree branches, dung, rice similar to that of children and adolescents). Elderly husks, and straw) are indistinguishable. Relative men also have significantly lower exposure than to these fuels, which are taken as the baseline, use elderly women. of kerosene subtracts about 90 µg/m3, LPG or LNG Predictably, fuel choice affects indoor pollution subtracts 103 µg/m3, and piped natural gas 136 µg/m3 levels. Kerosene and natural gas are significantly 17Colds, coughing, breathing problems, tuberculosis, headaches, eye problems, chest pains, and fever were compiled to give a broad idea of the respiratory problems associated with IAP. 18In most localities of rural Bangladesh, the soil has low sand content, and mud walls and floors are frequently recoated to prevent cracking. The effective mud seal permits little ventilation, compared to other common construction materials (e.g., thatch or corrugated iron). 30 5580-CH04.pdf 30 5580-CH04.pdf 30 3/11/09 10:55:06 AM 3/11/09 10:55:06 AM 4 Household Gains from Energy Use Figure 4.3 Figure 4.3 Pollution Factors and PM10 Concentrations Source: Dasgupta et al. (2004). cleaner than biomass fuels. But econometric the pollution levels of higher-income households with results strongly suggest that certain household highlyeducatedadults.Theexposureofyoungchildren characteristics--construction materials, space and poorly educated women in poor households is configurations, cooking locations, and use of doors quadruple that of men in higher-income households and windows--matter as much or perhaps more than organized by more highly educated women. fuel choice in determining PM10 concentrations. In some biomass-burning households, concentrations Possible Gains from Cooking are scarcely higher than in those using natural gas. with Modern Stoves Across households, family income and adult Use of improved stoves and more modern forms educationlevels(particularlythatofwomen)significantly of cooking energy has several classes of benefits, affect choices of cooking fuels, cooking locations, some of which are private and others public. This construction materials, and ventilation practices. As a section considers mainly the private benefits, which result,thepoorest,least-educatedhouseholdshavetwice 31 5580-CH04.pdf 31 5580-CH04.pdf 31 3/11/09 10:55:06 AM 3/11/09 10:55:06 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities are fairly self-evident. Because improved stoves cooking, avoided health costs, and environmental require less fuel, using them reduces either the time benefits,amongothers.ForatypicalhouseholdinSouth spent collecting biomass fuels or the money spent to Asia, the benefits of switching exclusively to improved purchase them. Cooking with modern fuels involves stoves or from biomass energy to LPG or kerosene total cash expenses, but also entails many benefits. about US$30 per year. Some of the best new-generation In most rural economies, biomass fuels are models, lasting 5 to 10 years, cost about US$50. Thus, collected from the local environment; thus, the households that use such stoves can expect a positive benefits mainly involve the reduction in collection return on investment within two years. time for household members, especially women. As The benefits of improved stoves, with certain reported in Chapter 3, annual biomass-collection time caveats, significantly outweigh the costs, at least in for rural households in Bangladesh totals about 200 theory. Unfortunately, the rural reality tells a much hours (150 of which are contributed by women). This different story. If rural households perceived that even means that using an improved stove saves about 25 a fraction of the benefits developed under the national percent of both fuel and collection time, representing program were true, the program would not require a significant annual savings. For those that purchase significant subsides. Rural residents would be willing biomass fuels, the benefits are similar to, but more topurchasesuchstovesattheirretailcost.Butimproved straightforward than, time savings. If 25 percent less stoves must be designed to deliver a range of cooking fuel is used for the same amount of cooking, then services that people want: fuel efficiency, reduction in avoided expenses equals a 25 percent monetary gain. IAP, ability to cook a wide array of food preparations, Similar potential gains can result from using modern cooking convenience, and even attractiveness in the fuels, such as LPG, for cooking. kitchen or room. Given the history of low adoption It is not as straightforward to estimate the rates, it is clear that past government subsidies used to private health benefits of using improved stoves and promote the program have been problematic. modern cooking fuels (e.g., avoided eye irritation and Despite the effectiveness of improved stoves illnesses or early deaths caused by exposure to high to reduce indoor cooking smoke, programs from levels of IAP and improved cleanliness of the home). around the world reveal a wide range of successes and Most households are well aware of the benefits of failures. Obviously, if a household decides to revert eliminating smoke from their kitchens and other to a traditional stove or uses an improved stove that rooms. Household members, especially women, know develops cracks and leaks, the desired improvements the drudgery involved in keeping pans and walls free in cooking efficiency and reduced IAP will not of the extensive carbon produced by traditional stoves. be achieved. In addition, even after a household Assuming the same level of service and ease of use, adopts an improved stove, indoor pollution levels households should, at least theoretically, be willing remain significantly higher than those advocated to pay more for a stove that involves less eye and by international development organizations, such as throat irritation and less cleaning of pots and pans. the World Health Organization and environmental However, the health and avoided death benefits are agencies. Clearly, Bangladesh requires more options more difficult to grasp.19 with regard to the types of improved stoves promoted Despite these difficulties, the World Health and potential use of modern cooking fuels. Organization is making efforts to quantify some of the public and private benefits of using improved stoves or substitutingpetroleumforbiomasscookingfuels(Hutton From Kerosene to Electric Lighting and Rehfuess 2006; Hutton et al. 2006; WHO 2006). The high-quality lighting afforded by electricity These studies develop economic valuation methods has important consequences for Bangladeshi to evaluate time savings involving fuel collection and household welfare. This study applied the concept 19Health evidence is just reaching the point where people are realizing the long-term health consequences of IAP. 32 5580-CH04.pdf 32 5580-CH04.pdf 32 3/11/09 10:55:07 AM 3/11/09 10:55:07 AM 4 Household Gains from Energy Use Table 4.2 able 4.2 Annual Energy Use by Number of Students in Household Students in Lighting Domestic Uses Household (no.) Kerosene (liter) Grid Electricity (kWh) Kerosene (liter) Grid Electricity (kWh) 0 23 52 25 92 1­2 28 84 30 153 3­4 34 121 36 209 > 4 37 232 40 415 Average 27 81 29 144 Source: BIDS Survey (2004). of consumer's surplus to measure the welfare gains Table 4.3 able 4.3 for lighting resulting from transitioning from less Study Time by Household Electrification Status efficient kerosene-using devices to those using (average number of hours) electricity to obtain the same amount of benefit (Annex 3). Similar measurements could be made Student Study Household Status for other services (e.g., entertainment), which were Time With Electricity Without Electricity beyond the scope of this study. Daily 2.72 2.13 Evening 1.32 0.96 Gains in Education Source: BIDS Survey (2004). Mounting global evidence suggests that rural electrification is closely tied to higher income and improved education. Various studies have confirmed the link between rural electrification and education, lighting, use of kerosene lamps varies considerably suggesting that children's school attendance improves less by number of students (see Table 4.2). as households adopt electricity (Kulkarni, Barnes, and The most plausible explanation for such increases Parodi 2007). Many rural surveys have found that in electricity use is that schoolchildren use electric children study better when the high-quality lighting lights to extend their study time. According to the made possible by electricity is available (World survey, students in households with electricity study Bank 2002b). A recent study on rural electrification longer (see Table 4.3). and development in Bangladesh indicates that Total hours of study time (both daily and evening- households with electricity have higher literacy rates hour) are greater in households that use electric and increased school enrollment than those without lighting versus those without electricity, which use electricity (Barkat et al. 2002). kerosene lamps. Clearly, electricity contributes to Evidence from this study suggests that human-capital formation, leading to a longer-term the number of school-going students in rural increase in productivity. Bangladeshi households influences the amount of electricity consumed (Asaduzzaman and Moving to Better Lighting: Consumer's Latif 2005). If the number of students in a family Surplus household increases by one, annual electricity In rural Bangladesh, the most common household consumption grows by about 51 kWh. This change lighting appliances are kerosene-using kupi and is large, given that average annual household hurricane lamps and electricity-using incandescent consumption is 114 kWh for domestic uses and bulbs, fluorescent tubes, and compact fluorescent 81 kWh for lighting. Indeed, compared to electric lamps. In this study, households were categorized into 33 5580-CH04.pdf 33 5580-CH04.pdf 33 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities three groups: (1) those that light only with kupi lamps, lamp holdings are far more common than kupis, (2)thosewithhurricanelampsthatmayormaynotown and patterns across income classes are not clearly akupilampbutdonothaveelectricity,and(3)thosewith discernible. Households with electric lamps are electricity that sometimes use an electric lamp and may about 29 percent of households, reflecting those that or may not use any type of kerosene lamp. The gains have electricity; and their proportion increases with in consumer's surplus were estimated for the transition income. from the kupi- to the hurricane-based system, and from One conservative measure of the improvement the hurricane- to the electricity-based system. in consumer welfare is avoided expenditures from To understand the results of the estimates switching from kerosene to electricity for lighting. of welfare gains, it is necessary to observe the Although this is a rather rudimentary measure, it distribution of households by their lamp-holding is interesting to compare kerosene expenditures for characteristics. This study's findings indicate that households with and without electricity. Because kupi-based households are concentrated more households without electricity spend more on among poorer households (see Table 4.4). Hurricane kerosene, some fuel substitution occurs (see Table 4.5). Generally, households with electricity spend 200 to 500 takas less per year on kerosene. Although Table 4.4 able 4.4 this avoided expenditure is an interesting benefit, it Households by Lighting System and Income severely underestimates the true benefits because households with electricity have significantly more Income Type of Home Lighting System Quintile (% households) light compared to those with kerosene lamps. (thousands Hurricane- Electricity- A better way to measure the welfare gains of of Tk/year) Kupi Only based based switching from kerosene to electricity involves a < 25 41.4 43.6 15.1 concept called consumer's surplus, which is based 25­50 19.5 57.3 23.2 on a demand survey for lighting (see Annex 3). 50­75 13.3 53.0 33.7 The gain in welfare, as measured by consumer's 75­100 10.6 51.2 38.2 surplus, is substantial for the average consumer > 100 5.0 40.6 54.4 switching from any of the kerosene lamps to electric Average 20.1 50.8 29.0 lights (see Table 4.6). The estimated benefit is 40 to Source: BIDS Survey (2004). 45 percent of household income (a substantial gain in terms of Bangladeshi takas). The reason is the inherent efficiency of electric lamps. The price of each kilolumen-hour from incandescent bulbs or Table 4.5 able 4.5 Household Expenditure by Electrification Status on Kerosene and Other Energy Sources, by Income Group (Tk per year) Households without Electricity Households with Electricity Income Category Other Energy Other Energy (thousands of Tk) Kerosene Expenditure Expenditure Kerosene Expenditure Expenditure < 25 467.8 2,466.0 281.9 3,054.7 25­50 645.9 3,613.8 319.2 3,865.4 50­75 794.4 4,715.2 415.6 4,179.2 75­100 883.6 5,886.1 418.8 4,095.9 > 100 1,133.8 5,944.9 559.9 5,270.3 All 688.9 3,905.0 408.8 4,227.2 Source: BIDS Survey (2004). 34 5580-CH04.pdf 34 5580-CH04.pdf 34 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM 4 Household Gains from Energy Use Table 4.6 able 4.6 Welfare Gains from Switching Lighting Systems System Transition Income Group Kupi Only to Hurricane- to Kupi Only to (thousands of Tk/year) Hurricane-based Electricity-based Electricity-based < 25 56 (4.0) 838 (59.4) 894 (63.5) 25­50 67 (2.2) 1,333 (43.4) 1,401 (45.6) 50­75 83 (1.7) 2,243 (44.0) 2,326 (45.6) 75­100 58 (0.8) 2,156 (29.8) 2,214 (30.6) > 100 80 (0.5) 3,197 (20.9) 3,276 (21.4) Average 72 (1.4) 2,057 (40.1) 2,129 (41.5) Source: BIDS Survey (2004). Note: Left-hand figures represent takas per month; right-hand figures in parentheses are ratios of the gains in consumer's surplus to the relevant group's average monthly income. fluorescent tubes is only about 1.43 percent of the cost for supplementary irrigation, which lessens the of comparable lighting service from kerosene lamps risks associated with rainfed systems and enables (see Table 3.5). increasing crop and pasture productivity, as well as Interestingly, when consumers are divided into switching to higher-value crops" (United Nations income classes, the relative income gains are higher for 2005). One should interpret the gains discussed here poorer groups. But as expected, because of the higher with caution, however, as more intensive data is density of lamps found in wealthier households, their needed to conduct an appropriate analysis.20,21 absolute gains are two to three times higher than those of poorer groups. These findings have major Improved Productivity rural energy policy implications for Bangladesh. Bangladesh's agriculture is characterized by three rice- cropping seasons (Chapter 5). From the viewpoint of From Manual to Mechanized Farming energy inputs, the most important is the dry season, when irrigated winter rice (boro) is grown; boro rice This study hypothesized that using modern energy accounts for roughly 50 percent of rice production. The for farming ultimately leads to gains in agricultural other is the partly wet/partly dry season, when aman productivity. According to the United Nations rice, which may require supplementary irrigation, is Millennium Project: "At the household scale, grown. In all cropping seasons, both bullocks and modern energy services directly contribute to power tillers are used for land preparation. As a economic growth and poverty reduction. They create result, four basic types of energy-using technology opportunities for income generation, reduce unit are classified by energy-use importance in production: costs, and enable increased income from agriculture low tillage­low irrigation, low tillage­high irrigation, or animal husbandry by permitting pumping high tillage­low irrigation, and high tillage­high 20Farmers accounted for 61 percent of the survey sample; they were involved in some form of cultivation during the 12 months preceding the survey. 21For example, while data on crops grown and output, including by-products, have been requested from respondents on a disaggregated level, cost figures have been taken on an aggregated level. In such cases, total costs of fertilizer purchase may be accurate overall; but aggregate information on labor costs may have a significant margin of error. Unfortunately, collection of labor-related data is a detailed exercise; thus, gross, rather than net, returns have sometimes been used. Furthermore, labor use, which usually differs by tenure arrangement, adds to the complexity in determining net returns. 35 5580-CH04.pdf 35 5580-CH04.pdf 35 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 4.7 able 4.7 Farm Household Distribution by Tillage/Irrigation Technologies Energy Technology Rice Cropping Season (%) Combinations Aus Aman Boro Overall Low tillage­Low irrigation 98.1 93.1 41.9 39.4 Low tillage­High irrigation 1.3 4.9 56.3 58.2 High tillage­Low irrigation 0.5 1.4 0.4 2.4* High tillage­High irrigation 0.1 0.5 1.4 * Source: BIDS Survey (2004). * Indicates merger of these two groups. irrigation (see Table 4.7). Low irrigation is defined as Table 4.8 able 4.8 the absence of irrigation or manual irrigation, while Gross Productivity of Paddy by Tillage/Irrigation high irrigation refers to use of mechanized (diesel or (Tk/decimal) electric) pumps. Low tillage is defined as the use of bullocks in land preparation, while high tillage refers Rice Cropping to the use of power tillers or tractors. Season Modern energy use is particularly relevant during Energy Technology Ladder Boro Aman the boro season and, to a considerable extent, in the Low tillage­Low irrigation 118a,b 84a,b aman season. Rural Bangladesh has only two major Low tillage­High irrigation 145a 96a combinations of energy technology: low tillage­low High tillage­Irrigation (low and high) 140b 68b,a irrigation and low tillage­high irrigation. Empirically, Source: BIDS Survey (2004). it is the difference between the first two categories that Note: 1 decimal = 50 square yards; comparisons are for the same crop (within same column) between technology levels. indicates the productivity-raising effect of modern aFigures differ at 1 percent significance level. energy,particularlyduringtheboroseasonandinpaddy bFigures differ at 5 percent significance level. cultivation, which accounts for most farmed land. The study found productivity differences by energy type for boro and aman paddy for owned land education of household head, size of owned land, and cultivated by the sample respondents (see Table 4.8). regional variations are accounted for, up to 15 percent Farm productivity is significantly higher during the improvement in productivity can be attributed to boro and aman seasons (22 percent and 17 percent, switching from low tillage­low irrigation to the higher respectively). Thus, as is commonly known, irrigation irrigation scenario. In absolute terms, the figure totals provides a significant improvement in productivity. nearly Tk 1,800 per acre. For aman paddy, by contrast, estimated improvements are substantially smaller Distribution of Irrigation Gains or may not be clearly discernible because irrigation by Land Ownership may be supplemental, depending on rainfall level in Although likely to have higher energy use, larger a particular location or farmer's land conditions. farms generally have lower land productivity than In reality, the gains may be even more substantial smaller ones. To purge the confounding influence and inequitably distributed. Table 4.9 illustrates of such factors on land yield, this study conducted the issue for boro paddy production. Both sets of a regression analysis for the average gross yield of observed and estimated yield figures for low-energy boro paddy land. Results of the analysis suggest that, technology users exhibit the classic pattern of negative on average, when the influence of such factors as size­productivity relationship. But the observed yields 36 5580-CH04.pdf 36 5580-CH04.pdf 36 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM 4 Household Gains from Energy Use Table 4.9 able 4.9 Paddy Productivity in Boro Season by Land Ownership (gross value in Tk/decimal) Technology Land Ownership Observed Estimated (decimal)* Low-energy High-energy Low-energy High-energy 1­49 164 147 (­10) 140 149 (6) 50­249 141 145 (3) 123 144 (17) 250­500 96 146 (52) 99 138 (39) > 500 80 141 (76) 89 130 (46) Average 118 145 (23) 107 141 (32) Source: BIDS Survey (2004). Note: Figures in parentheses represent percentage change over the low-energy situation. * 1 decimal = 50 square yards. for high-energy technology users are similar for every section examines more general income gains that can land-ownership class. The change in land productivity be derived from moving up the energy ladder, many of from energy use is likely masked by the influence of which overlap with those already described. In short, other factors. Although both observed and estimated the gains described in this section are a more general yields fall with rising land ownership, the one high indicator of the value of the changes that result from in energy use declines at a slower rate than the one improved cooking and living standards made possible low in use, indicating increasingly greater gains from by better ways of using energy. higher farm technology (energy) use. The percentage differences in estimated yields for high energy use are Cross-sectional Correlations of Electricity somewhat less than those for the observed figures, but Use and Income the patterns are similar. Furthermore, the average gain This study survey measured household income is 32 percent (Tk 3,400), nearly double that suggested separately for business and wage-related activities. initially by the regression equation. More specifically, there are four measures of The conclusion is that farm households that income, including agriculture (farm-production move up the energy ladder enjoy a substantial gain in activities), non-agriculture (entrepreneurial activities), productivity and can provide irrigation or purchase agricultural wage labor, and non-agricultural wage irrigation water for their farms. About 50 percent labor. This section examines income levels for of the average gain in transitioning from lower- to households with and without electricity, while the higher-energy technology is derived from switching next one explores income gains from using more to more intensive irrigation using mechanized (diesel modern energy, controlling for many other factors. or electric) irrigation pumps, with complementary Overall, households with electricity have factors accounting for the other 50 percent. significantly higher incomes than those without electricity (see Table 4.10). This is generally true with Overall Income Gains: Moving Up the regard to income from non-agricultural activities, except in divisions where irrigation is more prevalent, Energy Ladder notably Rajshahi. In other divisions, the relationship Thus far, this chapter has examined some of the gains may be reversed. Regarding agricultural wage labor, derived from using more modern energy for the the relationship is counterintuitive, as it relates specific areas of lighting, cooking, and irrigation. This negatively to household electrification. Finally, 37 5580-CH04.pdf 37 5580-CH04.pdf 37 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 4.10 able 4.10 Household Income Differentials by Electrification Status (Tk per year) Electricity Division Income Source Connection Chittagong Dhaka Khulna Rajshahi All Divisions No 31,776a 17,348 20,395b 19,751 a 22,227 Agriculture Yes 21,251a 20,169 28,898b 35,418a 23,980 No 23,443a 24,729a 18,706a 13,538a 19,545a Non-agriculture Yes 78,455a 58,236a 43,747a 29,504a 59,208a Agricultural No 4,821a 3,086a 4,606a 6,439a 4,868a wage labor Yes 2,240a 1,775a 1,168a 2,883a 2,085a Non-agricultural No 3,865 5,972 2,969 2,430 3,624 wage labor Yes 2,915 6,472 4,569 2,490 4,334 No 63,906a 51,135a 46,677a 42,159a 50,264a Total Yes 104,860a 86,652a 78,383a 70,296a 89,607a Source: BIDS Survey (2004). aDifferences are statistically significant at 1 percent level or less. bDifferences are statistically significant at 5 percent level. non-agricultural wage labor is shown to have little control for many factors. To quantify the effects of relevance to electrification. Thus, it is mainly income energy use on household income or consumption from non-agricultural activities that has the most while controlling for other important factors, a stable relationship with electrification. Given the low regression technique is most appropriate. In this use of electricity for irrigation in Bangladesh, this is section, the results are presented only for the energy not an unexpected finding. and outcome variables. The detailed models used for Electricity is not the only modern energy source this analysis can be found in Annex 2. used in agriculture. The main fuel used for irrigation For this analysis, we divided household energy in Bangladesh is diesel. As discussed previously, rural use into two relevant categories: (1) non-lighting and households can gain substantially by moving up the (2) lighting. Both types can be used for consumption energy ladder--whether switching from kerosene and income generation. We further categorized energy kupis to electric lamps, manual irrigation to electric use by major sources, namely, biomass, kerosene/ pumpsets, or inefficient biomass stoves to improved diesel, and electricity. Thus, the main outcome models using cleaner-burning fuels. Similarly, farm variables for the study are both farm and non-farm productivity gains from switching to mechanized income (see Table 4.11). Perhaps surprisingly, in irrigation are estimated at 17 to 22 percent for any rural Bangladesh, non-farm income is higher than cropping season. farm income. In addition to income, household consumption is considered an outcome.22 Net Gains after Controlling for The study estimates the benefits of various forms Other Variables of energy, especially modern energy, on household It should be noted that correlation does not signify consumption (net of energy consumption) and income causality, and the tables presented thus far do not obtained from both farm and non-farm production. 22Household energy expenditure is excluded from this consumption figure. 38 5580-CH04.pdf 38 5580-CH04.pdf 38 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM 4 Household Gains from Energy Use Table 4.11 able 4.11 Summary Statistics of Outcomes and Energy Use Variables of Household Welfare Regressions Variable Mean Standard Deviation Outcome (Tk/month) Consumption 3,581 2,449 Farm income 2,229 2,702 Non-farm income 2,889 4,997 Total income 5,118 5,950 Monthly energy use Non-lighting (kgoe) 95.7 70.7 Lighting (klm-hr) 115.6 267.7 Biomass (kgoe) 90.1 65.7 Kerosene/diesel (kgoe) 3.5 9.5 Electricity (kWh) 25.7 252.3 Source: BIDS Survey (2004). Table 4.12 shows the effects of energy use on household lighting energy, such as diesel, is more important consumption and income by types of use (lighting and for farm income. After all, diesel can be used for non-lighting) and source (biomass, kerosene/diesel, irrigation, which, as noted previously, is strongly and electricity).23 The findings are interesting and in correlated with farm productivity. line with what one would expect from both farm and Theeffectsofdifferenttypesofenergyconsumption non-farm households moving up the energy ladder. (biomass,kerosene/diesel,andelectricity)areevenmore To a certain extent, the findings reflect the interesting. Biomass energy is the fuel most commonly efficiency with which a fuel is used. Biomass energy used by households in Bangladesh. An increase of is generally burned in inefficient stoves, ovens, or 1 kgoe of biomass energy increases monthly household boilers. Kerosene and diesel use generally has energy income by about Tk 8, most of which can be attributed efficiency ratios of about one-third. At the end-use to farm income. Moving up the energy ladder yields point, electricity is a very efficient form of energy. greater income gains. Use of kerosene or diesel fuel One would expect its effect to be higher than that of significantly raises farm income, likely resulting from other fuels on both expenditure (consumption) and diesel-powered irrigation. Use of 1 kgoe (about 1 liter income. The results confirm that electricity has the offuel)increasesfarmincomebyapproximatelyTk149, highest level of impact on both energy expenditures significantly higher than its cost. Of all fuels, electricity and the equivalent measure of income. has the greatest effect; 1 kWh contributes to raising As might be expected, lighting energy has a non-farm income by Tk 4.7 and total income by almost greater effect on non-farm income than farm income. Tk 16, which is equivalent to more than Tk 185 for 1 Conversely, non-lighting energy has a greater effect kgoe, a gain of more than 20 times that achieved from on farm income than non-farm income. Thus, lighting biomass use. Much of the non-farm income gain from is more important for non-farm income, while non- electricity use can be attributed to lighting, the most 23Tables A2.2 and A2.3 (Annex 2) present complete regression results. These are marginal gains, which differ from the gross average gains calculated earlier. 39 5580-CH04.pdf 39 5580-CH04.pdf 39 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities common rural use of electricity. But one should not households switch from kerosene to electric lighting. discount electricity's importance to farm income in Land productivity and agricultural income increase rural Bangladesh, despite its limited use in irrigation. when farm households switch from manual to Income and other welfare gains are indicators of mechanized tillage and irrigation. improved energy use by rural households.24 Given The indirect link between electrification and that biomass is used only for cooking, kerosene education is also positive. Households with electricity for both cooking and lighting, and electricity for have higher levels of education, although the causal lighting only, one can infer that household welfare direction is unclear. Children who live in households would increase significantly by switching from with electricity appear to study longer, thereby biomass to kerosene for cooking and from kerosene laying the foundation for longer-term, human-capital to electricity for lighting. As previously discussed, formation.Sincemainlythepoorlackelectricity,amajor real improvement in household income and concern is that children in poor households miss out on consumption can be achieved by switching from the educational benefits that electricity affords. biomass to kerosene or from kerosene to electricity. Another concern is IAP, caused by the incomplete The evidence is fairly conclusive that switching to combustion of biomass for cooking and heating. modern energy sources can increase a household's Although the effects are not yet clearly understood, income and consumption both directly and indirectly. switching to cleaner-burning fuels, if affordable, and Increased income leads to greater consumption and using improved stove designs, can mitigate the health more diversified investment, which, in turn, creates risk to rural households. more income and household welfare. Modern energy's returns to household welfare and farm productivity are substantial, even after controlling for such factors as education, assets, and Summing Up household location. Switching to modern energy and This chapter has demonstrated that moving up the the modern use of biomass also contributes to the energy ladder can contribute significantly to rural productivity of rural industries and microenterprises, income growth. Large welfare gains result when the subject of the next chapter. Table 4.12 able 4.12 Estimates of Household Welfare (Tk per month) Monthly Energy Use Consumption Farm Income Non-farm Income Total Income Model 1 Non-lighting (kgoe) 4.89** 9.61** 3.80 11.20** Lighting (klm-hr) 4.32** ­0.19 9.68** 9.39** Model 2 Biomass (kgoe) 3.43** 13.32** -10.37 8.18** Kerosene/diesel (kgoe) -18.00 149.07* -151.61 ­54.59 Electricity (kWh) 7.43** ­2.17 4.72** 15.73** Source: BIDS Survey (2004). Note: Figures represent changes in outcome (consumption and income) due to one unit change in energy use. * = significance level of 10 percent. ** = significance level of 5 percent or stronger. 24The consumer or producer's surplus approach used to calculate consumption gain does not represent a real increase in household or farm income; rather, it estimates potential gain; that is, the savings a household could realize by switching from a lower-tier energy source to a higher one (Annex 3). 40 5580-CH04.pdf 40 5580-CH04.pdf 40 3/11/09 10:55:08 AM 3/11/09 10:55:08 AM 5 Energy Consumption and Rural Production Aroundtheglobe,microenterpriseshavedemonstrated informationonenterprisetype,assets,andemployment-- their ability to facilitate economic growth and reflects the need to conduct separate surveys on rural development. China's development of small- and household and growth-center microenterprises. A medium-sizedenterprises,forexample,hascontributed brief section on energy and farming highlights the tothecountry'srapideconomicadvance.Suchprograms importance of irrigation for productivity. have been complemented by rural electrification and otherruralinfrastructureinvestments.Indeed,without Enterprise Types and Distribution theunderpinningsofelectricityandotherdevelopment Bangladesh's rural enterprises can be grouped into inputs, rural microenterprises would struggle. But three location-based categories: growth center, village, simply introducing modern energy into the rural and home. The survey findings reveal, perhaps economy is unlikely to produce an economic miracle. surprisingly, that most home-based businesses Substantial evidence suggests that other supporting center on manufacturing, while village and growth conditions are needed for businesses to grow and center industries are involved more in trading. On flourish. Under the right conditions, long-term growth closer examination, the reasons are clear. Many can be sustained. home businesses are run by women involved in such This chapter charts the course of modern energy production activities as basket weaving and sewing. use in Bangladesh's rural microenterprises, exploring Outside the home, village enterprises, run mostly by theirenergy-usecharacteristicsandtheeffectsofenergy men,areinvolvedinstoreoperationssellinggoodsand consumption on business growth and development. services (see Table 5.1). The rich data set collected by this study--including Table 5.1 able 5.1 Microenterprise Distribution by Location and Activity Type Microenterprise Type Agriculture and Food Manufacturing Trading Services and Other All Location No. % No. % No. % No. % No. % Home 5 3.6 73 53.3 36 26.3 23 16.8 137 100 Village 63 18.4 12 3.5 208 60.8 59 17.3 342 100 Growth center 197 10.9 191 10.6 1,116 62.0 297 16.5 1,801 100 All 265 11.6 276 12.1 1,360 59.6 379 16.7 2,280 100 Source: BIDS Survey (2004). Note: In addition to the main authors, Rashid Faruqee and Hussain Samad were contributing coauthors of this chapter. 41 5580-CH05.pdf 41 5580-CH05.pdf 41 3/11/09 10:55:27 AM 3/11/09 10:55:27 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Growth-Center Microenterprises from the rural cooperatives or PBSs (Palli Bidyut Samities), and 18 percent from the Bangladesh Power Bangladesh has formally classified some 2,078 growth Development Board. Although only 38 percent had centers (large-sized markets).25 This study focused on phone connections, nearly 67 percent had access to rural enterprises randomly selected from 115 formal mobile phones. growth centers across 40 subdistricts (upazilas). Eighty percent of the selected growth centers were located 15 kilometers or more from the nearest district Types and Distribution headquarters, while subdistrict centers were also Growth-center microenterprises typically focused situated at a considerable distance (about 15 percent of on retail services. Trade and business (food and non- growth and upazila centers were indistinguishable). food) predominated, together comprising 61 percent Most growth centers had access to roads in fair of enterprises; these were followed by services (27 condition. percent) and manufacturing (10 percent). About 38 Concentration of enterprises varied greatly percent of rural enterprises were linked to agriculture, by growth center. Some centers had no more than mostly in the retail business. There was little regional 15 enterprises, while others had as many as 808. All variation (see Table 5.2). enterprises in the centers surveyed were self-standing Inthetradeandbusiness(food)category,groceries (none included household-based enterprises). and general stores predominated, followed by tobacco Seventy-five percent of the selected growth centers (biri), betel quid (paan), and beetle nut shops. In the had been recently established, as evidenced by trade and business (non-food) category, the main enterprise distribution by age of the growth centers. establishments were traders of pharmaceutical and None of those established within one year of the medical goods; cosmetics and toiletries; and textiles, survey involved manufacturing. Those established clothing, hosiery, footwear, and leather goods. In the 5 to 10 years earlier had the highest concentration of services category, tea stalls, restaurants, tailor shops, such enterprises. and beauty salons were the major businesses. The The study found that, compared to other rural manufacturing category mainly included non-grain areas, growth centers enjoyed more extensive access processing (e.g., jewelry, wood furniture, sawmills, to electricity. Nearly 94 percent of those studied andwood-processingunits).UnlikeotherSouthAsian had electricity; 76 percent obtained a connection countries, where rice mills are common, Bangladesh Table 5.2 able 5.2 Distribution of Growth-Center Microenterprises by Region Establishments by Region Chittagong Dhaka Khulna Rajshahi Total Enterprise Type No. % No. % No. % No. % No. % Trade and business (food) 153 33 160 33 142 34 145 31 600 33 Trade and business (non-food) 116 25 139 29 118 28 143 31 516 28 Services 128 28 120 25 123 29 123 26 494 27 Manufacturing 54 11 58 12 34 8 45 9 191 10 Source: Data International Survey (2002). 25The Government of Bangladesh, through its Engineering Department and local government, maintains a regularly updated list of rural growth centers. 42 5580-CH05.pdf 42 5580-CH05.pdf 42 3/11/09 10:55:27 AM 3/11/09 10:55:27 AM 5 Energy Consumption and Rural Production mills rice either in households or using mobile enterprises nationwide,27 it is nonetheless helpful threshing machines. to compare current findings with earlier data on The survey results compare favorably with those Bangladesh and other countries. For example, Islam of other recent studies. For example, the National (2001) notes that eateries and tea stalls dominated Private-sector Microenterprise Survey, conducted in earlier commercial activities in Bangladesh. This 2003, revealed a similar pattern of micro-, small-, and survey shows that such activities remain significant, medium-sized enterprises. It showed that 27 percent but others have emerged, reflecting a diversification of enterprises (excluding fisheries) were agriculture- of business operations. based;26 while 14 percent were in manufacturing, compared to 11 percent in this survey. Since the 2003 Energy Use survey included medium-sized enterprises, trade and Analysis of energy-consumption patterns in services sectors combined represented only about Bangladesh's growth centers underscores the need 50 percent of enterprises, while they predominated to consider the context of energy-services delivery in this survey. and the relative efficiency of the energy sources used. This survey showed that 95 percent of rural The survey confirms that firms use multiple energy enterprises have sole owner operators, while 5 percent sources, depending on their specific business needs are partner owned. Current owners initiated most of (see Table 5.3). the businesses, signifying their recent establishment. Nearly all firms use some form of electricity, Similarly, the 2003 survey found that 94 percent of mainly for lighting. If owners cannot access electricity enterprises had sole owners, of which 90 percent were from the cooperatives or national grid, they generate owner operated. their own or purchase it from small firms. Because This survey's findings revealed that all electricity is used more efficiently than other energy establishments are owned and operated by men. The sources, its consumption appears lower (9 percent). averageoperatingdayis12hours;unlikeagricultural Kerosene and diesel are major sources of backup activities, there is no peak business season. Rural lighting during power outages, while candles are microenterprises are small, run primarily by also used. owners (with an average of two employees). Even One surprising finding is that only 6 percent the size of manufacturing establishments is small of firms--111 out of 1,801 enterprises--use biomass (averaging only four employees). Similarly, the energy, mainly for heat and steam. Charcoal is used 2003 survey revealed an average of 3.3 workers for by only 4 percent of firms to meet similar needs (see all types of rural enterprises. This survey shows Table 5.4). that women are an extremely small percentage of Modern energy sources are used mainly for employees (1.5 percent); by contrast, the 2003 survey lightingandheating(includingthepreparationoffood reported that women represented about 9 percent for sale). Non-lighting uses include manufacturing of the work force. and catering services and powering of small The value of enterprise assets is limited. Only appliances. Use in drive-shaft power for productive one in six owns the land on which the operation purposes is not yet widespread. was established. Nearly 67 percent have no assets; that is, the total value of assets, on average, is roughly equivalent to the annual revenue of the Energy Costs enterprise. For most growth-center microenterprises, energy is Although the rural-enterprise composition a significant but not major cost, averaging less than of growth centers may not be representative of 5 percent of total operating expenses (from less than 26If fisheries had been included, agriculture would have comprised 32 percent, compared to 38 percent in the current survey. 27Limiting business coverage to the growth centers means that many common, larger-sized enterprises (e.g., brick kilns, cold storage, fisheries, poultry, and hand looms), some of which are energy intensive, are not captured. 43 5580-CH05.pdf 43 5580-CH05.pdf 43 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 5.3 able 5.3 Distribution of Energy Sources by Activity Type User Microenterprises (%) Trade and Trade and Users (no. of Business Business Energy Source firms) (food) (non-food) Services Manufacturing All* Biomass 111 3.8 0.2 15.6 5.2 6.2 Charcoal 73 0.0 0.4 6.3 20.9 4.0 Candle 1,034 58.2 64.1 52.6 49.2 57.4 Kerosene/diesel 986 59.8 39.7 59.9 66.0 54.7 Dry cell 196 11.2 13.6 8.9 7.8 10.9 Storage cell 57 2.5 4.3 4.0 0.0 3.2 Captive electricity 343 19.5 21.7 18.0 13.1 19.0 Grid electricity 1,472 79.0 86.6 79.6 82.7 81.7 Miscellaneous 56 3.2 4.1 2.0 1.6 3.1 None 26 2.7 0.8 1.0 0.5 1.4 Total 1,801 600.0 516.0 494.0 191.0 NA Source: Data International Survey (2002). * Sum of percentages is more than 100 because enterprises use multiple energy sources. Table 5.4 able 5.4 Major Energy Sources by Use Type Microenterprises that Use Fuels Use as Major Lighting Source Use as Major Non-lighting Source % of Total % of Energy % of Energy Energy source No. of Users Respondents No. of Users Users No. of Users Users Biomass 111 6 0 0 71 63 Charcoal 73 4 0 0 47 64 Candle 1,034 57 66 6 -- -- Kerosene/diesel 986 54 352 35 180 18 Generator electricity 343 19 122 35 0 0 Grid electricity 1,472 81 1,120 76 208 14 Source: Data International Survey (2002). 2 percent in the food trade to about 10 percent for Although energy comprises a comparatively services) (see Table 5.5). As already illustrated, not lesser share of total expenses, energy costs vary by all firms use all sources of energy for lighting and energy intensity of the business. For example, retail non-lighting purposes. If such a distinction is made, shops incur small lighting costs, while diesel- or the proportion of energy costs rises appreciably, electricity-powered heavy machinery shops must particularly for non-lighting uses. budget more for energy. Explaining such cost 44 5580-CH05.pdf 44 5580-CH05.pdf 44 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM 5 Energy Consumption and Rural Production Table 5.5 able 5.5 Energy as a Percentage of Total Costs by Activity Type Microenterprise Type (%) Energy Source No. of Firms Trade (food) Trade (non-food) Services Manufacturing All Firms Biomass* 110 3.9 1.8 5.3 4.3 4.9 Charcoal 71 0 16.2 11.0 1.9 5.8 Candle 1,034 0.2 0.3 0.7 1.0 0.5 Kerosene/diesel 985 1.0 1.3 4.7 2.6 2.4 Dry cell 180 0.1 0.2 0.8 0.1 0.2 Storage cell 57 0.3 0.8 1.9 0 1.1 Generator electricity 343 0.6 0.7 4.5 1.7 1.7 Grid electricity 1,472 0.8 1.4 5.4 6.1 2.8 All sources 1,775 1.7 2.3 10.1 8.3 4.9 Source: Data International Survey (2002). * Biomass sources include fuelwood, sawdust, tree residue, straw, rice husk, jute sticks, bagasse, and other crop residue, animal dung, and briquette. variations requires examining the level and effects electricity when it becomes available in their of relative energy pricing for various fuels used by respective communities and otherwise generate rural enterprises. their own electricity or purchase it locally from Energy cost is relative to what businesses pay small generator firms. for it. Predictably, based on standard energy units, the price of electricity is higher than that of all Entrepreneurial Attitudes other energy types, mainly because of the higher The study survey asked business owners and value of the energy services provided. Measured operators a variety of energy-related attitudinal by energy content, biomass is the cheapest source, questions. Findings show that business entrepreneurs while electricity is the most expensive. Although the consider electricity the best source of lighting, but price of charcoal is low, its use is specialized (e.g., also perceive that reliability of supply is a significant making jewelry and tobacco [biri]). Conversely, for problem. Few respondents consider kerosene superior lighting services (measured in kilolumen-hours), to electricity for lighting. Most agree that electricity electricity is the cheapest source (Tk 0.003 per provides better illumination than kerosene and makes klm-hr versus Tk 1.4 for kerosene and Tk 41.3 for reading easier. They also perceive that grid electricity candles). Indeed, for the same lighting service, is superior to batteries for powering television (see kerosene is about 45 times costlier than grid-based Table 5.7). electricity (see Table 5.6). Entrepreneurs perceive irregular electricity To summarize, patterns of energy choice are, supply and voltage fluctuations as a negative use in large part, determined by the relative fuel price factor. Some 67 percent of respondents are convinced for the desired end use. For heating, biomass is the that PBS power distribution and management least expensive fuel. For lighting, grid electricity is quality have declined in recent years. Indeed, nearly by far the cheapest source. Thus, it is not surprising 75 percent consider fatal accidents as a disadvantage that most rural microenterprises switch to grid of use. About 67 percent view electricity as expensive, 45 5580-CH05.pdf 45 5580-CH05.pdf 45 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 5.6 able 5.6 Price per Unit of Energy for Non-lighting Business Use Enterprise Type (Tk/unit) Energy Unit of Trade All Source Energy No. of Users Trade (food) (non-food) Services Manufacturing Enterprises kgoe 71 3.1 1.3 3.3 3.9 3.3 Biomass* klm-hr 0 -- -- -- -- -- kgoe 47 -- 3.5 4.2 4.9 4.6 Charcoal klm-hr 0 -- -- -- -- -- Kerosene/ kgoe 180 22.6 23.0 22.3 23.5 22.7 diesel klm-hr 301 1.3 1.0 1.4 1.5 1.4 Grid kgoe 208 55.9 86.4 66.1 53.4 66.3 electricity klm-hr 1,173 0.04 0.02 0.03 0.03 .03 Candles klm-hr 72 48.2 41.8 39.2 37.9 41.3 Source: Data International Survey (2002). * Biomass sources are fuelwood, sawdust, tree residue, straw, rice husk, jute sticks, bagasse, and other crop residue, animal dung, and briquette. Table 5.7 able 5.7 Perceived Advantages and Disadvantages of Electricity Use Response Choices (%) Survey Statement Strongly Agree Agree Indifferent/Neutral Strongly Disagree Disagree Electricity provides better illumination than kerosene oil (n = 1,801) 70.7 28.9 0.1 0.2 0.0 Reading is easier with electric lamp than kerosene lamp (n = 1,801) 64.4 34.8 0.8 0.1 0.0 It is [more] advantageous to run TV by electricity than battery (n = 1,766) 60.9 35.1 3.5 0.2 0.3 Source: Data International Survey (2002). and most consider it an unattractive option for 50percentdidnotconsiderprevailingratesexorbitant, cooking, probably because of its high cost compared 30 percent thought they were too high, and about to biomass fuel. However, nearly all respondents agree 19 percent were indifferent. This finding is consistent that the advantages of using electricity far outweigh with respondents' reactions to a similar statement the disadvantages. on affordability of connection. About 59 percent of Enterprises without electricity were asked non-users said that, if they were to obtain access, several questions related to the costs and benefits the monthly payments would not be too high; about of electricity use. There was no consensus on the 44 percent perceived it as unaffordable. In short, cost of obtaining a business connection: About despite the overwhelming perception regarding net 46 5580-CH05.pdf 46 5580-CH05.pdf 46 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM 5 Energy Consumption and Rural Production Table 5.8 able 5.8 Perceived Advantages and Disadvantages of Fuelwood Use Response Choices (%) Survey Statement Strongly Agree Agree Indifferent/Neutral Strongly Disagree Disagree Fuelwood is readily available to collect (n = 1,801) 31.2 33.3 0.3 17.9 17.3 Fuelwood can be obtained free of cost from own trees (n = 1,801) 30.3 32.6 2.3 15.1 19.8 Fuelwood creates smoke that creates breathing problems (n = 1,801) 40.1 45.1 5.1 4.3 5.3 Cooking with fuelwood creates health hazards (n = 1,750) 25.0 33.4 10.5 16.6 14.5 Fuelwood is very expensive for cooking (n = 1,798) 28.3 32.5 1.6 15.3 22.3 Source: Data International Survey (2002). gains, most respondents believed the advantages of expensive, but few respondents were current users, electricity outweighed the problems. reflecting the perception of high price versus actual With regard to fuelwood, the opinions expressed cost. Many answers to LPG-related questions were varied markedly, reflecting the widespread problem indifferent, suggesting that most respondents were of deforestation and thus uneven resource availability. unaware of the fuel (see Table 5.9). Because of scarcity and moderate shortages, about 50 The picture that emerges from this survey is percent considered fuelwood expensive. At the same that most growth-center microenterprises in rural time, many rural entrepreneurs perceived fuelwood Bangladesh are not energy intensive. They consist as a readily available resource. More than 90 percent mainly of small retail stores and shops, with some expressed concern with regard to deforestation. Rural manufacturing. The main energy use is lighting, entrepreneurs preferred fuelwood to straw, dung, for which virtually all businesses use electricity. and other biofuels, which were used sparingly. But If a particular growth center does not yet have about 85 percent of respondents were aware that the electricity, business owners organize their supplies, smoke emitted from fuelwood could cause respiratory either by buying their own generator or purchasing problems, and some 50 percent agreed that it may small amounts from a local firm. Small agricultural cause other health problems (see Table 5.8). industries and restaurants that require energy for heat Respondents' opinions differed on the use rely mainly on biomass or kerosene. of kerosene and liquefied petroleum gas (LPG). Kerosene, although readily available, was viewed as a health hazard. Rural entrepreneurs perceived its price Home and Village Enterprises as relatively high, but were unclear why. More than The survey showed that many rural microenterprises 55 percent considered LPG a good source of cooking are located outside the growth centers. Out of 2,400 energy. About 38 percent considered LPG as very households, 137 had home-based enterprises; in 47 5580-CH05.pdf 47 5580-CH05.pdf 47 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 5.9 able 5.9 Perceived Advantages and Disadvantages of Kerosene Use Response Choices (%) Survey Statement Strongly Agree Agree Indifferent/Neutral Strongly Disagree Disagree Cooking with kerosene is easy 17.2 24.5 4.1 30.6 23.6 (n = 1,801) Kerosene is not expensive for lighting 14.1 21.6 2.2 38.6 23.4 (n = 1,801) Kerosene is very expensive for cooking 59.6 30.5 3.8 3.4 2.6 (n = 1,801) Cooking with kerosene creates health hazards 46.8 35.7 11.2 3.1 3.2 (n = 1,801) Source: Data International Survey (2002). Table 5.10 able 5.10 Microenterprise Distribution by Location and Length of Operation Years in Home-based Village-based Total Operation No. % No. % No. % < 1 11 8.1 58 17.2 69 14.2 1­3 26 19.3 91 27.0 117 24.8 3­5 14 10.4 48 14.2 62 13.1 5­10 29 21.5 75 22.3 104 22.0 > 10 55 40.7 65 19.3 120 25.4 Total 135 100.0 337 100.0 472 100.0 Source: BIDS Survey (2004). addition, 342 village-based enterprises were located decade or longer, although interesting differences outside rural households. are evident between categories. For example, about 40 percent of home-based enterprises are more than 10 Types and Distribution years old, twice the proportion of village-based ones. Few home-based enterprises have been in operation As already mentioned, most home-based businesses for one year or less, while more than 17 percent of are women run and involve manufacturing (basket village enterprises have been recently established (see weaving and sewing), while village industries, run Table 5.10). Home-based manufacturing enterprises mainly by men, focus more on trading (selling of likely involve handicraft activities carried down by goods and services). The age of home and village families over generations. microenterprises is spread over a number of years. Some 50 percent of home- and village-based More than 25 percent have been in operation for a microenterprises are run by only one person, although 48 5580-CH05.pdf 48 5580-CH05.pdf 48 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM 5 Energy Consumption and Rural Production Table 5.11 able 5.11 Microenterprise Distribution by Number of Employees and Location Workers in Peak Home Village All Firms Periods (no.) No. % No. % No. % 1 40 30.7 182 58.3 222 50.2 2 41 31.5 93 29.6 134 30.3 3­5 43 33.0 34 10.9 77 17.4 > 5 6 4.0 3 0.9 9 2.0 Total 130 100 312 100 442 100 Source: BIDS Survey (2004). differences between categories are apparent. During while village enterprises use non-fuelwood biomass; peak periods of operation, about 31 percent of home cow dung accounts for 8 percent of village enterprises' enterprises employ only one person, compared to energy basket, compared to only 1 percent for home some 58 percent of village enterprises. In fact, home enterprises. Similarly, village enterprises derive enterprises are more evenly distributed in terms of size 49percentoftheirenergyfromcropresidue,compared ofoperation(proxiedbynumberofworkersduringpeak to only 26 percent for home enterprises (see Figure 5.1). periods).Thisfindingreinforcesthedistinctionbetween A major reason is that manufacturing businesses, manufacturingcenteredmainlyinhomeenterprisesand mostlyhome-based,requirebetterheatcontrolandare trading in village-based ones (see Table 5.11). thus more likely to use superior forms of biomass. A portion of the needed resources may be collected from Energy Use and Costs the owner's homestead, implying an easier supply for The survey findings show that more than 70 percent of home-based, versus village, enterprises. homeenterprisesandvirtuallyallvillageenterprises-- Findings also reveal differences in other portions most of which operate during extended evening of the energy basket. With regard to liquid fuels, home hours--use energy for lighting. Beyond lighting, enterprises use more diesel than kerosene; conversely, however, these small handicraft and retail firms use village enterprises use more kerosene than diesel. little energy. About 67 percent of home enterprises Home enterprises, on average, consume only about and 64 percent of village ones use no energy for non- 6 kgoe of kerosene, compared to nearly 68 kgoe for lighting purposes. Some report using no energy types village enterprises. With regard to electricity, village for either lighting or non-lighting purposes, implying enterprises use more than 30 kgoe, compared to only thattheirbusinessesinvolvemanuallaborandoperate 3.9 kgoe for home enterprises (see Figure 5.1). Clearly, mainly during the daytime. trade-oriented village enterprises require more Compared to village enterprises, home-based lighting services than do home-based businesses. businesses use significantly less energy; on average, Given that non-lighting energy use predominates home enterprises consume 332 kgoe per year, in home and village-based enterprises--as evidenced comparedto608kgoeforvillage-basedones(including by the critical role biomass plays in the energy balance non-users). Predictably, home businesses depend of both--it is of interest to compare the output and more on biomass than do village enterprises. But costs of energy sources for non-lighting and lighting surprisingly, within the biomass portion of the energy uses (see Table 5.12). On average, village enterprises basket, home enterprises depend more on fuelwood, consume more energy per enterprise for non-lighting 49 5580-CH05.pdf 49 5580-CH05.pdf 49 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Figure 5.1 Figure 5.1 Comparison of Energy Use Structure by Enterprise Location Source: BIDS Survey (2004). Table 5.12 able 5.12 Comparison of Energy Output and Cost by Use and Location Home Village Lighting Non-lighting Lighting Non-lighting Amount Amount Amount Amount (klm-hr/ Price (kgoe/ Price (klm-hr/ Price (kgoe/ Price Energy Source year) (Tk/klm-hr) year) (Tk/kgoe) year) (Tk/klm-hr) year) (Tk/kgoe) Biomass 0.00 -- 317.5 4.34 0 -- 546.2 3.68 Candle 0.32 43.33 0 -- 1.1 54.68 0 -- Kerosene or diesel 23.40 79.15 40.7 29.11 37.7 17.11 95.1 32.08 Grid electricity 461.50 0.77 14.2 185.62 2,877.0 0.27 16.0 55.52 All sources 474.20 17.92 372.4 27.21 2,915.8 10.50 657.2 29.20 Source: BIDS Survey (2004). purposes than do home enterprises (657 versus 372 biomass is by far the least expensive source, which kgoe per year). Indeed, for each enterprise type, accounts for its dominance as a heating fuel. Because village enterprises consume higher levels of energy. village enterprises are larger than home-based ones, Furthermore, the lighting services consumed by they use significantly more biomass energy for non- village enterprises are more than six times that of lighting uses. home-based enterprises, suggesting the importance The price for lighting is similar to the results of lighting for village businesses and the small size presented for rural households in Chapter 3. For home of home enterprises, some of which lack access to enterprises, the average price of electric lighting is electricity. Tk 0.27 per klm-hr and more than Tk 40 per klm-hr for Thepriceofenergyperkilogramsofoilequivalent candles and kerosene; findings show similar prices for varies markedly by fuel source. For non-lighting uses, village-based enterprises. Given the wide disparity in 50 5580-CH05.pdf 50 5580-CH05.pdf 50 3/11/09 10:55:28 AM 3/11/09 10:55:28 AM 5 Energy Consumption and Rural Production the price of lighting, it is not surprising that businesses Table 5.13 able 5.13 use more electricity when they need lighting; electric Summary Statistics of Outcomes and Explanatory lights are used primarily, with kerosene lamps and Variables of Enterprise Energy Demand Regressions candles as backup sources. The benefits of switching to a higher-tier source Variable Mean along the energy ladder extend beyond profitability, Enterprise energy demand particularly for lighting services. The cost of one Fuelwood (kg/month) 73.25 unit of lighting service from electricity is far cheaper Kerosene (liter/month) 164.72 than that from kerosene. For home and village-based Diesel (liter/month) 60.39 enterprises, the respective costs of electric lighting are Electricity (kWh/month) 218.58 Tk 0.77 and 0.27 per klm-hr. The corresponding costs Explanatory for kerosene lighting are Tk 30.40 and 17.11 (nearly 40 Enterprise assets and 63 times greater). Clearly, a major reason for the Land (acres) 26.34 higher profitability of electricity-based enterprises Non-land (Tk 10,000) 1.25 involves such cost differences for lighting. Village/growth center price Fuelwood (Tk/kg) 1.36 Enterprise Energy Demand and Kerosene (Tk/liter) 19.71 Profitability: A Quantitative Analysis Diesel (Tk/liter) 21.39 Electricity (Tk/kWh) 4.50 Rural microenterprises, like rural households, are If the village/growth center has electricity 0.89 influenced by a range of enterprise- and community- Sources: Data International Survey (2002) and BIDS Survey level factors when selecting from among alternate (2004). energysources.Theirchoices,inturn,eventuallyaffect the enterprise outcome; that is, revenue and profit. Enterprise factors may include years that a business has been in operation, assets, and types of activities; higher price of electricity outweighs the positive community factors may include prices of alternate relative price effect of electricity on demand for energy sources and infrastructure variables. Table 5.13 diesel. Thus, diesel and electricity remain substitutes provides summary statistics of energy variables and in energy use by rural enterprises. The demand for selectedinfluencingvariables,whileTable5.14presents kerosene, which appears unaffected by these factors, regression results of energy demand.28 is included here for completeness (see Table 5.14). An enterprise's non-land assets have a significant Given the demand for alternate types of energy in positive effect on consumption of biomass (fuelwood), non-farm production, one can estimate the effects of diesel fuel, and electricity. Although the price of energy use on enterprise revenue and profit. Energy fuelwood has no effect on any type of energy source, use is expressed in taka value adjusted by village the prices of diesel and electricity significantly and growth-center price indices. Table 5.15 provides affect energy demand. The price of diesel positively summary statistics and Table 5.16 regression results.29 affects electricity consumption, thereby showing that As Table 5.16 shows, energy consumption positively diesel and electricity are substitutes in enterprise affects enterprise profit. One taka worth of energy consumption. If this statement is true, however, a consumption increases enterprise profit by more higher electricity price should lead to greater diesel than half a taka. When the effect is differentiated consumption.Butresultsshowanegativeeffect,which by energy source, one discovers that kerosene/ is possible if the negative income effect caused by the diesel consumption has no effect on either profit or 28Table A2.4 (see Annex 2) presents the complete regression for energy demand. 29Tables A2.5 and A2.6 (see Annex 2) provide complete regressions. 51 5580-CH05.pdf 51 5580-CH05.pdf 51 3/11/09 10:55:29 AM 3/11/09 10:55:29 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 5.14 able 5.14 Estimates of Enterprise Energy Demand (N = 2,290) Energy Demand (monthly) Explanatory Variable Fuelwood Kerosene Diesel Electricity Enterprise assets Land (acres) -0.15** 0.04 -0.04* 0.11 Non-land (Tk 10,000) 3.99** 1.14 6.12** 25.51** Village/growth center price Fuelwood (Tk/kg) 5.25 1.17 0.93 -10.46 Kerosene (Tk/liter) 11.43* -1.03 -3.76 -14.56 Diesel (Tk/liter) -4.79 -11.17 3.29 23.38* Electricity (Tk/kWh) -6.07 -15.87 -13.26* 25.68 If village/growth center has electricity 10.47 -26.91 -18.19** 418.72** Source: Data International Survey (2002) and BIDS Survey (2004). Note: * = significance level of 10 percent; ** = significance level of 5 percent or stronger. Figures represent changes in energy demand caused by unit changes in explanatory variables. Table 5.15 able 5.15 Table 5.16 able 5.16 Summary Statistics of Outcomes and Energy Use Effects of 1 Tk Increased Energy Use on Enterprise Variables of Enterprise Profitability Regressions Outcome (annual Tk per enterprise) (N = 2,290) (N = 2,290) Energy Source (Tk/yr)a Revenue Profit Variable (Tk/year) Mean Model 1 Outcome Total energy use 0.86 0.55** Revenue 555,817.6 Model 2 Profit 72,590.8 Energy use Energy Biomass 5.72* 1.22** Total use* 6,674.82 Kerosene -3.82 -0.27 Biomass 1,129.62 Electricity 4.11 1.35** Kerosene 2,701.26 Source: Data International Survey (2002). Electricity 2,622.92 Note: * = significance level of 10 percent; ** = significance level of 5 percent or stronger. Figures represent changes in outcome Source: Data International Survey (2002). (consumption and income) caused by unit changes in explanatory * Includes the cost of miscellaneous sources, in addition to the variables. three major ones. a Total energy cost used in Model 1 is slightly higher than the sum of individual energy costs used in Model 2 because it includes other miscellaneous sources. revenue, but biomass and electricity have a significant positive effect, especially on profit. One taka spent on biomass consumption increases profit by Tk 1.22; a similar amount spent on electricity increases it by The conclusion is that use of either high-quality Tk 1.35. Thus, the effect is slightly higher for electricity biomass or purchased electricity dramatically affects than for biomass as an energy source in enterprise the profitability of rural Bangladeshi enterprises. To income generation; that is, enterprises would gain by reiterate, biomass is usually the least expensive fuel switching from biomass and diesel to electricity as an and is used primarily for heating, while electricity energy source in production. is used for lighting and other purposes. Thus, 52 5580-CH05.pdf 52 5580-CH05.pdf 52 3/11/09 10:55:30 AM 3/11/09 10:55:30 AM 5 Energy Consumption and Rural Production energy source matters in the profitability of rural methodsthatdependonmodernenergyusefarcostlier, businesses. but more efficient, equipment for large tracts. The three basic equipment pieces are low-lift pumps for surface water irrigation, surface tube wells (including deep- Energy for Farming set shallow tube wells), and deep tube wells. These may be run with either a diesel- or electricity-operated Farming is a critical energy use for rural households. motor. Low-lift pumps are owned either privately or The country's potential for increased irrigation, collectively. Surface tube wells are privately owned; though not as developed as that of other South Asian while deep tube wells are usually owned collectively countries, is significant. The most obvious energy (for reasons of higher cost, as well as custom). inputs are diesel engines and electric motors for After irrigation are harvesting and transport. pumping irrigation water. In addition, a variety of Most harvesting is done manually. Threshing, small machines, including power tillers, are used traditionally done manually, depends more on increasingly in land preparation, alleviating much of diesel- or electricity-powered machines. Husking the drudgery associated with such work. increasingly uses machines that run on diesel and occasionally kerosene. Crops may be transported to Farming Stages and Energy Inputs market by bullock or by using mechanized methods AsdiscussedinChapter4,Bangladeshischaracterized that depend on petroleum fuels. by three rice-cropping seasons: (1) aus (rainfed summer rice planted in March­April), (2) aman Land Tillage Methods (rice planted in July­September and harvested in Over the past decade, acreage under mechanized November­December), and (3) boro (irrigated winter tillage has risen significantly. In 1996, 4.2 million rice planted in December­February and harvested acres--20 to 25 percent of the country's cultivated in April­June).30 Aus requires little energy for land--was tilled using diesel- or petroleum-powered tillage (except bullock power) and irrigation; boro tillers or tractors. When the 2004 household survey necessitates energy for both tillage and irrigation, and was conducted, that proportion had risen to more aman falls in between. than 66 percent, although regional variations were The major rice-farming stages involve land substantial (see Table 5.17).32 preparation, irrigation, harvesting, and transport to Although use of power tillers and tractors is market.31 During land preparation, bullock-driven common, ownership is not. In the study sample, only ploughs or power tillers/tractors are used. While 35 farmers owned mechanized tillage equipment, bullock power may be quantified, Bangladesh has while 1,068 rented it; another 186 households rented little relevant data available. Power tillers/tractors draft bullocks, while the remainder used their own require petroleum fuel (mainly diesel). bullocks. Irrigation, the next major stage, overlaps with land preparation; that is, when rice is transplanted, the land Irrigation Methods must be thoroughly flooded and puddled. Irrigation Mechanized irrigation using modern energy is of is also required at later stages of rice-growing. The growing importance in rural Bangladesh. In 2002­ capacitytoirrigatelandislimitedtomanualimplements 2003, 73 percent of the 11.68 million acres irrigated (e.g., swing baskets and hand tube wells). Mechanized used mechanized tube wells; another 17 percent used 30Details are available from the International Rice Research Institute (www.irri.org/irrc/weeds/closing.asp). 31To simplify data collection and analysis, this study emphasized farmers who directly manage their own land. Thus, analysis anddiscussion on energy use and its characteristics are tentative, as key data may be lacking. More definitive answers call for more in-depth studies. 32In 2005, a large-scale survey found that, since 1996, the number of power tillers and tractors had risen from slightly more than 150,000 to 711,000 (BBS 2005). 53 5580-CH05.pdf 53 5580-CH05.pdf 53 3/11/09 10:55:30 AM 3/11/09 10:55:30 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table 5.17 able 5.17 Comparison of Tillage Methods by Division Division (%) Tillage Method Chittagong Dhaka Khulna Rajshahi Total Power tiller/tractor 45.9 75.3 61.5 86.6 66.1 Draft bullock 54.1 24.7 38.5 13.4 33.9 Source: Village censuses (2004). Table 5.18 able 5.18 Irrigation Method and Energy Type by Division Division (%) Method or Type Chittagong Dhaka Khulna Rajshahi All Divisions Irrigation Mechanized 52.0 78.5 53.4 91.5 68.1 Hand tube well 1.1 1.2 0.9 0.8 1.0 Indigenous 10.1 15.0 19.8 1.1 11.2 method None 36.8 5.3 25.9 6.6 19.7 Energy Diesel 50.0 93.3 66.7 76.7 70.4 Electricity 19.2 6.7 -- 20.0 10.4 Diesel and 2.6 -- 7.4 3.3 3.2 electricity Not applicable 34.2 -- 25.9 -- 16.0 Source: BIDS Survey (2004). power pumps, while 3 percent used canals (or in The village survey found that 68 percent of combination with power pumps) and the remainder villages used mechanized irrigation methods. Of traditional methods (BBS 2005). these, 70 percent used diesel power, 10 percent Although most mechanized irrigation equipment used electricity, and 3 percent used both (see is diesel-driven, the percentage division between Table 5.18). diesel and electricity is unclear. In 2005, the country Of the 1,459 farm households surveyed, only had 7.54 million mechanized irrigation sets (including 62 percent irrigated their land (see Table 5.19). Of low-lift pumps, shallow tube wells, deep-set shallow these, 97 percent used mechanized means. Only 130, tubewells,anddeeptubewells)(BBS2005).According or 9 percent of all respondents, owned mechanized to the Rural Electrification Board, in February 2006, irrigation sets.34 Ownership of mechanized pumps irrigation connections totaled slightly more than was concentrated in Rajshahi--77 out of 130 farming 195,000.33 Even assuming that one connection could households--where 18 percent of such households serve more than one set (which is seldom the case), owned modern irrigation equipment. But in Dhaka, electrically driven irrigation sets are uncommon. half of non-farmers owned irrigation sets. 33Details are available at www.reb.gov.bd. 34Several households owned more than one piece of equipment. 54 5580-CH05.pdf 54 5580-CH05.pdf 54 3/11/09 10:55:30 AM 3/11/09 10:55:30 AM 5 Energy Consumption and Rural Production Table 5.19 able 5.19 Farmer Irrigation Method by Division Division Chittagong Dhaka Khulna Rajshahi Total Method No. % No. % No. % No. % No. % Manual 1 0.3 12 3.7 9 2.8 1 0.2 23 1.6 Mechanized 159 40.3 203 63.2 158 48.5 356 85.4 876 60.0 All 160 40.5 215 67.0 167 51.2 357 85.6 899 61.6 Source: BIDS Survey (2004). Table 5.20 able 5.20 Energy Consumption in Farming (average owner-user households per year) Electricity Diesel Consumption Cost (Tk) Consumption Cost (Tk) Equipment Type Yearly kWh Households Yearly Per kWh Yearly liters Households Yearly Per liter Irrigation pump 3,643.47 9 9,762 2.8 347.95 103 7,056 20.5 Power tiller/tractor -- --- -- -- 238.24 35 4,799 20.2 Thresher -- -- -- -- 78.13 17 1,698 20.8 Source: BIDS Survey (2004). Fuel Choice for Irrigation for purchased irrigation water in rural areas. As Asnotedpreviously,mechanizedirrigationequipment expected, Rajshahi had the highest proportion of may be powered by electricity or diesel. According farmerownershipofirrigationequipment.Onaverage, to available information, only 112 of current farm a farm household required about 348 liters of diesel households use their machines for land under their per year for irrigation, reflecting an annual expense of control (i.e., self-cultivated owned land and other land nearlyTk7,056.Thus,irrigationisextremelyimportant under lease agreement or other form of user rights). Of for farm productivity, as well as profitability from the these, 103 are operated by diesel and 9 by electricity sale of irrigation water to neighbors. (8 of which are in Rajshahi). It is also noteworthy that Clearly, farm households that could benefit from about 70 percent of diesel-pump users are in villages electricity-powered irrigation have not used electricity without electricity. Average electricity consumption as extensively as they might, perhaps, in part, because per farm and the costs borne by farmers indicate an of their remote location. Further research is needed to expense of nearly Tk 2.8 per kWh, far lower than for identify the factors that have prevented farmers from domestic purposes (Tk 7.7 per kWh in Rajshahi and adopting electricity-driven pumps, particularly given Tk 5.5 for all divisions). Average diesel consumption the difficulty of moving large diesel pumps between per farm and farmer costs are slightly more than fields and deciding whether government-subsidized Tk 20 per liter. The cost per liter is similar across types electricity for irrigation is a sound policy. of use, but varies slightly by division (see Table 5.20). The study found that many farmers used water Summing Up from mechanized irrigation equipment, whether owned or leased. The percentages were higher for In rural Bangladesh, microenterprise entrepreneurs farmers owning pumps, reflecting the active market own and operate a range of businesses, from 55 5580-CH05.pdf 55 5580-CH05.pdf 55 3/11/09 10:55:30 AM 3/11/09 10:55:30 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities trade and commercial services to small-scale Increasingly, they view traditional lighting sources as manufacturing. Although small, these enterprises inefficient and of poor quality. The lighting analysis represent a substantial increase in rural people's confirms that candles and kerosene lamps are many incomes. The three types of microenterprises times more expensive than electric lighting. examined in this chapter--growth-center, village, and Microenterprises with electricity enjoy benefits home-based--use a wide array of energy sources. All andcostadvantagesrelatedtolightingandprofitability. rural businesses in the growth centers use electricity, Indeed, the average cost of electric lighting can be 60 times less expensive than kerosene, the next either from the grid or local generators; and most have cheapest source. Unreliability of electricity supply, electric lighting. By contrast, not all free-standing however, forces businesses to depend on kerosene village and home-based enterprises have access and diesel as backup lighting sources. to electricity; those that do may choose not to use Thesurveyfindingsindicatescopeforimprovement electricity, even for lighting. Such enterprises depend in the energy-use patterns of rural Bangladeshi heavily on biomass energy, mainly for heating and microenterprises.36 Analysis of the relationship manufacturing needs. Though inconvenient, biomass between modern energy sources, as defined by the is the least expensive heating fuel, a major factor in energy ladder concept, indicates that rural businesses explaining rural businesses' continued reliance on it. that use modern energy generally have more revenue Another factor is that most rural areas have little or and are more profitable than those that rely on no access to LPG.35 traditional energy sources. Thus, a viable rural energy Bangladeshi entrepreneurs perceive that biomass strategy for Bangladesh must consider the best ways is inexpensive compared to other fuels and that it to promote modern energy, as well as the modern use can be collected from the environment. At the same of biomass. To this end, it is critical to examine the time, they are aware of the health problems associated institutional underpinnings of energy distribution, with traditional fuelwood use. They also agree that which are the subject of the next chapter. local biomass collection is becoming more difficult. 35Generally unavailable in rural areas, LPG uses 60 percent of cooking energy, compared to only 15 percent for traditional biomass stoves. 36Although energy for non-lighting use is limited, improvements are possible through the use of charcoal and more efficient kerosene stoves. 56 5580-CH05.pdf 56 5580-CH05.pdf 56 3/11/09 10:55:30 AM 3/11/09 10:55:30 AM 6 Institutional Framework for Rural Energy Supply Promoting better rural energy policies in Bangladesh Institutional Challenges and Potential is inextricably linked to national poverty reduction and rural development. Extending both grid and off- To advance the development goals of poverty grid electricity systems improves rural livelihoods reduction and human development, the institutions and enables delivery of critical human services, from responsible for delivery of modern energy services clean water and sanitation to health and education. must be responsive to the unique needs of rural The results of this study demonstrate that access populations. The inefficiencies associated with to improved biomass stoves, electric lighting, and Bangladesh's current energy-sector providers, which mechanized tillage and irrigation creates the enabling have shifted financial burden to the government and conditions for informal and small-enterprise sectors rural consumers, call for more effective institutional to grow, thereby empowering the poor. 37 arrangements and coordination. This chapter reviews Rural energy must be viewed as a basic input the current structure of Bangladesh's energy sectors, to the rural economy, in line with its role in rural highlighting past institutional and policy deficiencies productivity and income generation. Achieving a and opportunities for building a more integrated supply-and-demand balance is critical at all levels.38 approach to meeting current and future rural energy Given rural residents' heavy reliance on traditional demand. biomass energy, an urgent priority must be to increase biomass supply. To achieve a sustainable supply, Biomass Sector biomass must be used in more efficient ways that This study suggests that Bangladesh's heavy mitigate damage to human and global environmental reliance on the unsustainable use of traditional health. Complementing these efforts, more diversified biomass, especially fuelwood, represents a significant modern energy sources must be made available to opportunity cost and serious health risk for millions fulfill unmet rural household and business demand. of the rural poor. Compounding their diminished To this end, appropriate pricing policies are vital: quality of life is the accelerated commercialization They must be market based to ensure that suppliers of all forms of biomass--including low-quality tree can sustain the higher cost of rural operations,39 yet leaves and grass.41 This section highlights lessons remain affordable to the poor.40 from past and ongoing efforts to improve biomass Note: In addition to the main authors, M. Iqbal, M. Khaliquzzaman and Grayson Heffner were contributing coauthors of this chapter. 37In 2005, Bangladesh's National Strategy for Accelerated Poverty Alleviation set forth eight strategic priorities, most of which imply a role for promoting better energy policies to improve rural access. 38At the local level, for example, transportation costs and seasonal irrigation requirements call for a degree of local supply­demand balance or self-reliance. 39The government's current pricing of electricity, oil, and gas is far below the cost of supply. 40In deciding whether subsidies are required for particular types of energy or end users, policy makers must consider both the social and economic costs and periodically revisit the results as these factors change. 41Paradoxically, the rural buying and selling of biomass also creates jobs. Indeed, employment from trading biomass exceeds that from trading substitute fuel products (e.g., kerosene, LPG, and electricity). Thus, in the push to promote modern energy, one must also consider the potential loss to rural employment, which could lead to even greater poverty. 57 5580-CH06.pdf 57 5580-CH06.pdf 57 3/11/09 10:55:46 AM 3/11/09 10:55:46 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities energy efficiency to increase rural productivity and · Bangladesh University of Engineering and quality of life. Technology · Grameen Shakti Improved Biomass Stoves · Bangladesh Rural Development Board From the late 1980s to 2001, the Institute of Fuel · Bangladesh Rural Advancement Committee Research and Development promoted improved · Proshika cookstoves through several donor-supported projects. Although an estimated 400,000 improved stoves were Although large-scale plant production appears distributed, commercialization failed, in large part, economically and financially feasible, several key because projects focused on targets and subsidies issues must be addressed. First, it is unclear whether rather than monitoring. Today the potential market the subsidy culture that characterized past programs for improved stoves is enormous. Better-designed will create future difficulties. Second, it is likely that stoves could save millions of days in lost productivity, biogas plants will benefit better-off households more decrease early mortality, and save energy. But to than the poor (Ghimire 2005). Third, the number of succeed, programs must promote stoves that people potential plants is unknown.42 want. Thus, future institutional support should focus on stove design, manufacture, and testing, as well as Power Sector training and quality monitoring. Bangladesh'spowersectorhaslaggedoveralleconomic Biogas Production growth, resulting in chronic unmet energy demand. Only 40 percent of the total population and less than 30 Basic biogas technology holds special promise percent of rural residents have access to electricity. Low for rural Bangladesh. Biogas can be harnessed for connection rates and unreliable power supply result cooking, lighting, and mechanized irrigation through in fewer opportunities for economic development or village-based production and distribution networks. improved rural livelihoods. Though rural households Such arrangements can enhance rural residents' self- grasp the benefits of electricity, they perceive it as sufficiencyduringperiodsofloadsheddinganddiesel expensive, in part, because of frequent power outages. shortages. Currently, tens of thousands of household- As Figure 6.1 illustrates, total effective power- and village-level biogas plants are in place throughout generation capacity has fallen below electricity the country. demand, resulting in extensive daily load shedding. Over the past 30 years, an array of government In turn, load shedding has obscured the reality of agencies, university research institutions, and non- unmet demand. Over the next nine years, electricity governmentalorganizations(NGOs)haveparticipated demand is expected to double (at 8 percent annual in the development, piloting, and implementation of growth). The investment required to generate household- and enterprise-scale biogas programs. sufficient capacity to accommodate future demand Key entities are: growth is immense. · In response to this challenge, the state-owned Rural Electrification Board · BangladeshPowerDevelopmentBoardhasputforward Local Government Engineering Department · anambitiousexpansionplancallingforinstallationofan Bangladesh Council for Scientific and Industrial additional4,210MWofgenerationcapacityby2010.The Research and Institute of Fuel Research and plan, which relies on natural gas, consists of combined- Development (a Council agency) · cycle, gas-turbine baseload units and simple-cycle, gas- Renewable Energy Research Centre (Dhaka turbine peaking units (Nexant 2006). University) 42Although the daily cow-dung input for cooking three meals a day is known, reports conflict regarding the number of cattle owned by rural households (BBS 2005). Anecdotal evidence suggests seasonal, rather than year-round, ownership. In addition, use of power tillers and tractors is widespread. 58 5580-CH06.pdf 58 5580-CH06.pdf 58 3/11/09 10:55:46 AM 3/11/09 10:55:46 AM 6 Institutional Framework for Rural Energy Supply Figure 6.1 Figure 6.1 Annual Load Shedding, 1993­2004 Source: Nexant (2006). Power-sector reform, initiated in 1994, envisions purchasingfrompublicandprivategeneratorsand the separation of sector regulation from ownership selling to distributors. and operation and independent, commercially · PGCB. The Power Grid Company of Bangladesh, oriented sector entities. It calls for the unbundling a wholly owned subsidiary of the BPDB, operates of generation, transmission, and distribution the national transmission grid, schedules grid functions and increased private-sector participation operations, and wheels energy to distributors. in generation and distribution. Formerly a monopoly · DESA. The Dhaka Electricity Supply Authority operator,theBangladeshPowerDevelopmentBoardis (not yet functional) distributes energy and being converted into a holding company. The Energy conducts commercial operations in Dhaka RegulatoryCommission, established in2004, istoplay and adjoining areas, except for Mirpur and a vital role in a range of sector operations, as noted Gulshan. in the following sections, but has been hampered by · DESCO. The Dhaka Electric Supply Company slow organizational approval processes and lack of distributes energy and conducts commercial funding. operationsintheMirpurandGulshanjurisdictions Key power-sector entities and their functions are of the Dhaka metropolitan area. as follows: · REB. The Rural Electrification Board oversees operations of consumer-owned rural electric · MPEMR. The Ministry of Power, Energy, and cooperatives (PBSs). It performs supervisory and Mineral Resources oversees sector management. regulatorydutiestoensurethattechnicalstandards · ERC. The Energy Regulatory Commission has are met and performance is monitored. authority over consumer protection, approval · WZPDC. The West Zone Power Distribution of tariffs and pricing, issuance of generation Company is responsible for regional distribution and distribution licenses, and promotion of in Khulna. competition. · IPP. There are some independent power suppliers · Power Cell. Within the MPEMR power division, (IPP) in Bangladesh that sell electricity to the the Power Cell oversees power-sector reform. transmission company. · BPDB. The Bangladesh Power Development Board · EA. Electrical Advisor. operates most publicly owned generators and · CEI. Chief Electric Inspector. some urban distributors; it acts as a single buyer, 59 5580-CH06.pdf 59 5580-CH06.pdf 59 3/11/09 10:55:46 AM 3/11/09 10:55:46 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Over the next several years, the newly constituted into 70 cooperatives or PBSs (Palli Bidyut Samities), Rural Power Company, which runs a 140-MW whichbenefitanestimated6.78millionhouseholds--a generator and sells power to the Rural Electrification remarkable achievement in a country of 140.5 million Board, plans to own and operate a substantial people. Yet most of the PBSs are not financially viable. portion of the power sector's generation assets. As Massive operating and capital subsidies are required part of the unbundling process, regional distribution to keep the program functioning and expanding. companiesarebeingestablishedinKhulna(WestZone Although the PBSs are allowed to set their own tariffs, Power Distribution Company), Rajshahi (Northwest reflecting local costs, most agricultural and household Distribution Company), Chittagong and Comilla consumption is highly subsidized. (SouthZoneDistributionCompany),andMymensingh Another key issue is that the rural power sector and Sylhet (Central Distribution Company). relies heavily on the Bangladesh Power Development Board for its power supply and the Power Grid Toward a Pro-Poor Approach Company of Bangladesh for load scheduling. As TheRuralElectrificationBoard(REB),asemiautonomous a result, rural load shedding exceeds urban load governmentagency,isresponsibleforelectrifyingrural shedding, even though the REB network accounts for Bangladesh. Established in 1977, REB operations are only 40 percent of national electricity consumption. based on the U.S. model of consumer-owned rural As this survey's results highlight, providing electriccooperatives.Today,ruralBangladeshisdivided village access to electricity does not guarantee access Figure 6.2 Figure 6.2 Bangladesh Power Sector Organization Bangladesh Power Sector: Current Structure MPEMR ERC (Ministry of Power, Energy and Mineral Resources) Power Cell EA & CEI Generation BPDB IPP Small IPP PGCB (An enterprise of BPDB) Transmission Distribution BPDB (South/Central DESA DESCO REB/POSs WZPDC REB/POSs /North Zone) Source: Nexant (2006). 60 5580-CH06.pdf 60 5580-CH06.pdf 60 3/11/09 10:55:47 AM 3/11/09 10:55:47 AM 6 Institutional Framework for Rural Energy Supply at the household level. The government's Poverty 10 to 15 percent makes its way beyond the peri-urban Reduction Strategy Plan has identified various centers to more remote rural areas. measures to improve the pro-poor orientation of Promoting rural access to natural gas could result rural electrification. These include: (1) setting up insignificantwelfaregainsformanyruralhouseholds. a microcredit mechanism to help poor households For example, by switching from biomass to LPG to surmount the hurdle of first connection, (2) breaking meet a portion of their cooking needs, rural family up the REB into regional boards to create a more members, especially women, could realize large locally driven approach to the economic development time savings (through fewer hours spent collecting aspects of rural electrification, and (3) increasing biofuels) and better health (through reduced exposure PBS financial viability by their taking over peri- to indoor pollution). urban enclaves still served by the Bangladesh Power Rural populations' access to pipeline gas is also Development Board. inadequate. Indeed, rural residents represent only 6 percent of national coverage. By contrast, in major Oil and Gas Sector urbanareasservedbyregionaldistributioncompanies, Natural gas is by far the most important commercial non-bulksaletocommercialandresidentialcustomers fuel in Bangladesh, accounting for more than isextensive.Oneexceptiontothepoorruraldistribution 70 percent of primary commercial energy production. network is a US$300 million project financed by Most oil is imported in the form of refined products. the Asian Development Bank that plans to expand Since 1998, urban marketing and use of bottled gas transmission infrastructure into the western region has grown rapidly. In and around the capital city of surrounding Rajshahi by 2010. Dhaka, bottled compressed natural gas (CNG) is now a cheap substitute for gasoline in the transport sector. Sector Organization Total in-country production of liquefied petroleum The oil and gas sector is organized into four gas (LPG) is 23,000 tons per year, with another 35,000 segments: exploration, production, transmission, and to 45,000 tons imported and locally bottled. Yet only distribution (including marketing) (see Figure 6.3). Figure 6.3 Figure 6.3 Oil and Gas Sector Organization Source: McKenzie (2006). 61 5580-CH06.pdf 61 5580-CH06.pdf 61 3/11/09 10:55:47 AM 3/11/09 10:55:47 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Key sector institutions and their functions are production facilities and network expansion.43 as follows: Although LPG is widely available in most peri-urban areas, it is estimated that only 10 to 15 percent finds · Petrobangla. Successor to the Bangladesh Mineral its way to rural areas for consumption. In more Oil and Gas Corporation, Petrobangla is a remote rural areas, where LPG prices are fixed, it is public-sector holding corporation responsible not unusual for local dealers to charge significantly for oversight and coordination of 11 subsidiary higher prices. gas companies. The government has agreed to a roadmap for · BAPEX. The Bangladesh Petroleum Exploration sector reform, which emphasizes clear separation of Company is a state-owned entity that manages policy making, regulatory, and service provisions exploration and production. through the creation of well-structured, adequately · BGFCL. The Bangladesh Gas Fields Company funded entities. Key decisions include transferring Limited is a state-owned entity that manages Petrobangla's quasi-regulatory functions to the production. newly formed Energy Regulatory Commission and · SGFL. The Sylhet Gas Fields Limited is a state- resolving retail-market arrears to alleviate fiscal drain owned entity that manages production. on Petrobangla. · IOCs. The International Oil and Gas Companies are private-sector exploration and production Renewable Energy Sector entities; they include Unocal Corporation, Cairn Non-biomass renewable-energy systems also hold Energy, Niko Resources, and Tullow Oil. promiseforruralBangladesh.Solarphotovoltaics(PV) · GTCL. The Gas Transmission Company Limited are used widely throughout the country, with more manages midstream. than 80,000 reported household- and enterprise-level · TGTDCL. The Titas Gas Transmission and installations (3.5 MW total capacity). Solar PV could Distribution Company Limited is a state-owned, be used within a broader rural electrification program regional distribution entity. if affordable products that meet consumer needs were · BGSL. The Bakhrabad Gas Systems Limited is a supplied and supported. state-owned, regional distribution entity. Otherpotentialrenewable-energysourcesinclude · JGTDCL. The Jalalabad Gas Transmission and wind, hydropower, geothermal, tidal, and wave. Distribution Company Limited is a state-owned, Although wind energy has made some inroads, its regional distribution entity. potential is likely limited to coastal areas and islands · PGCL. The Pashchimanchal Gas Company with strong wind regimes; currently, the country Limited is a state-owned, regional distribution has 2 MW of installed wind turbines in operation. entity. The potential for micro- and mini-hydropower is · RPGCL. The Rupantarita Prakritik Gas Company also limited (with the exception of Chittagong and Limited is a CNG marketing entity. the Chittagong Hill tract); the country has one large facility at Kaptai, installed some 40 years ago, which Downstream Deficiencies and Reform produces 1,000 GWh per year. The downstream gas sector includes pipeline Until recently, renewable energy was developed gas, bottled LPG, and CNG. Downstream sector on an ad-hoc basis. Over the past 20 years, various performance is hampered by network losses of government agencies, research institutes, NGOs, about 20 percent, as well as pricing and commercial and private-sector companies have participated in performance deficiencies. Fiscal drain due to development, demonstration, and implementation. inadequate pricing prevents operating companies Current institutional arrangements reflect the sector's from reinvesting in required exploration and nascent state. 43 Petrobangla buys IOC gas at US$2.6­2.9 per MCF but charges domestic customers only US$1.40 per MCF. 62 5580-CH06.pdf 62 5580-CH06.pdf 62 3/11/09 10:55:48 AM 3/11/09 10:55:48 AM 6 Institutional Framework for Rural Energy Supply Key sector institutions and their focal areas are milestone in visualizing the country's future energy as follows: needs. But it failed to comprehensively address rural energy as a unique and important energy-sector issue. · Grameen Shakti. A subsidiary of the Grameen Although the policy covered such critical areas as Bank, Grameen Shakti focuses on solar home rural electrification, rural penetration of commercial systems (SHS), having installed 5,000 systems fuels, and reforestation, it overlooked improved use over the past two years; household biogas of biomass, biogas production, and other programs systems; and improved stoves. equally vital to energizing rural Bangladesh. · REB. The Rural Electrification Board centers on The2006policyrevisiontakesamorecomprehensive household- and village-scale, off-grid PV systems; approach, focusing on integration across subsectors, it installed nearly 2,000 SHS through the PBSs. demand-side issues, more effective institutional · IDCOL. The Infrastructure Development arrangements, and policy monitoring. Among the Company Limited focuses on household- and critical rural energy issues identified are the financial village-scale, off-grid PV systems. viability of loss-making PBSs, indoor air pollution, and · IFRD. The Institute of Fuel Research and the effects of commercialized biomass supplies on the Development, within the Bangladesh Council poor. for Scientific and Industrial Research, centers on Tackling rural energy issues in a more integrated small- and medium-sized biogas digesters. way calls for establishing a Rural Energy Steering · Local Government Engineering Department. The Committee within the Ministry to improve linkages Department handles small- and medium-sized between energy and rural development institutions. biogas digesters. The Sustainable Energy Development Authority · BRAC. The Bangladesh Rural Advancement (SEDA), the proposed secretariat for the Committee, Committee, the largest national NGO, centers on would coordinate with ministries and government SHS, having supplied nearly 1,200 systems, and agencies to develop and oversee policy and program biogas installations. implementation. It is envisioned that the new · Proshika. A leading NGO, Proshika focuses on Committee and SEDA would work together to biogas installations. recommend ways in which to improve coordination between all levels (national, district, subdistrict, and To date, Grameen Shakti and private-sector local) and form local organizations that could plan companies have led much of the work in renewable and implement sustainable energy projects. energy. In addition, the World Bank has supported Summing up, the many diverse institutions the scaling up of solar PV systems. Other donors have currentlyaddressingruralenergyissuesinBangladesh included the United Nations Development Program are committed to providing rural residents better and the U.S. Agency for International Development. energy services. But the multitude of institutional players creates a significant role for improving inter- Building an Integrated Approach institutional coordination. That this role has often gone unfulfilled may explain why rural energy The 1996 National Energy Policy, developed by the remains invisible to many policy makers. Suggested Ministry of Power, Energy, and Mineral Resources steps toward filling this institutional gap are the focus as part of power-sector reform, represented a policy of the final chapter. 63 5580-CH06.pdf 63 5580-CH06.pdf 63 3/11/09 10:55:48 AM 3/11/09 10:55:48 AM 5580-CH06.pdf 64 5580-CH06.pdf 64 3/11/09 10:55:49 AM 3/11/09 10:55:49 AM 7 Policy Recommendations This study--the first in 30 years to examine energy With the exception of rural electrification, access in rural Bangladesh--reveals several stark Bangladesh's energy policies and institutions have realities. Rural residents continue to rely heavily focusedmainlyonurbanmarkets.In2006,thecountry on traditional biomass to supply most of their still lacked a comprehensive, unified rural energy energy. But biomass used in traditional ways has policy with appropriate institutional support. As a energy efficiencies of only 15 to 25 percent; thus, result, the rural energy issue remains invisible to extensive time is required for fuel collection and many policy makers. meal preparation. Furthermore, biomass cooking The major recommendations for the rural energy fuels result in indoor air pollution (IAP), which is sector in this study focus on developing the capacity linked to respiratory disease and related health risks. for implementing high-quality programs rather than Rural residents face an increasing fuelwood shortage, specific policies. Energy policies cannot be static, meaning that many are relying on residues, including given the ever-changing energy markets around the dung, leaves, and grass, to meet their basic cooking world. The main recommendations are as follows: needs. Rural businesses also depend on biomass as one of their main energy sources. Thus, the problems · There is a need to further develop the institutional associated with biomass energy are widespread. capacity to deal with biomass energy problems, Significant bottlenecks prevent most forms of including the strengthening of the Infrastructure modern energy--a main alternative to biomass Development Company Limited (IDCOL) and fuels--from reaching many rural people. Although perhapsotheragenciestodiversifyintosuchareas well managed, rural electrification has reached less as biogas, improved stoves, and other household than one-third of the population. Households and energy technologies. In addition, more attention businesses fortunate enough to have electricity must needs to focus on improving local biomass supply contend with frequent power outages, which imply in and around farms. added expenditure on backup diesel or kerosene fuel · The recent progress in off-grid electrification and and hindrances to rural productivity and quality renewable energy through the IDCOL program of life. That liquefied petroleum gas (LPG) is used is encouraging and should be continued and little for cooking reflects its high cost and lack of diversified. availability. · The Rural Electrification Board needs to maintain But the rural energy picture is not all bleak. As its distance from political pressures. It also needs this study reveals, significant benefits, including to deal with the problem of brownouts and more lucrative rural businesses and better educated blackouts prevalent in the program and assess rural households, can result from promoting access to ways to financially strengthen the country's modernandsustainableenergy.Thus,thedevelopment weaker rural electric cooperatives or PBSs. of energy infrastructure is sound development policy Given the spread between village and household that policy makers should actively promote. electrification and the low number of agricultural pump connections, an assessment of whether 65 5580-CH07.pdf 65 5580-CH07.pdf 65 3/11/09 10:56:04 AM 3/11/09 10:56:04 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities financially viable electricity intensification members, but especially women, spend significant policies might be pursued is needed. amountsoftime(nearly200hoursannually)collecting · Given the extensive commercialization and all forms of biomass. Such arduous chores divert increasing scarcity of quality biomass energy time from competing tasks and important human- in rural areas, there appears to be a market for development activities. Moreover, poorly designed liquefied petroleum gas (LPG), with the right biomass-burning stoves without chimneys have low marketing strategies. Currently, the reach of LPG energy efficiency. Used indoors, such stoves emit high into rural areas is low compared to other South levels of smoke and other products of incomplete Asian countries. combustion that collect at high levels indoors. The · Finally, given the government's weak capacity to result is significant health problems for all family assess rural energy policies in an objective way, it members, especially women and children, who spend isrecommendedthatapolicygroupbecreatedvia many hours indoors within the vicinity of the stove. an independent agency, part of another agency, Mitigating the ill effects of biomass burning or a consortium of existing research institutions. requires both demand- and supply-side solutions. It is also recommended that a more complete Because all biomass shortages vary by region, the set of energy questions be incorporated into the solutions must be local; thus, one should caution national income and expenditure surveys, as against any blanket interventions covering the entire purchase and collection of biomass energy are country. Surveys during various times of the year nearly absent entirely from existing ones. might be undertaken that ask household members a range of questions regarding the availability of The sections that follow bring the broad array biomass (both collected and purchased), alternative of fuels that sustain rural people's lives into focus. fuels, cooking devices, food preparation methods, Tackling the constellation of issues surrounding and kitchen practices. Various proposed solutions their effective use is critical to the sustainability of could then be offered householders, local leaders, Bangladesh's economic future. But the task will not be and non-governmental organizations (NGOs). The easy. It will require complementary efforts on many participationofwomenwouldbeespeciallyimportant diverse fronts. as they are the ones who deal daily with biomass collection, processing, and cooking. With these Problems and Potential of groups' active participation and cooperation, the most Biomass Energy suitable solutions should be found. Biomass demand can be reduced by using Biomass--a vital energy source in rural Bangladesh-- improved cookstoves and kitchen practices. Where is often overlooked by energy policy makers. As this possible, fuel substitution including the use of study shows, biomass is the country's primary rural kerosene or LPG should be encouraged. Biomass household fuel. More than 95 percent use it to cook all supply can be increased by planting trees around or part of their meals. Moreover, most household- and homes, maintaining and improving local natural village-based enterprises use biomass for heating. resources, and increasing agricultural productivity. As the survey findings illustrate, beneath this Although the impetus for improvement should come biomass dependence, an energy crisis is simmering. from rural residents, the government needs to play In regions experiencing a continuing decline in the a supportive role through such activities as stove availability of quality biomass, many people seem testing and quality control, providing seedlings to be turning from fuelwood to dung, straw, tree and other inputs, and offering extension advice and leaves, and grass. As the previous chapter discussions demonstrations of relevant technologies. underscore, quality biomass in the form of wood is Because biomass is an important fuel in rural fast becoming monetized, a reflection of its scarcity. industry and service sectors, improving intermediate In areas where wood is not purchased, all family and end-use efficiency in these sectors is essential. 66 5580-CH07.pdf 66 5580-CH07.pdf 66 3/11/09 10:56:04 AM 3/11/09 10:56:04 AM 7 Policy Recommendations The increased commercialization of biomass fuels is a improved kitchen practices, and better ventilation can double-sided issue: In areas where commercialization mitigate the negative health effects, such measures has accelerated, many rural residents face shortages cannot reduce biomass demand. More sustainable of locally available, quality fuelwood; conversely, a solutions call for efforts that increase stove efficiency certain proportion earns income through biomass and vent smoke from people's living areas.45 There growing, production, transport, and trade, which are some recent estimates in Bangladesh that IAP should assist in alleviating poverty.44 might account for as much as 8 percent of the burden In short, promoting a sustainable supply-demand of disease. balance of biomass energy, supported by appropriate In Bangladesh, most intervention programs to policies, should be encouraged. To this end, rural popularize improved stoves were proven ineffective residents can be encouraged to move toward more or were small in scope. The largest program, efficient use of biomass energy, in combination with terminated in 2001, was that of the government of tree planting on farmsteads and improved local forest Bangladesh. The program was supported by various management. organizations, including the Institute of Fuel Research and Development, the Bangladesh Council for Increasing Biomass Efficiency and Use ScientificandIndustrialResearch,Ansar-VDP,andthe Bangladesh Rural Development Board. By December This section reviews past and ongoing initiatives that 2002, an estimated 300,000 Bangladeshi households have aimed to improve biomass energy efficiency had received access to improved stoves. Based on and use in rural areas, highlighting lessons and these penetration rates, one could expect that about potential solutions that can mitigate risks to human half a million households would have gained access and environmental health. by 2007. This figure is less than 0.5 percent of the Improved Stoves: Potential for population; thus, there is much room for improvement Revitalization (Hossain 2003). Some 20 million Bangladeshi households own up to More recently, various agencies have continued 40millionbiomassstoves,mostofwhichareinefficient to promote improved stoves, albeit on a smaller scale. and produce excessive smoke and other products Current NGO initiatives are not widespread for lack of incomplete combustion. As this study shows, of trained personnel and funds. Grameen Shakti has millions of people in close proximity to such stoves been working throughout the country on initiatives suffer adverse health effects. At the same time, the fundedbyinternationaldonors,wherebybeneficiaries products of incomplete combustion contribute to the are provided microfinance on demand; to date, its accumulation of greenhouse gases. For these reasons, accomplishments have been modest. The Village all relevant ministries--health, environment, energy, Education Resource Center, another NGO striving to and forestry--should be much concerned about the alleviatepovertyinBangladesh,isamongthefewlarge human and environmental damage caused by using ones actively working on improved stove initiatives. biomass energy in inefficient, polluting ways. The Center operates from its head office in Savar and Worldwide, 1.5 million people die prematurely 14 branch offices located across various districts. Its each year because of indoor air pollution (IAP), caused main focal areas are water, sanitation, hygiene, and byusinginefficientbiomassstoves(WHO2006).Many improved stoves. It has pioneered a participatory millions of days are lost through sickness caused by assessment method to mobilize communities and inhaling excessive smoke. Although biomass drying, generate village demand for improved stoves, house 44This study examined the purchase of biomass energy, but did not investigate the market structure of fuelwood or other biofuels; thus, more research in this area is needed. 45It should be cautioned that even the most effective improved stoves being sold or distributed in developing countries do not lower IAP levels to those recommended by international environmental agencies; but it is a step in the right direction. In developed countries, there are stoves for heating that eliminate virtually all indoor pollution and vent little pollution outside the home. 67 5580-CH07.pdf 67 5580-CH07.pdf 67 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities modifications, and behavioral change to transform stove design and testing; demonstration; research villages into smoke-free ones. The Center's demand- and development; quality control; promotion in all oriented approaches and stoves' designs cater to media forms; making simple tools available to stove consumer requirements. Even so, accomplishments to makers; providing advice on material sources, use, date have been modest. Finally, the Local Government and suitability (especially if clay); training of stove Engineering Department, the technical arm of the makers, repair-service operators, and extension Local Government Ministry for Rural Infrastructure workers; monitoring and evaluation; and loans. Development with nationwide reach, is implementing End-user feedback on the pros and cons of various work on improved stoves. stove designs should be ongoing, and suggestions Various reasons account for the ineffectiveness of should be integrated into the further refinement of many earlier government initiatives. In some cases, stove design. the stoves were inappropriate to the needs of the cook. Stove initiatives should be part of a larger Inothers,programsweredrivenbyinstallationtargets program to improve village life, especially the health with little follow-up activities. Many programs were and welfare of community members. Part of this subsidized, with benefits going to the organizers or program should provide training on biomass use (e.g., installers rather than the recipients. In still other cases, drying and using in appropriate sizes and discussion the improved stoves were no better than the originals, on food preparation and energy-saving methods). but were less versatile. Many such initiatives could be initiated within the Where improved stove programs have succeeded, community; through these efforts, the entire village cooks have been an integral part of the process. could participate and take pride in its achievements. Advice can be offered about the pros and cons of The overall program should promote market- interventions. Saving time and money are usually based approaches that rely on local producers and top priorities; at the same time, comparing the health distributors who profit only from satisfied customers. effects of various stove designs and kitchen layouts Direct stove subsidies should be avoided, as they should be stressed. Improved stoves need not be create market distortions and ultimately hinder stove complicated; they can be made from local materials. commercialization. Instead, subsidies should be Local women's groups can be organized and trained directed toward activities that support the program to make and install improved stoves on a commercial (e.g., greater government backstopping, active NGOs, basisafterdecidingonhouseholdcookingneeds.Such small-scale enterprises, and using women's groups groupscanfabricatechimneypieces(ifnotpurchased) to promote improved stoves for rural households from mud, using straw as a reinforcing agent and and microenterprises). Finally, a wide range of stove burnt clay inserts of standard sizes, thereby reducing designs (some of which would be suitable for the the cost US$5­10 per stove. service sector [e.g., canteens, restaurants, and hot-food Evaluation research suggests that better- shops]) is needed. coordinated efforts could revitalize the potential of Biogas Digesters for Cooking and Farm improved stoves, which would reduce pressure on Productivity traditionalbiomasssuppliesandsignificantlyimprove the health of biomass-reliant rural populations. It is Currently, some 25,000 biogas digesters are operational suggested that improved stove programs take a more in Bangladesh. Each digester requires dung from about pragmatic approach with regard to stove benefits six cows (i.e., assuming that each produces 10 kg of wet and performance, including demonstrations and dung per day, equivalent to 1.15 kg of air-dry dung). If promotion to increase awareness of the dangers of 25percentofcowdungcouldbeutilized,approximately IAP and need to conserve biomass fuels. 1millionfamily-sizeddigesters,eachsupplyingcooking Undertaking a sustained nationwide program fuel for five to six people, could be built. Unfortunately, requires sufficient administrative and technical many farmers lack access to the required amount of infrastructure.Specificactivitiesincludethefollowing: dung, which hinders large-scale introduction. 68 5580-CH07.pdf 68 5580-CH07.pdf 68 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM 7 Policy Recommendations Passing dung through digesters has several baking. Non agro-based industries included brick advantages. First, the gas produced is easy to light and and tile manufacturing, lime burning, road tarring, control; its efficiency for cooking is 50 to 60 percent, and soap making. Wood was the main input for depending on the stove. Used for lighting, the gas charcoal production. Most industries were based in emits a brighter light than kerosene. Most pathogens rural areas, relying on inexpensive, readily available in the dung are destroyed during the production biomass supply, characterized by low fuel efficiency. process, and the resulting slurry makes an excellent It is recommended that the government provide a fertilizer. If dung is applied directly to paddy fields, range of technical assistance, including energy audits, methane, a potent greenhouse gas with 20 times the to improve fuel efficiency. In addition to improved warming effect of carbon dioxide (CO2), is produced stove initiatives, interventions should be pursued for andventedintotheatmosphere.Digesterscapturethis charcoal production and other activities where solid methane and use it for cooking; furthermore, slurry biomass is used or produced. is a better fertilizer than dung. Technical assistance should focus on a wide array The Bangladesh Council for Scientific and of areas: improving institutional- and service-sector Industrial Research (BCSIR) runs the government's stoves; improving brick, pottery, and lime kilns; active but modest biogas digester program, which conducting energy audits at sugar factories and focuses on relatively large units for which farmers sawmills to assess energy-saving approaches; listing receive advice and a subsidy of Tk 7,500 (US$120) agro-processing factories that use biomass energy per digester. Because these digesters are relatively (e.g., tea drying, rice parboiling, and fish smoking); expensive to build, they must be properly maintained and examining ways to improve energy efficiency to perform effectively. A fixed-dome, 5 m3 unit costs for all biomass users. To succeed, these initiatives US$240­290. Cheaper black plastic (PVC) models require quick payback periods. Charcoal producers that fit into trenches are variously sized according could be trained to better manage wood resources to the number of cattle. The cost of such digesters and improve production processes. If forests are the (with appliances) is about US$100. Farmers with main wood source, the forest service should work only one or two cows could use this type of digester with charcoalers to plan a cutting cycle. Cooperating to meet a portion of their cooking needs and, at the with producers would enhance forest sustainability same time, produce an excellent fertilizer. Financing and diminish illegal cutting. could be arranged through the active program of a With regard to briquetting of difficult-to- Bangladeshi microenterprise. Farmers or villagers burn residue (e.g., husks and sawdust), many past could cooperate to build community digesters, which initiatives have failed because of inappropriate, could provide enough fuel to run electric generators sophisticated equipment or lack of sound marketing for part of each day. and management. The cost of briquetting is generally Over the next several years, the Infrastructure expensive, and the product may not be competitive Development Company Limited (IDCOL) plans to with unprocessed biomass. Hand machines and facilitate the construction of thousands of biogas locally made, inexpensive binders, as well as plants. It might be beneficial for two complementary potential markets for the finished product, should agencies, such as the BCSIR and IDCOL, to foster be investigated. Densification may offer a solution for biogas development. straw, which is bulky and quick-burning, but price is a constraint. Various types of densifying equipment, Higher Bioenergy Efficiency for Industry including hand presses, as well as binders, should be In 2000, the formal and informal industrial sector-- compared. If unemployment and underemployment mostly agro-processing industries--accounted are problems, simple hand devices may prove more for more than 20 percent of total biomass energy appropriate than high-tech machines. Recently, this demand. Energy uses included parboiling of rice, area has become more widespread, with several sugar production, tea drying, tobacco curing, and hundred factories operating across the country. 69 5580-CH07.pdf 69 5580-CH07.pdf 69 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Increasing Biomass Supply resource base. Adopting an effective strategy is A complementary solution to Bangladesh's biomass urgently needed, especially in rural areas, to ensure energy problem is to increase or maintain supply an ongoing increasing supply of renewable biomass through local forest management or growing of energy. Implementation of this strategy should be trees, bamboo, and shrubs on and around farms guided by powerful institutional voices within the and along roads. Trees grown outside the forest are government. Thus, in addition to the Ministry of a principal source of rural fuelwood. Such trees are Power, Energy, and Mineral Resources, the ministries usually planted close to the demand center and are of agriculture, forestry, health, rural development, intensively managed, especially if privately owned. women's affairs, and others are also essential. They provide added income by yielding multiple Especially important is the proposed Renewable products (e.g., fruits, medicines and herbs, and Energy Development Agency (REDA), backed by the nectar and leaves used as feedstock in honey and silk Ministry's renewable energy division. production). The upsurge in tree-planting activities over the past 20 years has resulted from the support Proposed Strategy of NGOs, government, and international donors, Given Bangladesh's significant dependence on as well as the active involvement of rural people. biomass energy, we reiterate some of the previously Through participatory silviculture, it will be possible cited suggestions to conclude this section. A strategy to increase tree planting and management both in and to improve biomass energy prospects in rural areas outside the forest. could take various approaches, four of which are Itcomesasnosurprisethatagriculturalconversion highlighted here. First, the intermediate and end- threatens a portion of rural Bangladesh's forests and use efficiency of biomass energy production should grasslands. Rural population growth is 2.5 million be increased, especially at the household level. per year, more than 62 percent of the country's The aim would be not only to reduce unit energy annual growth rate of 4 million. Unless agricultural consumption, but, more importantly, to decrease productivity keeps pace with population growth, products of incomplete combustion, which damage forested lands will be cleared for food production. human health and contribute to the accumulation of Short-rotation, nitrogen-fixing trees could be planted greenhouse gases. Second, biomass production (of in and around farmland to maintain, if not increase, both annuals and perennials) should be increased on productivity and provide farm animals browsing all land formations so that the country can keep pace material. Such trees yield stick wood; and increased with demand for biomass and its products. Ideally, productivity and browsing may result in more increased productivity should outpace population agricultural residue and dung. Thus, it is essential growth so that more biomass can be used for all that the relevant ministries, particularly agriculture purposes, especially energy. Third, use of modern and forestry, work jointly to facilitate increased energy and more convenient forms of biomass should agricultural productivity, without which the biomass be promoted, despite ongoing use of traditional forms resource base will surely diminish. of biomass, which will continue for many years to Biomass is rural Bangladesh's most important come. (Modern forms of energy include solid, liquid, cooking, heating, and industrial fuel and will remain and gas products used as fossil-fuel substitutes; so well into the future. For perennials, this implies biomass and its products could be used to generate management that removes not more than annual electricity, where appropriate). Finally, buildings and growth over the long term; for annuals, it means, at habitats should be designed to conserve energy. Trees a minimum, that soil fertility be maintained so that planted around buildings and along streets would land does not deteriorate to a state of marginal or no reduce the ambient temperature, which would, in productivity. turn, reduce the need for cooling in offices and homes TheexpecteddoublingofBangladesh'spopulation while periodically providing biomass. Trees would over the next 30 years will further strain the natural also mitigate erosion and improve the environment. 70 5580-CH07.pdf 70 5580-CH07.pdf 70 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM 7 Policy Recommendations Within the country's current rural energy policy otherwise would remain unconnected for years will and institutional framework, these recommendations be served. may be difficult to implement. Thus, the last section This section explores ways in which to improve of this chapter turns to the complexity of institutional both grid and off-grid electrification programs. issues that must be tackled in order for policy implementation to succeed. However, we now turn Grid and Off-grid Programs to a successful program to promote electricity in Both grid and off-grid electrification programs Bangladesh. are important to the socioeconomic development of rural Bangladesh. The evidence is convincing Rural Electrification and Rural that electrification translates into substantial gains Development in household welfare and a higher quality of life. Indeed, the benefits of lighting alone are highly In 1976, Bangladesh was a newly independent country valued by rural households that adopt electricity. grappling with the challenge of creating national As this study shows, household- and village-based policies and programs amid an underdeveloped microenterprises that connect to the grid are more infrastructure, rapidly growing population, and profitable than comparable businesses served by frequent natural disasters. Few would have predicted private or locally purchased generators. that 25 years later, this poor South Asian country would have succeeded in providing electricity to Cooperative Grid Electrification: Role of the nearly 30 percent of its rural households. Rural Electrification Board Bangladesh's story of rural grid electrification is One of Bangladesh's most successful development important for several reasons. First, despite relatively programsoverthepast20yearshasbeentheextension low coverage rates compared to other developing of grid electricity in rural areas. As discussed in countries, an additional 600,000 rural customers previous chapters, the Rural Electrification Board receive electricity services each year. This annual (REB) promotes the development of electricity increase exceeds the total rural population of many distribution companies based on the rural cooperative countries. Second, having a well-managed, well-run model adopted in the United States. In 1977, it was program is somewhat unique in South Asia. The argued that enlarging the public-sector electricity rural cooperatives have low distribution system company's focus to include low-density, poorer rural losses of only 15 to 17 percent; their 95 percent areas was unfeasible, given the company's already revenue-collection rate is high by developing country overextended efforts to serve urban areas. Thus, standards. Third, though one of the world's poorest the REB was created in 1978 as a semi-autonomous countries, Bangladesh demonstrates how rural public agency to promote electricity in rural areas and electrification programs can succeed under adverse implement a financially viable program. economic conditions. As an agency of the Ministry of Power, Energy, A complementary rural electrification story is and Mineral Resources, the REB is responsible for Bangladesh's recent effort to strengthen capacity in planning and implementing all investments in rural decentralized or off-grid service. Although the grid electrification infrastructure, overseeing performance is expanding service by large numbers every year, of the rural electric cooperatives or PBSs, and systemsexpandoutwardfromgenerationstations,and regulating prices. Over its 25-year history, the REB transmission lines serve the most highly populated has constructed the infrastructure for Bangladesh's areas first. As a result, some 50 percent of rural entire national rural electrification system. In this villages--70 percent of rural people--remain without capacity, a major responsibility has been to manage a connection. But a host of renewable technologies loans and grants provided by international donors now available for decentralized generation and with which to finance an infrastructure development provision offer promise that many households that program. Another essential function has been to 71 5580-CH07.pdf 71 5580-CH07.pdf 71 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities finance short-term capital needs of newly established preferential treatment to companies serving poorer PBSs. Finally, the board has provided the PBSs vital regions, such as permitting higher tariffs or providing technical assistance and monitoring of technical and electricity at lower bulk prices. financial performance, thus helping them evolve into The conclusion is that Bangladesh's rural modern distribution utilities. electrification program has been both active and Despite the REB's long success in promoting effective in its positive development outcomes. The rural electrification, recent problems call for urgent national survey on rural electrification confirms attention. As this study's national survey reveals, electricity's significant effect on rural development. rural consumers perceive that electricity supply In 2006, the PBSs were generally well managed; yet has become unreliable. Some 80 percent of rural the program has a nationwide generation problem. households report daily outages, while 60 percent Rural customers have reported significant outages, report significant power fluctuations. Survey results which, if they persist, may lead to dissatisfaction with also bring into question the REB's selected coverage PBS performance. In addition, many cooperatives approach, as 20 percent of households that receive in more remote areas are not yet financially viable. electricity indirectly from the PBSs--most commonly Policy solutions will require load promotion to via a neighbor--are billed at a higher rate than that increase revenue or price restructuring and subsidies. charged by the PBSs. While not illegal in the strictest A recent initiative is studying REB management sense (i.e., electricity is metered and paid for by the practices and the board's potential as a freestanding collecting household), this practice suggests that some corporation. Such studies stress the REB's key role in type of barrier prohibits households from receiving supervising and promoting rural electrification for electricity directly from the PBSs. It may be that Bangladesh and its need to remain free from political households are beyond the required distance from manipulation. the grid or informally connected ones want to avoid connection charges. Off-grid Electrification: Role of the The national survey also reveals a significant Infrastructure Development Company potential to replace diesel-powered irrigation motors Limited with electric ones. Investment in infrastructure to Only 3 percent of Bangladesh's nearly 17 million deliverelectricitytoruralareascouldbecomplemented rural households are gaining access to the national by development of daytime loads. Since most current grid each year, creating an enormous potential for rural demand is during evening hours, revenue from off-grid systems. Opportunity is greatest in locales the additional agricultural load based on availability where grid extension is uneconomical. Many rural of groundwater potential could significantly improve residents can afford electricity if the costs are spread PBS finances. out over many years. Given that many components Recent financial analysis of the PBSs has brought of renewable energy systems last 20 years or more, into question the REB's pricing and subsidy policies. it makes sense to allow households to spread out In 2006, the PBS service territories differed markedly; purchases of renewable energy equipment over a some were characterized by lucrative households, period of years. Such an arrangement could be made while many others served poorer regions. The REB's in conjunction with rural development programs standard subsidy and pricing policies mean that the aimed at improving local incomes. financial viability of PBSs in poorer service territories In2002,theInfrastructureDevelopmentCompany tends to be poor. Amid the constant struggle to meet Limited (IDCOL) was made the country's focal financial targets, these cooperatives lack the financial agency for coordinating the off-grid program, the resources to serve their customers properly. Thus, first phase of which has focused on promoting solar careful review of the policies of uniform subsidies photovoltaic (PV) systems. To date, most work has and similar tariffs for all PBSs is recommended. One been done by NGOs specialized in microfinance potentialsolutionmightincludemechanismsthatgive and microenterprise development. The program has 72 5580-CH07.pdf 72 5580-CH07.pdf 72 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM 7 Policy Recommendations succeeded beyond initial expectations. Working with companies were created to import, refine, and market 16 partner organizations, including Grameen Shakti kerosene and other petroleum products. For the past and other NGOs involved in microcredit and solar 30 years, the Bangladesh Petroleum Corporation home systems (SHS), the program has installed more and its subsidiaries have controlled most aspects of than 80,000 systems over a three-year period. petroleum supply, including its pricing system. The The program has operated on the premise that following sections suggest pricing policies for specific partner organizations, who have gained the trust of fuels, where appropriate, to promote more equitable rural residents, are the most efficient SHS delivery rural use. agents. Their collection history has proven strong enough to develop a credit line. IDCOL's supportive Kerosene role has focused on developing consumer awareness Bangladesh's kerosene market consists of government of SHS and their potential for rural lighting. suppliers, known as oil-marketed companies, and Additional responsibilities have included registering private retailers. Theoretically, kerosene is priced participating organizations eligible for assistance, uniformly across the country, but transport-cost establishing standards, and refinancing up to adjustmentsaremadeformarketdistancesgreaterthan 80 percent of partner organizations' customer loans. 40kilometersfromasupplydepot.Priceincreasesoccur This innovative program has demonstrated at each distribution stage: from dealer to subdealer to that, with careful planning and adequate support retailer.Themoreremotethedestination,thehigherthe in the initial stages, partner organizations can gain fuel price. But the price difference between rural and proficiency in equipment supplier dealings and after- urban areas is not great, and the system for containing sale customer support. In addition, the significant kerosene within a price range works well throughout portionofincomethatpoorruralhouseholdshavebeen the country. Because most rural households use the willing to allocate for basic lighting has underscored fuelforlighting,itisaccessibletomostatworld-market the high value they place on the service. rates; in addition, it is available in quantities sufficient Initially, the program aimed to remain for cooking. Thus, kerosene pricing policies require no technology neutral. Indeed, IDCOL was selected as substantial revision. the implementing agency because it did not promote a single technology. Although SHS has composed most of the program's lending demand to date, a Liquefied Petroleum Gas tremendous unmet need for electricity remains. In 1978, the Bangladesh Petroleum Corporation Recently, several other types of off-grid systems have began commercial production and marketing of been implemented, but financing has been limited. LPG. Over the next 20 years, use of LPG grew As a result, a significant need remains to diversify slowly. In the late 1990s, the government began product lines to microgrids, which could be connected to allow private companies to import and market directly to the national grid system once the PBSs the fuel. Today, Bangladesh has three public and reach these remote communities. six private LPG marketing companies. Private companies, which must pay import taxes and Toward Policies for Household charges associated with the fuel, compete with public companies, which are exempt from such Petroleum Fuels taxes and fees. Despite this price disadvantage, the Before independence in 1971, all of Bangladesh's private companies are viable and are expanding petroleum products were imported and marketed by from large urban centers into smaller towns. private-sector companies. Although these companies Notwithstanding such progress, LPG, which is competed, they were heavily taxed, which translated readily available in urban markets, has largely failed into higher fuel prices. After independence and to reach rural households. Because LPG is a clean fuel, nationalization of the petroleum sector, public-sector it could contribute to alleviating IAP. By having the 73 5580-CH07.pdf 73 5580-CH07.pdf 73 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities fuel on tap, rural people could reduce the amount of field for all companies marketing LPG. In this way, time spent collecting biomass. it is hoped they would be willing to innovate and The difficulties of promoting LPG in rural areas expand their markets into even more remote areas are well documented. Lower-income rural residents not currently served. with little cash on hand usually cannot afford the high upfront costs associated with LPG. They are unlikely to pay cash for the typical 12.5 kg cylinder or double- Natural Gas (Methane) burner stove (costing Tk 2,500­3,000). In addition, Piped natural gas from fields in the Bay of Bengal is LPG transport and storage costs (unlike kerosene's usedmainlybyhouseholdersinDhakaandotherlarge narrower range of prices) are prohibitive for many. towns. Indeed, the government has reserved natural However, in other South Asian countries, LPG is gas for household use. Under current pricing policy, penetrating rural areas. In India, for example, more households are charged a (subsidized) flat monthly than 20 percent of rural households use LPG to meet rate, irrespective of the amount of gas consumed. Such at least part of their cooking needs; thus, increased a policy leads to abuse and waste and discriminates availability appears to lead to greater use. Most such against households without connections in smaller households do not switch completely to LPG; rather, towns and rural areas. they use the fuel for quick-heating items, such as tea. It is recommended that the government reverse One should note that the poorest rural Bangladeshi the flat-rate tariff so that households are charged per households will not adopt LPG extensively; but unitofconsumption.Also,subsidies,whichinvariably more prosperous rural areas have greater scope for benefit wealthier households, should be removed. promotion. The savings from such changes could be redirected Increased use of LPG in rural areas could be to rural areas to encourage rural electrification, encouraged in several ways. First, commercial food increase biomass availability, and foster its more enterprises could be targeted initially; they currently efficient use. purchase biomass for food preparation, which At the same time, the brick-making industry requires significant cash outlays. Second, as in many has requested a pipeline connection to enable year- other countries, the upfront cost of LPG stoves and round brick and tile production, which would reduce cylinders, which are necessary for fuel use, could dependence on imported Indian coal and fuelwood be partially subsidized or paid for in installments; (officially banned from use in brick making).46 lower upfront costs would put stoves within reach Although the imported coal varies in quality, its of households who otherwise could not afford LPG. ash content is high, and thus its energy value is low Third, large-capacity cylinders, which target mainly (not much higher than that of dried wood). Because urban consumers, could be complemented by a the price of wood is lower than that of coal, wood variety of smaller cylinder sizes. Promoting smaller is used as a starter and booster fuel. But controlling cylinders in rural areas would reduce the cash outlay the temperature of the brick stacks is difficult, and for refills. All three recommendations would require the quality of baked bricks is not uniform. The kilns well-designed strategies in the safe promotion of LPG are usually shut down during the rainy season due to new markets. to fuel shortages caused by flooding in low-lying Because private companies must keep their areas. If the kilns were fired by natural gas instead of prices competitive with those of public companies, coal, they could operate year-round. In addition, they they have a disincentive to service rural areas. As a could be designed to provide uniform heat, ensuring result, they tend to target relatively better-off markets a superior product. not served by the main public companies. Thus, it Bangladesh's brick manufacturing association would be sound policy to promote a level playing has expressed a willingness to pay a non-subsidized 46Bricks and tiles require less energy than concrete and corrugated iron roofing; thus, brick making should be encouraged. 74 5580-CH07.pdf 74 5580-CH07.pdf 74 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM 7 Policy Recommendations price for natural gas, which would generate more succeeded in moving the country's electrification rate government revenue, save foreign exchange, and from 10 to 30 percent, continued support of its efforts reduce fuelwood use. It would also mean that a to fulfill its ambitious mandate is critical. Inevitably, percentageofbrickmakerscouldinstallmore-efficient such institutions face serious challenges as their kilns and produce more uniform bricks and tiles. It programs mature. It is imperative that international is recommended that the government examine the agencies stay involved to support them through the merits of supplying brick makers natural gas. critical transitions. As noted in previous chapters, the REB now faces an array of complex issues, including Enhancing Rural Energy Projects, differential subsidies for cooperatives that service poor-load areas, takeover of towns and small cities Policies, and Strategies within the boundaries of cooperatives, and the Rural energy is a complex issue, encompassing a transition to greater self-reliance. Another major broad and diverse spectrum of resources--from challenge identified by this study is access to reliable petroleum fuels and coal to biomass and renewable supplies of bulk electricity. energy--spanningmultiplesectors,includingforestry, Beyond the REB, institutions that support rural electricity, and health. To date, many of Bangladesh's energy have been working relatively independent of institutions involved in rural energy have not been one another. IDCOL's recent success in promoting coordinated. One major recommendation resulting solar PV systems suggests that renewable and other from this study is to develop the policy capacity to areas of rural energy require a level of support on tackle rural energy issues in all their complexity. This par with that of grid electrification. Unlike the REB policy capacity could be used to objectively analyze model, that of IDCOL is more decentralized, and there and promote rural energy solutions and would is no overlap between the two. Both models handle complement the existing institutional programs financing, provision of technical assistance, lobbying that are being used to implement rural energy in on regulatory issues, vendor approval, and support Bangladesh. of venture development. Neither the REB nor IDCOL Donors have often advocated projects with a limits its focus to technology dissemination. Both narrow technology focus. But experience teaches institutions are concerned with development of the that focusing on single technologies does nothing market and supporting environment for delivery of to further markets and private-sector companies to rural energy services. support rural energy development. Such an ad-hoc The recent move toward establishing REDA, arrangement is not conducive to capacity-building. while a welcome development, perhaps, does not go Whatever experience is gained under a specific far enough. The recommendation is to designate an project cannot be applied to subsequent ones for lack agencytohandletheentirerangeofruralenergyissues, of continuity. Moreover, single efforts generally seek sometimes acting as an executing agency and, at other exceptions to regulatory policies but are unable to times,playingacoordinationrole.GivenIDCOL'ssolid change them. As a result, it is not possible to see the track record, a first step might be to strengthen its off- long-term effects of such projects in the form of greater grid electrification unit, thereby enabling it to handle access to quality, rural energy services. Thus, a rural a broader range of rural energy issues. energy institution could provide government advice Because Bangladesh's entire range of rural energy on how to better focus use of donor funds. responsibilitiesisdividedamonganumberofagencies To date, successful rural energy programs in or institutions, development of an active policy unit Bangladesh have combined financing, institutional is needed to advise the government on the diversity support, local support, appropriate pricing, competent of rural energy problems. Such a policy unit, which implementing firms, and market development. They could be called the rural energy policy unit, would also have been characterized by a strong coordinating be responsible for formulating and promoting rural agency. A case in point is the REB; given that it has energy policies, ranging from biomass to electricity 75 5580-CH07.pdf 75 5580-CH07.pdf 75 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities and LPG. It would work hand in hand with various The Way Forward agencies with major responsibility for energy in rural areas, including the REB, IDCOL, and others. Rural energy's importance to the Bangladesh There is a significant need for a group in economy cannot be underestimated, given the Bangladesh dedicated to evaluating and promoting world's focus on globalization and market reform. ruralenergypolicies.ItisenvisionedthatBangladesh's This study underscores how the effects of rural rural energy policy unit would eventually be energy cut across multiple, diverse facets of rural independent or semi-independent of the government. life--from income and labor productivity to Over the short term, however, it might be practical to education and women's health. The problems rural house the unit within an existing agency or public people face in obtaining safe, clean, and reliable research group. Its role would be to promote sound energy supplies are not minor inconveniences. On policies for rural areas and innovative pilot projects; it the contrary, they represent a significant barrier to would not be involved in project development, which rural economic development and improved social would be left to other agencies. well being. A multifaceted approach to solving Inadditiontoprovidingpolicysupporttopromote Bangladesh's rural energy problems is not only rural energy, this new agency could compile and warranted; it is an essential building block to propel publish rural energy information. Analysis of data the country into the twenty-first century. from the national survey undertaken for this study-- The past two decades have witnessed many the first of its kind in 25 years--would provide future attempts to promote rural energy. Donor- and public- market opportunities by identifying priority locations sector supported projects that have introduced forprojectfinancing.Thiswouldbeagreatservicetoall and popularized improved biomass stoves have cooperating organizations interested in rural energy, yielded only limited success, despite the large including the many above-mentioned ministries and potential benefits of sustaining biomass supply agencies. It would be imperative to ensure that the and improving human health. Social afforestation technologies recommended were of sufficient quality programs initiated over the period have run their to meet rural energy needs. In addition, the program course. Although biogas programs have enjoyed scope would need to extend beyond renewable energy considerable success, they fall far short of realizing to include conventional off-grid power systems their considerable potential. Renewable energy in remote areas. Technology coverage should be efforts, especially the popularization of solar PV, broadened to include renewable energy (e.g., micro- have achieved a remarkable measure of success; hydropower, PV, and wind), improved stoves, tree even so, the technical and socioeconomic issues planting, and small diesel systems. associated with scaling up household and village Summing up, rural energy is a complex and, for electrification require capacity-building at national the most part, unprofitable business. But by using a and local levels. combination of loans and subsidies, both large- and The rural energy issue identified is this: small-scale businesses can become viable to better Bangladesh has a comprehensive need for better promote a wide range of rural energy services, from institutional coordination and attainment of a critical LPG and grid electricity to improved stoves and tree mass of technology and market development. With growing. Moving forward requires a combination effective institutional coordination, combined with of research, production and delivery, support, and market development, appropriate subsidy and monitoring and evaluation, all of which must be pricing policies, and government and donor support, done through a variety of businesses--including current and proposed programs can succeed beyond rural electric cooperatives, NGOs, private-sector expectations. The call for action is urgent, not only companies, and local municipalities--interested in for rural development, but for the country's equitable serving rural energy customers. economic growth. 76 5580-CH07.pdf 76 5580-CH07.pdf 76 3/11/09 10:56:05 AM 3/11/09 10:56:05 AM Annex 1 Selected Tables from the Household Survey Table A1.1 able A1.1 Survey Sample (number) Division Survey Type Dhaka Rajshahi Chittagong Khulna All Household 603 600 640 548 2,391 Home enterprise 50 27 31 29 137 Village enterprise 85 58 63 136 342 Institutions 38 32 36 50 156 Source: BIDS Survey (2004). Table A1.2 able A1.2 Household Distribution by Ownership of Agricultural Assets (percent households) Division Asset Type Dhaka Rajshahi Chittagong Khulna All Land (acres) < 0.50 57.9 59.5 56.1 48.7 55.7 0.50­2.49 28.0 27.5 27.0 32.9 28.7 2.50­4.99 8.6 7.0 8.4 12.4 9.0 5.00 5.5 6.0 8.4 6.0 6.5 Non-land Business enterprise/shop 9.8 5.3 6.3 8.6 7.4 Bicycle/motorcycle 15.1 30.7 13.1 28.1 21.5 Rickshaw/van 8.1 9.0 5.0 8.4 7.6 Push cart /bullock cart 1.8 1.0 2.2 4.2 2.3 Boat/engine boat 12.9 6.5 2.5 8.0 7.4 Irrigation pump 4.6 13.5 3.0 4.7 6.4 Tiller/tractor 1.3 3.2 0.0 1.3 1.4 Thresher 0.5 1.5 3.0 3.8 2.2 Rice/flour mill/cane crusher 1.2 0.5 0.0 1.6 0.8 Hand tube well 45.6 60.2 30.5 17.5 38.8 Other 7.6 4.3 17.3 10.9 9.7 None 32.7 24.5 42.5 42.2 35.4 Source: BIDS Survey (2004). 77 5580-CHAnnex1.pdf 77 5580-CHAnnex1.pdf 77 3/11/09 10:56:29 AM 3/11/09 10:56:29 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.3 able A1.3 Household Distribution by Various Characteristics (percent distribution) Division Household Characteristic Dhaka Rajshahi Chittagong Khulna All Education No schooling 54.6 54.0 44.5 35.4 47.4 Primary 22.1 21.7 31.9 35.0 27.6 Secondary 18.3 19.5 18.3 24.7 20.0 Higher secondary and above 5.0 4.8 5.3 4.9 5.0 Dwelling ownership status Own 95.2 97.6 98.0 95.6 96.7 Rent 0.3 0.2 0.8 0.4 0.4 Allowed to reside 4.5 2.2 1.2 4.0 2.9 House type Pucca 1.5 1.0 2.8 0.6 1.5 Semi pucca 3.3 6.2 7.0 20.1 8.9 Kacha (but tin roof) 88.4 76.6 62.1 60.9 72.1 Kacha (thatch roof) 6.8 16.2 28.1 18.4 17.5 Latrine type Pucca 11.1 9.2 13.3 6.5 10.1 Slab 37.0 22.5 47.8 39.8 36.9 Kutcha 48.6 67.8 38.4 53.3 51.8 Other 3.3 0.5 0.5 0.4 1.2 Drinking water source Tap inside/outside home 0.5 0.3 1.1 0.2 0.6 Tube well 93.8 99.2 85.9 92.9 92.8 Dug well 0.4 0.3 5.3 -- 1.6 Pond/canal 5.3 0.2 7.7 6.9 5.0 Source: BIDS Survey (2004). 78 5580-CHAnnex1.pdf 78 5580-CHAnnex1.pdf 78 3/11/09 10:56:29 AM 3/11/09 10:56:29 AM Annex 1 Selected Tables from the Household Survey Table A1.4 able A1.4 Annual Household Income by Source (average Tk per household) Source Dhaka Rajshahi Chittagong Khulna All Agriculture 21,096 28,652 31,498 25,546 26,797 Crop 9,225 13,090 13,469 10,590 11,644 Non-crop 9,359 9,846 14,212 10,720 11,092 Wages 2,512 5,716 3,817 4,236 4,061 Non-agriculture 45,510 19,226 48,210 24,544 34,832 Processing/trade/ miscellaneous 15,824 7,277 13,109 8,828 11,349 Wages 6,191 2,442 3,420 3,141 3,810 Transport 3,042 3,047 3,681 2,786 3,156 Salaries/ allowances/pensions 5,900 3,079 6,807 4,897 5,205 Rent (including land mortgages) 2,556 1,915 1,454 1,357 1,825 Remittances 11,997 1,466 19,739 3,535 9,487 Total 66,606 47,878 79,708 50,090 61,629 Source: BIDS Survey (2004). Table A1.5 able A1.5 Household Distribution by Income Source (percent) Source Dhaka Rajshahi Chittagong Khulna All Agriculture 94.5 96.8 95.5 95.4 95.6 Crop 64.8 72.3 68.6 65.9 68.0 Non-crop 91.2 89.2 93.3 90.7 91.1 Wages 23.5 43.7 18.9 31.2 29.1 Non-agriculture 88.2 75.3 78.9 83.6 81.4 Processing/trade/ miscellaneous 35.0 42.8 34.2 41.4 38.2 Wages 29.9 19.2 14.8 21.7 21.3 Transport 10.0 14.3 7.3 13.1 11.1 Salaries/ allowances/ pensions 15.3 8.8 19.1 25.4 17.0 Rent (including land mortgages) 13.3 11.8 5.2 12.4 10.5 Remittances 27.7 11.0 28.6 13.3 20.2 Source: BIDS Survey (2004). 79 5580-CHAnnex1.pdf 79 5580-CHAnnex1.pdf 79 3/11/09 10:56:30 AM 3/11/09 10:56:30 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.6 able A1.6 Household Consumption Expenditure by Budget Item (Tk per year) Item Dhaka Rajshahi Chittagong Khulna All Food 34,511 22,606 39,706 25,693 30,893 Clothing/footwear 3,679 2,374 3,286 2,656 3,012 Consumer durables 5,450 3,375 4,921 3,527 4,347 Miscellaneous 12,264 7,783 14,473 9,013 10,986 All 55,904 36,138 62,386 40,889 49,238 Source: BIDS Survey (2004). Table A1.7 able A1.7 Household Distribution by Type of Energy Consumption (percent) Energy Type Dhaka Rajshahi Chittagong Khulna All Biomass 99.8 99.5 98.6 100.0 99.5 Fuelwood 85.1 67.2 95.8 88.7 84.3 Tree leaves 81.4 72.0 61.6 91.8 76.1 Crop residue* 81.4 93.2 53.6 75.4 75.5 Dung cake/stick 56.9 72.3 29.5 64.8 55.2 Sawdust 0.7 0.8 0.3 1.6 0.8 Non-biomass 100.0 96.5 100.0 100.0 99.1 Candle 2.5 0.8 10.6 4.4 4.7 Kerosene 99.0 91.5 98.4 100.0 97.2 Natural gas -- -- 0.9 -- 0.3 LPG/LNG -- 0.2 1.1 -- 0.3 Grid electricity 43.8 20.3 38.9 10.8 29.0 Solar PV 0.3 -- -- 1.5 0.4 Storage cell 1.3 -- 0.8 0.9 0.8 Dry-cell battery 39.1 50.7 42.3 61.3 48.0 Source: BIDS Survey (2004). * Also includes crop waste and weeds. 80 5580-CHAnnex1.pdf 80 5580-CHAnnex1.pdf 80 3/12/09 7:43:52 AM 3/12/09 7:43:52 AM Annex 1 Selected Tables from the Household Survey Table A1.8 able A1.8 Annual Household Energy Consumption: All Divisions (average per household) Heating Energy Type All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass (kg) Fuelwood 1,186.21 1,064.84 28.60 92.77 -- -- -- Tree leaves 501.51 470.67 29.99 0.85 -- -- -- Crop residue 708.18 538.86 164.41 2.72 -- -- -- Dung cake/stick 523.90 503.68 16.07 4.16 -- -- -- Sawdust 8.40 8.36 0.02 0.02 -- -- -- Non-biomass Candle (piece) 15.86 -- -- -- -- 15.86 -- Kerosene (liter) 28.98 1.76 -- 0.07 -- 27.16 -- Natural gas (Tk) 9.59 9.59 -- -- -- -- -- LPG/LNG (liter) 0.05 0.05 -- -- -- -- -- Grid electricity (kWh) 143.83 0.25 -- 4.00 49.50 80.74 9.34 Solar PV (kWh) 0.53 -- -- -- 0.04 0.48 0.01 Storage cell (kWh) 0.55 -- -- -- -- 0.14 0.41 Dry-cell battery (piece) 15.01 -- -- -- -- -- -- Source: BIDS Survey (2004). Note: Electricity use for motive power is negligible (0.28 kWh/ household/year). Only 2 households used briquettes (Chittagong), but only for 1­2 months. Only 1 household used charcoal, but in negligible quantities. Only 2 households used biogas. None used generator electricity. All categories include non-users so figures may vary due to the percentage of households not using a fuel. However, at the national level they should be representative for all households. Table A1.8.1 able A1.8.1 Annual Household Energy Consumption: Dhaka (average per household) Heating Energy Type All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass (kg) Fuelwood 1,022.74 972.05 44.56 6.13 -- -- -- Tree leaves 536.39 495.65 38.02 2.72 -- -- -- Crop residue 619.10 501.42 116.21 1.47 -- -- -- Dung cake/stick 484.26 456.77 22.35 5.14 -- -- -- Sawdust 3.19 3.19 -- -- -- -- -- Non-biomass Candle (piece) 1.87 -- -- -- -- 1.87 -- Kerosene (liter) 26.02 3.38 -- -- -- 22.63 -- Natural gas (Tk) -- -- -- -- -- -- -- LPG/LNG (liter) -- -- -- -- -- -- -- Grid electricity (kWh) 232.50 0.70 -- 6.48 95.21 114.07 16.04 Solar PV (kWh) 0.66 -- -- -- 0.15 0.48 0.03 Storage cell (kWh) 1.52 -- -- -- -- 0.56 0.96 Dry-cell battery (piece) 9.92 -- -- -- -- -- -- Source: BIDS Survey (2004). 81 5580-CHAnnex1.pdf 81 5580-CHAnnex1.pdf 81 3/11/09 10:56:30 AM 3/11/09 10:56:30 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.8.2 able A1.8.2 Annual Household Energy Consumption: Rajshahi (average per household) Heating Energy Type All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass (kg) Fuelwood 480.05 454.31 25.74 -- -- -- -- Tree leaves 365.47 348.10 17.07 0.31 -- -- -- Crop residue 847.10 723.86 118.47 4.78 -- -- -- Dung cake/stick 737.40 708.79 26.90 1.71 -- -- -- Sawdust 8.70 8.62 0.08 -- -- -- -- Non-biomass Candle (piece) 0.38 -- -- -- -- 0.38 -- Kerosene (liter) 22.08 0.68 -- 0.01 -- 21.39 -- Natural gas (Tk) -- -- -- -- -- -- -- LPG/LNG (liter) 0.02 0.02 -- -- -- -- -- Grid electricity (kWh) 64.10 0.30 -- 0.41 20.84 38.77 3.78 Solar PV (kWh) -- -- -- -- -- -- -- Storage cell (kWh) -- -- -- -- -- -- -- Dry-cell battery (piece) 15.50 -- -- -- -- -- -- Source: BIDS Survey (2004). Table A1.8.3 able A1.8.3 Annual Household Energy Consumption: Chittagong (average per household) Heating Energy Type All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass (kg) Fuelwood 2,069.84 1,714.93 18.70 336.21 -- -- -- Tree leaves 386.68 378.17 8.24 0.27 -- -- -- Crop residue 454.007 369.07 84.80 0.20 -- -- -- Dung cake/stick 235.95 232.82 2.42 0.70 -- -- -- Sawdust 4.50 4.50 -- -- -- -- -- Non-biomass Candle (piece) 51.94 -- -- -- -- 51.94 -- Kerosene (liter) 36.01 2.15 -- 0.06 -- 33.81 -- Natural gas (Tk) 35.81 35.81 -- -- -- -- -- LPG/LNG (liter) 0.15 0.15 -- -- -- -- -- Grid electricity (kWh) 226.15 -- -- 7.31 65.68 139.12 14.04 Solar PV (kWh) -- -- -- -- -- -- -- Storage cell (kWh) 0.25 -- -- -- -- -- 0.25 Dry-cell battery (piece) 15.57 -- -- -- -- -- -- Source: BIDS Survey (2004). 82 5580-CHAnnex1.pdf 82 5580-CHAnnex1.pdf 82 3/11/09 10:56:30 AM 3/11/09 10:56:30 AM Annex 1 Selected Tables from the Household Survey Table A1.8.4 able A1.8.4 Annual Household Energy Consumption: Khulna (average per household) Heating Energy Type All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass (kg) Fuelwood 1,107.29 1,076.16 25.75 5.39 -- -- -- Tree leaves 746.18 685.41 60.70 0.07 -- -- -- Crop residue 941.29 575.80 360.71 4.78 -- -- -- Dung cake/stick 670.05 647.03 13.23 9.78 -- -- -- Sawdust 18.36 18.28 -- 0.08 -- -- -- Non-biomass Candle (piece) 6.07 -- -- -- -- 6.07 -- Kerosene (liter) 31.60 0.70 -- 0.2 -- 30.70 -- Natural gas (Tk) -- -- -- -- -- -- -- LPG/LNG (liter) -- -- -- -- -- -- -- Grid electricity (kWh) 37.40 -- -- 1.32 11.69 21.84 2.55 Solar PV (kWh) 1.57 -- -- -- -- 1.57 -- Storage cell (kWh) 0.42 -- -- -- -- -- 0.42 Dry-cell battery (piece) 19.40 -- -- -- -- -- -- Source: BIDS Survey (2004). Table A1.9 able A1.9 Average Energy Consumption for Agriculture and Transportation (owner­users only) Agriculture Energy Type Irrigation pump Power tiller/tractor Thresher Transport Electricity (kWh) 3643.47 (9) -- 78.13 (17) -- Diesel (liter) 347.95 (103) 238.24 (35) -- 13,003.55 (14) Source: BIDS Survey (2004). Note: Figures in parentheses indicate the number of households that own and operate the relevant equipment. 83 5580-CHAnnex1.pdf 83 5580-CHAnnex1.pdf 83 3/12/09 7:44:07 AM 3/12/09 7:44:07 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.10 able A1.10 Energy Consumption by Business Type Home Enterprise Village Enterprise Institution Users Energy use/ Users Energy use/ Users Energy use/ Energy Type (% total units) unit/year1 (% total units) unit/year1 (% total units) unit/year1 Biomass (kg) 26.3 12.3 10.3 Fuelwood 19.7 395.12 7.3 311.46 9.0 657.00 Tree leaves 7.3 74.54 0.6 2.11 -- -- Crop residue2 15.3 277.62 2.0 949.09 3.2 5.67 Dung cake/stick 5.1 21.64 3.5 149.00 1.9 8.38 Sawdust 2.2 77.52 1.2 6.18 -- -- Briquette -- -- 0.6 12.81 -- -- Charcoal 1.5 2.85 1.5 3.95 -- -- Non-biomass 72.3 98.5 74.4 Candle (piece) 3.6 5.26 21.1 18.16 45.5 100.87 Kerosene (liter) 55.5 13.52 78.9 84.04 37.8 18.66 Diesel/petrol (liter) 5.1 45.59 4.4 60.50 1.3 2.35 Grid electricity (kWh) 25.5 194.68 51.2 360.04 42.3 439.97 Storage cell (kWh) 0.7 3.15 2.6 12.00 7.1 39.00 Dry cell battery (piece) 2.9 0.83 23.7 6.24 12.8 3.50 All 79.6 98.5 75.0 Source: BIDS Survey (2004). 1 Average over all units. 2 Includes crop waste and weeds. Table A1.11 able A1.11 Household Distribution of Cooking Stove Ownership (percent) Stove Type Dhaka Rajshahi Chittagong Khulna All Divisions Clay (fixed) 99.0 99.7 98.6 100.0 99.3 Clay (portable) 30.7 31.7 3.0 11.9 19.2 Kerosene 6.5 1.0 0.8 1.1 2.3 Gas 0.2 -- 1.3 -- 0.4 Electric heater 0.3 0.2 0.2 -- 0.2 Source: BIDS Survey (2004). 84 5580-CHAnnex1.pdf 84 5580-CHAnnex1.pdf 84 3/11/09 10:56:30 AM 3/11/09 10:56:30 AM Annex 1 Selected Tables from the Household Survey Table A1.12 able A1.12 Household Ownership of Cooking Stoves by Type (number per 100 households) Stove Type Dhaka Rajshahi Chittagong Khulna All Divisions Clay (fixed) 200.0 172.3 182.8 174.8 182.7 Clay (portable) 33.2 32.8 3.9 12.6 20.5 Kerosene 7.1 1.0 0.8 1.3 2.5 Gas 0.1 -- 1.3 -- 0.4 Electric heater 0.3 0.2 1.7 -- 0.2 All 240.7 206.3 190.5 188.7 206.3 Source: BIDS Survey (2004). Table A1.13 able A1.13 Household Distribution of Number of Meals and Cooking Times (percent distribution) Number of Meals Cooking Times Dhaka Rajshahi Chittagong Khulna All Divisions 4 2 0.5 -- 0.5 1.6 0.6 4 3 8.6 -- 0.3 -- 0.1 4 4 -- -- 0.2 -- -- 3 1 -- 3.3 8.6 32.5 12.8 3 2 77.4 81.0 47.6 63.1 67.1 3 3 11.6 12.7 38.9 2.2 17.0 2 2 0.7 1.6 0.3 0.4 0.8 2 2 1.2 1.2 3.6 0.2 1.6 1 1 -- -- -- -- -- All -- 0.2 -- -- -- Source: BIDS Survey (2004). Table A1.14 able A1.14 Hours of Operating Cooking Stoves by Type (daily) Stove Type Dhaka Rajshahi Chittagong Khulna All Divisions Clay (fixed) 2.97 2.96 4.01 2.93 3.23 Clay (portable) 1.42 2.63 2.77 1.39 2.01 All 3.36 3.79 4.07 3.06 3.59 Source: BIDS Survey (2004). Note: Kerosene stoves and electric heaters are mainly for occasional use, but more information is needed; data on gas stoves are lacking. 85 5580-CHAnnex1.pdf 85 5580-CHAnnex1.pdf 85 3/11/09 10:56:31 AM 3/11/09 10:56:31 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.15 able A1.15 Household Ownership of Lighting Appliances (number per 100 households) Appliance Type Dhaka Rajshahi Chittagong Khulna All Divisions Grid electrified households Kupi/cherag 153.0 162.3 212.4 161.0 176.7 Hurricane/lantern 80.3 77.0 95.6 81.4 85.3 Petromax 1.1 1.6 -- -- 0.7 Light bulbs 310.6 315.6 414.5 308.5 348.6 Tube lights 36.7 18.9 77.9 20.3 47.0 Charger (include charger torch) -- 0.8 19.3 1.7 7.2 Non-electrified and off-grid electrified households Kupi/cherag 162.5 160.5 210.2 196.7 182.8 Hurricane/lantern 89.7 90.2 98.2 84.7 90.3 Petromax 0.6 0.8 -- 0.4 0.5 Light bulbs -- -- -- -- -- Tube lights 2.9* -- -- 6.7* 2.5* Charger (include charger torch) -- -- 0.5 -- 0.1 All Households Kupi/cherag 158.4 160.8 211.1 192.9 181.0 Hurricane/lantern 85.6 87.5 97.2 84.3 88.9 Petromax 0.8 1.0 -- 0.4 0.5 Light bulbs 136.0 64.2 161.3 33.2 101.2 Tube lights 17.7 3.8 30.3 8.2 15.4 Charger (include charger torch) -- 0.2 7.8 0.2 2.2 Source: BIDS Survey (2004). *It should be noted that 11 households without an electricity connection have solar panels (10) biogas plant (1), which use tube lights. Table A1.16 able A1.16 Hours of Running Lighting Appliances (cases per day) Appliance Type Dhaka Rajshahi Chittagong Khulna All Divisions Kupi/cherag 1.58 (583) 2.05 (582) 2.24 (623) 2.18 (546) 2.01 (2334) Hurricane/lantern 3.32 (375) 4.15 (390) 2.97 (470) 3.63 (347) 3.49 (1582) Petromax 0.38 (4) 0.22 (4) -- 6.90 (2) 1.62 (10) Light bulbs 3.80 (259) 4.09 (121) 3.83 (247) 3.85 (54) 3.86 (681) Tube lights 3.54 (55) 4.65 (13) 3.91 (88) 3.67 (15) 3.82 (171) Torch/charger -- 0.50 (1) 1.74 (41) 2.00 (1) 1.71 (43) All 2.62 (1276) 3.03 (1111) 2.83 (1469) 2.83 (965) 2.82 (4821) Source: BIDS Survey (2004). Note: Figures in parentheses indicate number of cases. 86 5580-CHAnnex1.pdf 86 5580-CHAnnex1.pdf 86 3/11/09 10:56:31 AM 3/11/09 10:56:31 AM Annex 1 Selected Tables from the Household Survey Table A1.17 able A1.17 Household Ownership of Plug-in Electric Appliances (number per 100 households) Appliance Type Dhaka Rajshahi Chittagong Khulna All Divisions Radio/cassette recorder 16.7 4.5 195.3 5.5 118.4 VCR/VCP/television 23.1 6.7 15.8 4.7 12.8 Electric cooling fan 50.1 17.3 56.1 8.0 33.8 Other 9.6 1.2 21.1 1.5 8.7 All 99.5 29.7 288.3 19.7 173.7 Source: BIDS Survey (2004). Table A1.18 able A1.18 Hours of Running the Plug-in Type of Electric Appliances (daily) Appliance Type Dhaka Rajshahi Chittagong Khulna All Divisions Radio/cassette recorder 2.06 (97) 2.95 (26) 2.41 (118) 2.84 (30) 2.38 (271) VCR/VCP/television 3.06 (133) 3.55 (41) 3.76 (101) 3.21 (26) 3.39 (300) Electric cooling fan 7.21 (163) 7.09 (57) 7.75 (177) 8.12 (25) 7.47 (422) Other 4.58 (53) 0.35 (5) 2.81 (129) 0.52 (8) 3.36 (182) All 4.54 (446) 4.87 (129) 4.57 (525) 4.22 (89) 4.62 (1175) Source: BIDS Survey (2004). Note: Figures in parentheses indicate number of cases. Table A1.19 able A1.19 Ownership of Irrigation Pumps and Power Tillers/Tractors (percent households) Pump/Vehicle Type Dhaka Rajshahi Chittagong Khulna All Divisions Irrigation pump (electric) -- 1.3 0.2 -- 0.4 Irrigation pump (diesel) 1.8 12.2 0.5 4.0 4.6 Power tiller/tractor (diesel) 1.3 3.8 -- 1.6 1.7 Source: BIDS Survey (2004). Table A1.20 able A1.20 Hours of Running Pumps and Power Tillers/Tractors (annual hours per unit) Pump/Vehicle Type Dhaka Rajshahi Chittagong Khulna All Divisions Irrigation pump (electric) -- 1,382 1,050 -- 1,345 Irrigation pump (diesel) 441 404 341 283 381 Power tiller/tractor (diesel) 548 239 -- 856 440 Source: BIDS Survey (2004). 87 5580-CHAnnex1.pdf 87 5580-CHAnnex1.pdf 87 3/11/09 10:56:31 AM 3/11/09 10:56:31 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.21 able A1.21 Cost of Household Energy Consumption: All Divisions (Tk per household per year) Heating Energy Source All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass 3,798.50 3,379.57 242.17 176.76 -- -- -- Fuelwood 1,962.25 1,749.59 44.78 167.88 -- -- -- Tree leaves 470.40 440.22 29.14 1.04 -- -- -- Crop residue* 641.23 491.80 147.08 2.35 -- -- -- Dung cake/stick 716.34 689.72 21.15 5.47 -- -- -- Sawdust 8.28 8.24 0.02 0.02 -- -- Non-biomass 1,303.26 65.24 -- 10.56 134.33 -- -- Candle 7.83 -- -- -- -- 7.83 -- Kerosene 607.67 37.04 -- 1.38 -- 569.25 -- Natural gas 9.59 9.59 -- -- -- -- -- LPG/LNG 18.02 18.02 -- -- -- -- -- Grid electricity 487.69 0.59 -- 9.18 134.26 311.89 31.77 Solar PV 1.08 -- -- -- 0.07 0.99 0.02 Storage cell 10.73 -- -- -- -- 2.75 7.98 Dry-cell battery 160.65 -- -- -- -- -- -- All 5,101.76 3,444.81 242.17 187.32 134.33 -- -- Source: BIDS Survey (2004). *Also includes crop waste and weeds. Table A1.21.1 able A1.21.1 Cost of Household Energy Consumption: Dhaka (Tk per household per year) Heating Energy Source All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass 3,802.63 3,475.16 305.32 22.15 -- -- -- Fuelwood 1,610.82 1,530.99 70.18 9.65 -- -- -- Tree leaves 712.86 658.72 50.53 3.61 -- -- -- Crop residue 822.78 666.39 154.44 1.95 -- -- -- Dung cake/stick 653.75 616.64 30.17 6.94 -- -- -- Sawdust 2.42 2.42 -- -- -- -- -- Non-biomass 1,414.24 73.67 -- 13.68 226.97 -- -- Candle 7.64 -- -- -- -- 7.64 -- Kerosene 560.10 72.80 -- 0.17 -- 487.13 -- Natural gas -- -- -- -- -- -- -- LPG/LNG -- -- -- -- -- -- -- Grid electricity 710.35 0.87 -- 13.51 226.67 414.83 54.47 Solar PV 1.35 -- -- -- 0.30 0.99 0.06 Storage cell 18.33 -- -- -- -- 4.69 13.64 Dry-cell battery 116.47 -- -- -- -- -- -- All 5,216.87 3,548.83 305.32 35.83 226.97 -- -- Source: BIDS Survey (2004). 88 5580-CHAnnex1.pdf 88 5580-CHAnnex1.pdf 88 3/11/09 10:56:31 AM 3/11/09 10:56:31 AM Annex 1 Selected Tables from the Household Survey Table A1.21.2 able A1.21.2 Cost of Household Energy Consumption Rajshahi (Tk per household per year) Heating Energy Source All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass 2,541.51 2,361.38 174.11 6.02 -- -- Fuelwood 685.03 648.30 36.73 -- -- -- -- Tree leaves 279.58 266.29 13.06 0.23 -- -- -- Crop residue 648.03 553.75 90.63 3.65 -- -- -- Dung cake/stick 921.74 885.98 33.62 2.14 -- -- -- Sawdust 7.13 7.06 0.07 -- -- -- -- Non-biomass 906.26 22.03 -- 1.66 74.91 -- -- Candle 0.88 -- -- -- -- 0.88 -- Kerosene 474.24 14.56 -- 0.13 -- 459.55 -- Natural gas -- -- -- -- -- -- -- LPG/LNG 6.00 6.00 -- -- -- -- -- Grid electricity 284.24 1.47 -- 1.53 74.91 193.62 12.71 Solar PV -- -- -- -- -- -- -- Storage cell 4.92 -- -- -- -- 1.26 3.66 Dry-cell battery 135.98 -- -- -- -- -- -- All 3,447.77 2,383.20 174.11 7.68 74.91 -- -- Source: BIDS Survey (2004). Table A1.21.3 able A1.21.3 Cost of Household Energy Consumption Chittagong (Tk per household per year) Heating Energy Source All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass 4,901.92 4,177.90 111.01 613.01 -- -- -- Fuelwood 3,762.96 3,117.74 33.99 611.23 -- -- -- Tree leaves 299.67 293.08 6.38 0.21 -- -- -- Crop residue 351.90 286.03 65.72 0.15 -- -- -- Dung cake/stick 478.97 472.63 4.92 1.42 -- -- -- Sawdust 8.42 8.42 -- -- -- -- -- Non-biomass 1,793.87 141.15 -- 19.12 187.15 -- -- Candle 17.46 -- -- -- -- 17.46 -- Kerosene 732.15 43.63 -- 1.25 -- 687.27 -- Natural gas 35.81 35.81 -- -- -- -- -- LPG/LNG 61.71 61.71 -- -- -- -- -- Grid electricity 761.21 -- -- 17.87 187.15 507.43 48.76 Solar PV -- -- -- -- -- -- -- Storage cell 7.07 -- -- -- -- 1.81 5.26 Dry-cell battery 178.46 -- -- -- -- -- -- All 6,695.79 4,319.05 111.01 632.13 187.15 -- -- Source: BIDS Survey (2004). 89 5580-CHAnnex1.pdf 89 5580-CHAnnex1.pdf 89 3/11/09 10:56:31 AM 3/11/09 10:56:31 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.21.4 able A1.21.4 Cost of Household Energy Consumption: Khulna (Tk per household per year) Heating Energy Source All Uses Cooking Parboiling Other Cooling Lighting Amusement Biomass 3,881.41 3,456.80 400.33 24.28 -- -- -- Fuelwood 1,644.33 1,598.09 38.24 8.00 -- -- -- Tree leaves 611.87 562.04 49.77 0.06 -- -- -- Crop residue 771.86 472.16 295.78 3.92 -- -- -- Dung cake/stick 837.56 808.79 16.54 12.23 -- -- -- Sawdust 15.79 15.72 -- 0.07 -- -- -- Non-biomass 1,042.79 14.61 -- 6.88 35.77 -- -- Candle 4.42 -- -- -- -- -- -- Kerosene 660.73 14.61 -- 4.22 -- 641.90 -- Natural gas -- -- -- -- -- -- -- LPG/LNG -- -- -- -- -- -- -- Grid electricity 145.98 -- -- 2.66 35.77 99.73 7.82 Solar PV 3.22 -- -- -- -- 3.22 -- Storage cell 12.98 -- -- -- -- 3.32 9.66 Dry-cell battery 215.46 -- -- -- -- -- -- All 4,924.20 3,471.41 400.33 31.16 35.77 -- -- Source: BIDS Survey (2004). Table A1.22 able A1.22 Household Biomass Energy Behavior: All Divisions (percent distribution) Energy Source Production Gathering Purchasing Any Combination None Fuelwood 9.6 24.6 17.8 10.3 37.7 Tree leaves 30.3 16.7 0.4 29.5 23.1 Crop residue* 36.0 9.8 4.7 27.0 22.5 Dung cake/stick 0.3 0.2 6.7 2.1 90.7 Sawdust -- -- 0.9 -- 99.1 Source: BIDS Survey (2004). *Also includes crop wastes and weeds; but mainly it is crop residue. 90 5580-CHAnnex1.pdf 90 5580-CHAnnex1.pdf 90 3/11/09 10:56:31 AM 3/11/09 10:56:31 AM Annex 1 Selected Tables from the Household Survey Table A1.22.1 able A1.22.1 Household Biomass Energy Behavior: Dhaka (percent distribution) Energy Source Production Gathering Purchasing Any Combination None Fuelwood 10.1 18.1 24.9 11.3 35.6 Tree leaves 34.8 16.7 0.2 30.2 18.1 Crop residue 36.1 5.3 10.8 31.2 16.6 Dung cake/stick 0.2 -- 10.4 3.5 85.9 Sawdust -- -- 0.8 -- 99.2 Source: BIDS Survey (2004). Table A1.22.2 able A1.22.2 Household Biomass Energy Behavior: Rajshahi (percent distribution) Energy Source Production Gathering Purchasing Any Combination None Fuelwood 6.0 30.7 6.8 2.3 54.2 Tree leaves 17.8 24.2 -- 30.3 27.7 Crop residue 38.0 14.2 3.2 42.6 2.0 Dung cake/stick 0.5 0.2 2.8 1.4 95.1 Sawdust -- -- 0.8 -- 99.2 Source: BIDS Survey (2004). Table A1.22.3 able A1.22.3 Household Biomass Energy Behavior: Chittagong (percent distribution) Energy Source Production Gathering Purchasing Any Combination None Fuelwood 12.5 33.4 20.3 13.4 20.3 Tree leaves 27.3 10.5 0.6 23.8 37.8 Crop residue 32.7 10.8 1.1 9.0 46.4 Dung cake/stick 0.5 0.2 2.8 1.4 95.2 Sawdust -- -- 0.3 -- 99.7 Source: BIDS Survey (2004). 91 5580-CHAnnex1.pdf 91 5580-CHAnnex1.pdf 91 3/12/09 7:44:24 AM 3/12/09 7:44:24 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.22.4 able A1.22.4 Household Biomass Energy Behavior: Khulna (percent distribution) Energy Source Production Gathering Purchasing Any Combination None Fuelwood 9.7 15.0 19.2 14.2 42.0 Tree leaves 42.3 15.9 0.9 34.5 6.4 Crop residue 37.8 8.9 3.8 26.1 23.4 Dung cake/stick 0.4 0.2 2.4 1.4 95.6 Sawdust -- -- 1.8 -- 98.2 Source: BIDS Survey (2004). Table A1.23 able A1.23 Household Consumption of Selected Biomass Energies: All Divisions (per year) % Distribution by Source Energy Type Household Consumption (kg/yr) Own Production Gathered Purchased Fuelwood 1,186.21 10.96 49.76 39.28 Tree leaves 501.51 52.23 46.33 1.44 Crop residue 708.18 68.55 24.67 6.78 Dung cake/stick 523.90 72.22 5.25 22.53 Sawdust 8.40 -- -- 100.00 All 2,928.20 42.89 35.00 22.11 Source: BIDS Survey (2004). Table A1.23.1 able A1.23.1 Household Consumption of Selected Biomass Energies: Dhaka (per year) % Distribution by Source Energy Type Household Consumption (kg/yr) Own Production Gathered Purchased Fuelwood 1,022.74 6.30 26.86 66.84 Tree leaves 536.39 55.54 43.89 0.57 Crop residue 619.10 72.16 15.02 12.82 Dung cake/stick 484.26 23.19 9.23 67.58 Sawdust 3.19 -- -- 100.00 All 2,665.68 34.56 24.30 41.14 Source: BIDS Survey (2004). 92 5580-CHAnnex1.pdf 92 5580-CHAnnex1.pdf 92 3/11/09 10:56:32 AM 3/11/09 10:56:32 AM Annex 1 Selected Tables from the Household Survey Table A1.23.2 able A1.23.2 Household Consumption of Selected Biomass Energies: Rajshahi (per year) % Distribution by Source Energy Type Household Consumption (kg/yr) Own Production Gathered Purchased Fuelwood 480.05 7.32 55.42 37.26 Tree leaves 365.47 40.96 58.90 0.14 Crop residue 847.10 67.47 28.28 4.25 Dung cake/stick 737.40 24.98 17.12 57.90 Sawdust 8.70 -- -- 100.00 All 2,438.72 38.57 34.73 26.70 Source: BIDS Survey (2004). Table A1.23.3 able A1.23.3 Household Consumption of Selected Biomass Energies: Chittagong (per year) % Distribution by Source Energy Type Household Consumption (kg/yr) Own Production Gathered Purchased Fuelwood 2,069.84 14.02 61.90 24.08 Tree leaves 386.68 57.77 41.17 1.06 Crop residue 454.07 69.36 28.85 1.79 Dung cake/stick 235.95 82.77 3.58 13.65 Sawdust 4.50 -- -- 100.00 All 3,151.04 32.49 50.14 17.37 Source: BIDS Survey (2004). Table A1.23.4 able A1.23.4 Household Consumption of Selected Biomass Energies: Khulna (per year) % Distribution by Source Energy Type Household Consumption (kg/yr) Own Production Gathered Purchased Fuelwood 1,107.29 7.61 32.44 59.95 Tree leaves 746.18 55.51 40.33 4.16 Crop residue 941.29 66.58 25.66 7.76 Dung cake/stick 670.05 92.49 1.30 6.21 Sawdust 18.36 -- -- 100.00 All 3,483.17 50.10 26.14 23.77 Source: BIDS Survey (2004). 93 5580-CHAnnex1.pdf 93 5580-CHAnnex1.pdf 93 3/11/09 10:56:32 AM 3/11/09 10:56:32 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.24 able A1.24 Operational Land and Household Ownership of Gardens Characteristic Dhaka Rajshahi Chittagong Khulna All Divisions Operational land Households with cultivation (percent) 53.23 69.50 61.72 59.49 61.02 Operational area/household (decimal)* 214.58 376.50 321.76 276.69 276.43 Garden Household ownership (percent) 9.29 13.50 20.63 43.61 21.25 Area under garden/household (decimal)* 78.13 23.38 202.06 26.49 77.31 Source: BIDS Survey (2004). * Refers to ownership households; 1 decimal = 50 square yards. Table A1.25 able A1.25 Household Ownership of Trees and Bamboo Characteristic Dhaka Rajshahi Chittagong Khulna All Divisions Trees % household ownership 89.05 83.67 88.75 89.60 87.75 Number of mature trees/household* 37.18 22.24 79.65 70.19 52.83 Number of immature trees/household* 39.15 34.36 256.27 111.56 113.73 Bamboo groves % household own 47.76 44.33 38.28 44.89 43.71 Number of mature bamboo/household* 119.05 90.77 474.22 69.83 183.53 Number of immature bamboo/household* 74.26 93.92 499.03 79.95 180.19 Source: BIDS Survey (2004). * Refers to ownership households. Table A1.26 able A1.26 Household Ownership of Cattle Characteristic Dhaka Rajshahi Chittagong Khulna All Divisions Cows % household ownership 39.30 40.67 53.44 46.35 45.04 Number of cows/household* 2.63 2.83 2.77 2.48 2.68 Buffaloes % household ownership 0.17 0.83 0.31 1.46 0.67 Number of buffaloes/household* 1.00 2.00 2.00 3.50 2.69 Source: BIDS Survey (2004). *Refers to ownership households. 94 5580-CHAnnex1.pdf 94 5580-CHAnnex1.pdf 94 3/11/09 10:56:32 AM 3/11/09 10:56:32 AM Annex 1 Selected Tables from the Household Survey Table A1.27 able A1.27 Household Distribution by Duration of Electricity Connection (percent distribution) Duration of Connection Dhaka Rajshahi Chittagong Khulna All Divisions No connection 56.2 79.7 61.1 89.2 71.0 With connection (months) 43.8 20.3 38.9 10.8 29.0 01--06 4.6 1.7 2.5 0.6 2.4 07--12 7.3 1.3 3.4 0.7 3.3 13--24 9.6 2.0 5.6 2.6 5.0 25--36 4.5 2.3 5.6 2.9 3.9 37--60 6.3 4.5 4.8 1.6 4.4 61 or more 10.0 7.8 16.4 2.0 9.3 Date not reported 1.5 0.7 0.5 0.4 0.7 Source: BIDS Survey (2004). Table A1.28 able A1.28 Electricity Sources (percent of connected households) Source Dhaka Rajshahi Chittagong Khulna All Divisions Palli Bidyut Samity (PBS) 70.8 84.4 71.5 72.9 73.6 Neighbor with PBS connection 28.4 15.6 17.7 22.0 21.8 Power Development Board (PDB) 0.4 -- 5.6 -- 2.2 Neighbors with PDB connection -- -- 5.2 1.7 2.0 Other (own or local generation) 0.4 -- -- 3.4 0.4 Source: BIDS Survey (2004). Table A1.29 able A1.29 Basis of Charging for Electricity (percent of connected households) Basis Dhaka Rajshahi Chittagong Khulna All Divisions Kilowatt hours consumed 67.8 81.1 75.1 62.7 72.3 Number of electric appliances/bulbs/tubes 15.9 6.6 5.2 18.6 10.7 Monthly fixed charge 7.6 7.4 18.9 10.2 11.8 No charge 1.5 0.8 0.4 -- 0.9 Other 2.7 3.3 0.4 6.8 2.3 Not reported 4.5 0.8 -- 1.7 2.0 Source: BIDS Survey (2004). 95 5580-CHAnnex1.pdf 95 5580-CHAnnex1.pdf 95 3/11/09 10:56:32 AM 3/11/09 10:56:32 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.30 able A1.30 Household Distribution by Electricity Supply Problem (percent of connected households) Problem Dhaka Rajshahi Chittagong Khulna All Divisions Unscheduled power cuts 74.2 69.7 93.6 59.3 79.1 Daily 6.4 0.8 2.0 6.8 3.9 Weekly 10.2 27.1 4.0 20.3 11.8 Rarely 6.4 0.8 -- 10.2 3.5 Never 2.7 1.6 0.4 3.4 1.7 Not reported Voltage fluctuation 44.3 53.3 81.1 55.9 60.1 Daily 14.8 1.6 -- 3.4 6.2 Weekly 34.1 37.7 15.7 32.2 28.0 Rarely 4.2 5.7 2.8 5.1 4.0 Never 2.6 1.6 0.4 3.4 1.7 Not reported Source: BIDS Survey (2004). Table A1.31 able A1.31 Table A1.32 able A1.32 Cost of Kerosene--Using Devices Cost of Electric Bulbs and Tube Lights Monthly Repair Item Annual Cost/Household Purchase Value/ Cost/Household Bulbs/tube lights bought (no.) 5.23 (648) Device Household (Tk) (Tk) Cost (Tk) 111.79 (648) Kupi/cherag 14.05 (2,332) 0.09 (2,327) Source: BIDS Survey (2004). Hurricane lantern 86.09 (1,645) 2.66 (1,641) Note: Figures in parentheses indicate number of cases. Petromax 810.42 (12) 25.42 (12) Source: BIDS Survey (2004). Note: Figures in parentheses indicate number of cases. 96 5580-CHAnnex1.pdf 96 5580-CHAnnex1.pdf 96 3/11/09 10:56:32 AM 3/11/09 10:56:32 AM Annex 1 Selected Tables from the Household Survey Table A1.33 able A1.33 Women's Time Use for Various Activities (daily hours per person) Activity Type Dhaka Rajshahi Chittagong Khulna All Divisions Crop processing 0.24 0.48 0.44 0.55 0.43 Collecting energy 0.22 0.48 0.41 0.31 0.36 Cooking/washing/ feeding/eating 6.26 5.81 6.34 6.18 6.15 Fetching water/washing clothes 1.38 1.62 1.47 1.26 1.44 Cleaning house/bathing 1.87 1.80 1.55 1.48 1.68 Childcare 1.27 1.03 1.29 1.17 1.19 Gainful work 0.83 0.70 0.97 0.54 0.77 Religious work 0.98 0.91 1.22 0.76 0.98 Study/reading 0.07 0.06 0.07 0.04 0.06 Watching TV/listening to radio 1.41 1.37 1.24 1.43 1.36 Visiting neighbors 0.95 1.07 0.77 0.95 0.93 Sleeping 8.19 8.46 8.01 8.61 8.31 Other 0.32 0.20 0.21 0.72 0.35 Source: BIDS Survey (2004). Table A1.34 able A1.34 Table A1.35 able A1.35 Time Spent Making Clay Stoves (hours per stove) Time Spent on Clay Stove Maintenance Stove Type Adult Male Adult Female Times per Total Fixed clay 3.83 5.10 Month Each Repair Monthly Stove Is Hours per Repair Portable burnt clay 6.05 6.88 Stove Type Repaired Incident Hours Source: BIDS Survey (2004). Fixed clay 4.80 0.54 2.59 Portable burnt clay 2.60 0.41 1.07 Source: BIDS Survey (2004). 97 5580-CHAnnex1.pdf 97 5580-CHAnnex1.pdf 97 3/11/09 10:56:32 AM 3/11/09 10:56:32 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.36 able A1.36 Respondents' Attitude toward Electricity (percent distribution) Advantage/Disadvantage Rank­1 Rank­2 Rank­3 Advantage Electric light is superior to kerosene light 70.0 10.0 6.2 Easier to study by electric light than kerosene light 3.6 13.4 4.2 Children devote more time to study at night if electric light 6.5 16.0 12.2 People feel secure at night with electric light 5.3 26.2 18.3 Electric light is not as harmful to health as kerosene light 0.3 3.9 4.2 Running TV is easier by electricity than battery 2.4 5.7 8.5 Electricity costs less than battery to run TV 0.6 2.6 4.6 Hard to work at night without electricity 1.7 6.1 12.4 Easier to entertain guests at night if electricity -- 1.0 2.3 Electricity is important for water supply in our locality 1.1 1.0 2.0 Cooking with electricity causes no smoke 0.2 0.7 1.7 Life is easy with electricity 2.7 5.2 13.1 Disadvantage Children will waste reading time watching TV 16.3 12.1 5.8 Electricity often causes accident that ends in death 54.9 19.3 8.5 Cooking is not advantageous with electricity 4.9 8.7 7.5 Electricity is very expensive 9.3 24.8 19.4 Bulb/tube etc. is expensive 3.0 17.0 22.4 Electric supply is very irregular and voltage is often low 10.1 13.6 22.4 Source: BIDS Survey (2004). Table A1.37 able A1.37 Respondents' Attitude toward Fuelwood (percent distribution) Advantage/Disadvantage Rank­1 Rank­2 Rank­3 Advantage Fuelwood is easy to collect 40.9 22.0 13.3 Fuelwood is cheaper in the market 2.8 10.1 11.5 Fuelwood is obtained from own trees without any cost 25.3 26.1 12.4 Disadvantage Cooking with fuelwood is not advantageous 6.8 4.2 2.5 Fuelwood generates smoke that creates breathing problems 56.1 16.9 8.8 Cooking with fuelwood is harmful to health 3.7 20.5 13.6 Fuelwood is very expensive 12.8 19.2 8.9 Getting fuelwood in the market is difficult 1.5 6.4 7.7 Excessive use of fuelwood is the root cause of deforestation 13.0 19.2 29.0 Source: BIDS Survey (2004). 98 5580-CHAnnex1.pdf 98 5580-CHAnnex1.pdf 98 3/11/09 10:56:33 AM 3/11/09 10:56:33 AM Annex 1 Selected Tables from the Household Survey Table A1.38 able A1.38 Respondents' Attitude toward Kerosene (percent distribution) Advantage/Disadvantage Rank­1 Rank­2 Advantage Cooking is easier with kerosene 22.0 12.4 Kerosene is not expensive for lighting 35.9 31.1 Kerosene is the best way for lighting for our family 28.6 21.2 Disadvantage Use of kerosene is very expensive for cooking 57.4 18.9 Foods cooked with kerosene are harmful to health 24.0 48.8 Kerosene is not easily available in the market 1.8 4.5 Source: BIDS Survey (2004). Table A1.39 able A1.39 Respondents' Attitude toward LPG/LNG (percent distribution) Advantage/Disadvantage Rank­1 Rank­2 Advantage Cooking is easier with LPG/LNG 36.4 1.2 Foods cooked with LPG/LNG are not harmful to health 1.1 19.4 Disadvantage Use of LPG/LNG is very expensive for cooking 25.3 6.0 LPG/LNG is not easily available in the market 8.0 15.5 LPG/LNG cooking devices are expensive 2.0 11.1 Source: BIDS Survey (2004). Table A1.40 able A1.40 Respondents' Knowledge of Price of Kerosene (percent distribution) Factor Dhaka Rajshahi Chittagong Khulna All Divisions Kerosene is cheaper 1.8 0.5 0.5 1.3 1.0 Price is justified 3.8 11.2 9.7 4.6 7.4 Kerosene is costly 93.9 88.2 89.7 93.8 91.3 Do not know 0.5 0.2 0.2 0.4 0.3 Source: BIDS Survey (2004). Table A1.41 able A1.41 Respondents' Reasons for Costliness of Kerosene (percent of those who reported kerosene as costly) Reason Dhaka Rajshahi Chittagong Khulna All Divisions Kerosene is taxed 3.4 7.0 1.4 1.2 3.2 No subsidy on kerosene 0.4 4.7 0.5 0.2 1.4 Iraq--U.S. war 38.0 12.7 12.9 33.5 24.2 Do not know 58.3 75.6 85.2 65.2 71.2 Source: BIDS Survey (2004). 99 5580-CHAnnex1.pdf 99 5580-CHAnnex1.pdf 99 3/11/09 10:56:33 AM 3/11/09 10:56:33 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A1.42 able A1.42 Respondents' Knowledge of Price of Electricity (percent distribution) Factor Dhaka Rajshahi Chittagong Khulna All Divisions Electricity is cheaper 6.3 3.0 1.6 1.8 3.2 Price is justified 8.3 3.7 18.0 6.4 9.3 Kerosene is costly 70.5 54.8 68.4 51.5 61.6 Do not know 14.9 38.5 12.0 40.3 25.9 Source: BIDS Survey (2004). Table A1.43 able A1.43 Respondents' Reasons for Costliness of Electricity (percent of those who reported electricity as costly) Reason Dhaka Rajshahi Chittagong Khulna All Divisions Electricity is taxed 4.5 13.4 2.5 2.1 5.4 No subsidy on electricity 0.5 8.8 0.9 0.7 2.5 Other 74.1 70.5 87.9 82.6 79.0 Do not know 20.9 7.3 8.7 14.6 13.0 Source: BIDS Survey (2004). Table A1.44 able A1.44 Respondents' Knowledge of Renewable Energy (percent distribution) Whether Available in the Market Whether Interested in Purchasing (% of those who have heard about) (% of those who have heard about) Renewable Energy Type Heard About Yes No Yes No Personal biogas plant 16.3 75.1 24.9 78.9 21.1 Improved stove 8.8 59.0 41.0 71.1 28.9 Solar cooker 3.2 47.3 52.7 59.7 40.3 Solar PV light 12.0 31.8 68.2 72.1 27.9 Solar PV lantern 0.9 55.0 45.0 55.0 45.0 Biogas light 3.4 82.7 17.3 67.1 32.9 Solar PV water heater 0.6 61.5 38.5 35.7 64.3 Solar PV water pump 0.6 60.0 40.0 33.3 66.7 Windmill water pump 0.7 68.7 31.3 37.5 62.5 Pressure cooker 25.0 9.0 91.0 57.5 42.5 Source: BIDS Survey (2004). 100 5580-CHAnnex1.pdf 100 5580-CHAnnex1.pdf 100 3/11/09 10:56:33 AM 3/11/09 10:56:33 AM Annex 1 Selected Tables from the Household Survey Table A1.45 able A1.45 Respondents' Reasons for Interest in Purchasing Renewable Energy Devices (percent of those interested in purchasing) Reason Easily Can Buy on Renewable Energy Type Cheaper Available Prestige Advantage Credit Other Personal biogas plant 2.5 -- 3.7 82.7 1.2 9.9 Improved stove -- -- -- 93.1 -- 6.9 Solar cooker -- -- 3.4 82.8 6.9 6.9 Solar PV light -- 2.5 5.1 54.4 30.4 7.6 Solar PV lantern -- -- 11.1 77.8 -- 11.1 Biogas light 3.7 -- 3.7 77.8 -- 14.8 Solar PV water heater -- -- 11.1 77.8 -- 11.1 Solar PV water pump -- -- -- 80.0 -- 20.0 Windmill water pump -- -- -- 80.0 10.0 10.0 Pressure cooker 0.4 2.8 4.3 82.3 -- 10.2 Source: BIDS Survey (2004). Table A1.46 able A1.46 Respondents' Reasons for Disinterest in Purchasing Renewable Energy Devices (percent of those not interested in purchasing) Reason Not Known Unsure Cannot Where Lack of about Not Easily Buy on Renewable Energy Type Available Costly Money Utility Available Credit Other Personal biogas plant 10.3 9.6 23.0 25.9 4.6 2.1 24.5 Improved stove 22.0 12.9 12.9 42.4 3.0 0.8 6.0 Solar cooker 16.7 21.4 16.7 26.2 4.7 2.4 11.9 Solar PV light 7.4 22.3 29.2 13.4 1.0 2.0 24.7 Solar PV lantern 10.0 20.0 10.0 30.0 10.0 -- 20.0 Biogas light 4.0 4.0 34.0 22.0 2.0 2.0 32.0 Solar PV water heater -- -- 25.0 25.0 25.0 -- 25.0 Solar PV water pump -- -- 25.0 25.0 25.0 -- 25.0 Windmill water pump -- -- 25.0 25.0 25.0 -- 25.0 Pressure cooker 1.8 25.2 14.7 12.7 1.2 2.7 21.7 Source: BIDS Survey (2004). 101 5580-CHAnnex1.pdf 101 5580-CHAnnex1.pdf 101 3/11/09 10:56:34 AM 3/11/09 10:56:34 AM 5580-CHAnnex1.pdf 102 5580-CHAnnex1.pdf 102 3/11/09 10:56:34 AM 3/11/09 10:56:34 AM Annex 2 Statistical Models Chapter 3 Table A2.1 able A2.1 Tobit Estimate of Household Energy Demand (N = 2,388) Household Energy Demand Fuelwood Kerosene Diesel Electricity Explanatory Variable (kg/month) (liter/month) (liter/month) (kWh/month) Gender of household head (1 = male, 0 = female) 1.614 0.204 0.064 4.275 Age of household head (yrs.) ­0.051 0.005** 0.001 1.049 Maximum education adult males in household (yrs.) 1.460** 0.019* 0.001 1.485 Maximum education adult females in household (yrs.) 0.759 0.017 ­0.001 2.382 Log of household landholding (decimals)1 2.389** 0.105** 0.025** 4.640 Log of household non-land asset (Tk) 2.591** 0.031** 0.038** 4.448** Energy price Fuelwood (Tk/kg) ­34.732** 0.321** ­0.003 ­0.119 Kerosene (Tk/liter) 5.951** ­0.009 0.002 0.534 Diesel (Tk/liter) ­8.807** ­0.106** 0.001 ­0.411 If village has electricity 14.211** ­0.597** ­0.015 32.782** Proportion irrigated area in village ­2.039 ­0.242** 0.031* 1.739** Whether village has any Paved road 7.282** ­0.033 ­0.025* 9.068 Primary school 9.818** ­0.157* 0.016 ­7.571 Health center ­19.016** 0.053 ­0.017 6.986 Market ­8.738** ­0.002 ­0.018 ­0.972 Commercial bank ­47.877** 0.442* 0.048 31.080 NGO 18.844** ­0.306* ­0.024 5.418 Source: BIDS Survey (2004). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. 11 decimal = 1/100 acre. (continued) 103 5580-CHAnnex2.pdf 103 5580-CHAnnex2.pdf 103 3/11/09 10:56:50 AM 3/11/09 10:56:50 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A2.1 able A2.1 Continued Household Energy Demand Fuelwood Kerosene Diesel Electricity Explanatory Variable (kg/month) (liter/month) (liter/month) (kWh/month) Village price Rice (Tk/kg) ­2.226 ­0.168** 0.002 ­3.040 Flour (Tk/kg) ­5.524** ­0.027 ­0.006 ­2.390 Potato (Tk/kg) ­4.242** 0.016 0.005 9.337* Lentil (Tk/kg) ­4.163** 0.006 0.005* ­2.321 Fish (Tk/kg) ­0.093 ­0.001** ­0.001* 0.173 Beef (Tk/kg) ­1.196** 0.001** 0.001 1.163 Chicken (Tk/kg) ­0.694** ­0.001* 0.0001 0.720 Soybean oil (Tk/liter) 0.210 ­0.019 ­0.004 ­3.139* Milk (Tk/liter) 2.618** 0.001 ­0.002 0.599 Sugar (Tk/kg) 3.173** 0.057** 0.001 ­2.329 Salt (Tk/kg) 10.741** 0.039 0.011 7.966 Village wage (Tk/day) Male 0.704** 0.008** ­0.001 0.192 Female ­0.913** ­0.008** 0.002 ­0.648 Child 0.488** ­0.005 0.001 0.294 Region1 Dhaka 21.673** ­0.079 ­0.066** ­87.485** Chittagong 84.843** 0.172 ­0.157** ­104.459** Khulna 27.633** 0.387** ­0.041* ­88.941** Constant 326.571** 4.702** ­0.870** 4.842 Log likelihood ­11,984.917 1,439.135 ­876.239 1,342.894 Source: BIDS Survey (2004). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. 1 Among the regions Rajshahi is the excluded category. Chapter 4 that energy demand is influenced by energy prices, which can affect household income or consumption indirectly It is difficult to identify the causal effect of access to through demand. It is assumed that energy prices are alternate energy sources on household income and determined by supply and demand, without individual consumption. Since energy demand and household households having any significant effect. Thus, a two- consumption/income are jointly determined, it is difficult stage, least-squared instrument variable was used for this to correlate them without imposing certain restrictions estimation (see Tables A2.2 and A2.3), where demand for on the demand function. The restriction applied here is energy sources was determined in the first phase. 104 5580-CHAnnex2.pdf 104 5580-CHAnnex2.pdf 104 3/11/09 10:56:50 AM 3/11/09 10:56:50 AM Annex 2 Statistical Models Table A2.2 able A2.2 Effects of Energy Use on Household Outcomes by Use Type (N = 2,388) Household per Capita Household Income (Tk/month)1 Consumption Explanatory Variable (Tk/month) Farm Non-farm Total Gender of household head (1 = male, 0 = female) ­0.0440 0.598** ­1.049** 0.074 Age of household head (years) 0.0003 ­0.002 0.009** 0.001 Maximum education of adult males (years) 0.0140** ­0.023* 0.120** 0.024** Maximum education of adult females (years) 0.0005 ­0.022* 0.072** 0.021** Log of household landholding (decimal)1 0.0450** 0.309** ­0.160 0.066** Log of non-land household asset (Tk) 0.0110** 0.162** 0.060** 0.049** Non-lighting energy use (kgoe/year) 0.1270** 0.410** 0.025 0.208** Lighting energy use (klmn­hr/year) 0.1360** ­0.010 0.387** 0.212** If village has electricity ­0.0130 ­0.141 0.903** ­0.058 Proportion irrigated area in village 0.0120 ­0.153 ­0.002 0.021 Whether village has any Paved road 0.0080 0.005 0.136 0.070** Primary school 0.0290 0.158* ­0.047 0.063* Health center ­0.0060 ­0.368** 0.017 ­0.063 Market ­0.0120 0.079 0.132 ­0.021 Commercial bank ­0.0890 0.095 ­0.506 ­0.158* NGO ­0.0780 ­0.020 0.235 0.081 Village price Rice (Tk/kg) 0.0003 ­0.161** 0.210** 0.008 Flour (Tk/kg) ­0.0040 ­0.089** ­0.082 0.011 Potato (Tk/kg) 0.0110 ­0.013 ­0.026 ­0.005 Lentil (Tk/kg) ­0.0050 0.047** ­0.092** ­0.010* Fish (Tk/kg) 0.0010* ­0.004** ­0.003 0.001 Beef (Tk/kg) 0.0020* ­0.004 0.034** 0.007** Chicken (Tk/kg) ­0.0040** ­0.002 ­0.003 ­0.004** Soybean oil (Tk/liter) 0.0010 0.007 0.073** ­0.009* Milk (Tk/liter) 0.0090** 0.020 ­0.001 ­0.009* Sugar (Tk/kg) 0.0090* 0.028 ­0.063* ­0.001 Salt (Tk/kg) 0.0050 0.017 ­0.123* 0.010 Village wage (Tk/day) Male 0.0030** ­0.007** 0.013** 0.004** Female ­0.0020* ­0.001 ­0.009 ­0.003** Child 0.0001 ­0.002 ­0.003 0.0005 (continued) 105 5580-CHAnnex2.pdf 105 5580-CHAnnex2.pdf 105 3/11/09 10:56:50 AM 3/11/09 10:56:50 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A2.2 able A2.2 Continued Household per Capita Household Income (Tk/month)1 Consumption Explanatory Variable (Tk/month) Farm Non-farm Total Region2 Dhaka ­0.0510 ­0.263* 1.207** 0.088 Chittagong ­0.1950** 0.255 0.192 0.145* Khulna ­0.1920** ­0.403** 0.961** ­0.028 Constant 5.0630** 1.158 ­5.955** 6.170** R-squared 0.2220 0.340 0.199 0.423 Source: BIDS Survey (2004). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. Consumption and income variables are in log form. 1 1 decimal = 1/100 acre. 2 Among the regions, Rajshahi is the excluded category. Table A2.3 able A2.3 Effects of Energy Use on Household Outcomes by Energy Source (N = 2,388) Consumption Household Income (Tk/year) Explanatory Variable (Tk/year) Farm Non-farm Total Gender of household head (1 = male, 0 = female) ­0.0420 0.558** ­0.849** 0.040 Age of household head (years) 0.0002 ­0.002 0.011** 0.001 Maximum education of adult males (years) 0.0140** ­0.026** 0.140** 0.023** Maximum education of adult females (years) ­0.0003 ­0.021* 0.073** 0.023** Log of household landholding (decimal)1 0.0480** 0.303** ­0.066 0.057** Log of non-land household asset (Tk) 0.0110** 0.167** 0.075** 0.048** Energy use (kgoe/year) Biomass 0.0840** 0.528** ­0.323 0.144** Kerosene ­0.0160 0.232* ­0.182 ­0.037 Electricity 0.0520** ­0.025 0.042** 0.079** Proportion irrigated area in village 0.0080 ­0.151 ­0.091 0.034 Whether village has any Paved road ­0.0050 0.022 0.083 0.088** Primary school 0.0350 0.179* ­0.142 0.067* Health center ­0.0270 ­0.458** 0.424* ­0.068 Market ­0.0120 0.097 0.116 ­0.160 Commercial bank ­0.1020 ­0.010 ­0.060 ­0.220** NGO ­0.0760 0.036 0.062 0.095 Village price Rice (Tk/kg) ­0.0040 ­0.168** 0.159* 0.013 Flour (Tk/kg) ­0.0040 ­0.095** ­0.067 0.010 Potato (Tk/kg) 0.0110 ­0.011 ­0.010 ­0.008 Lentil (Tk/kg) ­0.0040 0.045** ­0.082** ­0.011** 106 5580-CHAnnex2.pdf 106 5580-CHAnnex2.pdf 106 3/11/09 10:56:51 AM 3/11/09 10:56:51 AM Annex 2 Statistical Models Table A2.3 able A2.3 Continued Consumption Household Income (Tk/year) Explanatory Variable (Tk/year) Farm Non-farm Total Fish (Tk/kg) 0.0010* ­0.004** ­0.0030 0.0010 Beef (Tk/kg) 0.0020 ­0.007 0.0460** 0.0060** Chicken (Tk/kg) ­0.0040** ­0.004 ­0.0005 ­0.0040** Soybean oil (Tk/liter) 0.0004 0.015 0.0370 ­0.0070 Milk (Tk/liter) 0.0080** 0.015 0.0010 ­0.0080 Sugar (Tk/kg) 0.0120* 0.027 ­0.0460 ­0.0050 Salt (Tk/kg) 0.0050 0.016 ­0.0930 0.0090 Village wage (Tk/day) Male 0.0030** ­0.005** 0.0170** 0.0040** Female ­0.0020** ­0.002 ­0.0110 ­0.0030* Child 0.0001 ­0.003 ­0.0020 0.0004 Region2 Dhaka ­0.0720* ­0.358** 1.2820** 0.1000* Chittagong ­0.2000** 0.200 0.2330 0.1670** Khulna ­0.1880** ­0.484** 1.2200** ­0.0460 Constant 5.2500** 2.142* 7.8920** 6.1630** R-squared 0.2100 0.327 0.0560 0.4310 Source: BIDS Survey (2004). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. Consumption and income variables are in log form. 11 decimal = 1/100 acre. 2Among the regions, Rajshahi is the excluded category. Chapter 5 Table A2.4 able A2.4 Tobit Estimate of Enterprise Energy Demand (N = 2,290) Energy Demand (Tk/month) Explanatory Variable Firewood Kerosene Diesel Electricity Age of enterprise (years) 0.156 0.967 ­0.215 0.354 Log of land asset (decimal)1 ­3.889** 0.925 ­1.062* 2.815 Log of non-land asset (Tk 10,000) 8.164** 2.331 12.522** 52.211** Enterprise type2 Agriculture and food 23.649** 88.441** ­25.595** 79.723** Manufacturing 27.822** ­123.354** 9.950* 266.723** Trade ­89.473** ­163.394** ­6.212 ­25.739 If village/growth center has electricity 10.471 ­26.907 ­18.193** 418.724** If village/growth center has paved road ­1.015 ­39.292* ­3.127 63.939 (continued) 107 5580-CHAnnex2.pdf 107 5580-CHAnnex2.pdf 107 3/11/09 10:56:51 AM 3/11/09 10:56:51 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Table A2.4 able A2.4 Continued Energy Demand (Tk/month) Explanatory Variable Firewood Kerosene Diesel Electricity Price of energy source Fuelwood (Tk/kg) 5.246 1.168 0.932 ­10.455 Kerosene (Tk/liter) 11.438* ­1.028 ­3.758 ­14.563 Diesel (Tk/liter) ­4.788 ­11.166 3.287 23.379* Electricity (Tk/kWh) ­6.065 ­15.865 ­13.262* 25.676 Enterprise location3 Growth center ­5.148 48.409 2.351 205.446** Village ­13.800 58.144 ­2.199 198.329** Region4 Dhaka 10.324 1.224 ­5.623 118.407** Chittagong 9.598 97.450** ­21.479** 125.175** Rajshahi 13.801 ­6.282 0.036 34.074 Constant ­277.362* 317.274 ­84.736 ­1,455.246** Log likelihood ­1,689.142 ­11,097.043 ­687.906 ­13,539.475 Source: Data International (2002). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. 1 1 decimal = 1/100 acre. 2 Service-based enterprises are excluded. 3 Home-based locations are excluded. 4 Khulna is excluded. The demand for enterprise energy is influenced effects model that controls for growth-center or village-level more by growth-center/village characteristics than by heterogeneity was used for enterprise outcome regressions the characteristics of individual enterprises; thus, a fixed- (see Tables A2.5 and A2.6). 108 5580-CHAnnex2.pdf 108 5580-CHAnnex2.pdf 108 3/11/09 10:56:51 AM 3/11/09 10:56:51 AM Annex 2 Statistical Models Table A2.5 able A2.5 Effects of Enterprise's Total Energy Use on Profitability (N = 2,290) Explanatory Variable Revenue (Tk/year) Profit (Tk/year) Age of enterprise (years) 1,210.35 196.58 Log of land asset (decimal)1 11,467.57 1,516.09 Log of non-land asset (Tk) 95,605.74** 10,241.58** Enterprise type2 Agriculture and food 108,172.43** 52,372.62** Manufacturing 165,259.12 14,183.76 Trade 236,121.32** 8,902.59* Total energy use (Tk/year) 0.86 0.55** Constant ­410,412.62 60,624.86* F-statistics (9,2061) 11.28 12.42 Source: Data International (2002). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. 11 decimal = 1/100 acre. 2Service-based enterprises are excluded. Table A2.6 able A2.6 Effects of Enterprise's Energy Use on Profitability by Source (N = 2,290) Revenue Profit Explanatory Variable (Tk/year) (Tk/year) Age of enterprise (years) 990.158 153.033 Log of land asset (decimal)1 11,273.32 1,424.647 Log of non-land asset (Tk) 95,669.08** 10,178.643** Enterprise type2 Agriculture and food 106,775.51** 49,891.651** Manufacturing 119,636.91 5,690.698 Trade 229,058.11** 7,567.878 Energy use (Tk/year) Biomass 5.716* 1.217** Kerosene/diesel ­3.872 ­0.271 Electricity 4.107 1.351** Constant ­403,568.82 62,085.22 F-statistics (11,2059) 9.55 11.68 Source: Data International (2002). Note: ** and * represent significance levels of 5 percent (or higher) and 10 percent, respectively. 1 1 decimal = 1/100 acre. 2 Service-based enterprises are excluded. 109 5580-CHAnnex2.pdf 109 5580-CHAnnex2.pdf 109 3/11/09 10:56:51 AM 3/11/09 10:56:51 AM 5580-CHAnnex2.pdf 110 5580-CHAnnex2.pdf 110 3/11/09 10:56:51 AM 3/11/09 10:56:51 AM Annex 3 Consumer's Surplus as a Measure of Welfare Consumer's surplus--a measure of well being that relies on Figure A3.1 Figure A3.1 the difference between what a person, household, or group is willing to pay for energy and what must be paid--has Consumer Demand Curve a long history in economics as a method for estimating the benefits of public projects.47 The first use of applying consumer's surplus for valuing the benefits of electricity in the World Bank was seen as early as 1975.48 This was followed by a more elaborate exploration of the concept in the mid-1980s.49 More recently, the increasingly popular approach is being applied in most rural electrification projects, including such countries as Bolivia, Lao PDR, Peru, and the Philippines.50 Although fairly easy to apply, the procedure is not always well understood and is not without its critics, even by those educated in its underlying principles. The following sections provide a brief primer on the concept and address some of the more frequent criticisms that emerge when applying this method. One criticism not addressed Source: Peskin (2006); and World Bank (2002c). is that money and well being are not necessarily the same thing; it therefore questions whether any monetary measure of benefit is valid. weather, the taste for energy-consuming items such as radio and TV, the need to support energy needs of business, and Consumer Demand most importantly, the consumer's income. Clearly, if one chooses to focus only on the relationship between energy The starting point for an understanding of consumer's prices and energy consumed, these other factors must surplus is the consumer demand curve (see Figure A3.1). somehow be held constant. In general, if all the non-price The demand curve shows the relationship between the factors remain fixed, the higher the price of a good (in this price facing the consumer and the quantity consumed at case energy), the less likely the consumer will demand it. that price. Of course, the quantity of energy consumed at The demand curve shows price­quantity relationship any point in time depends on far more than its price--the for an individual, a household, a group of individuals, or a Note: This section is adapted from Henry M. Peskin, "A Primer on Consumer Surplus and Demand: Common Questions and Answers," ESMAP Knowledge Exchange Series, No. 5, May 2006. 47See Alfred Marshall, Principles of Economics, 8th ed., MacMillan and Co., London, 1930; Hal R. Varian, Microeconomic Analysis, 3rd ed., W. W. Norton and Co., New York, 1978. 48See Dennis Anderson, Cost and Benefits of Rural Electrification: A Case Study of Costa Rica, World Bank Public Utilities Report No. RES 5, World Bank, Washington, DC, 1975. 49See David Pearce and Michael Webb, Economic Benefits of Electricity Supply, Energy Department Paper No. 25, World Bank, Washington, DC, 1985. 50See Kyran O'Sullivan and Douglas F. Barnes, Energy Policies and Multitopic Household Surveys: Guidelines for Questionnaire Design in Living Standards Measurement Study, Energy and Mining Sector Board, Discussion Paper No. 17, World Bank, Washington, DC, 2006. 111 5580-CHAnnex3.pdf 111 5580-CHAnnex3.pdf 111 3/11/09 10:57:09 AM 3/11/09 10:57:09 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities group of households. The interpretation of the curve differs between the original Q , P combination and the new Q , 0 0 1 if it shows an individual or a group, especially if the group P combinations, the total of all gains is represented by the 1 contains a mix of incomes and tastes. shaded triangle a,b,c plus the rectangular area P , P , a, b. 1 0 For a heterogeneous group, the tendency for higher In brief, energy is demanded for the service that prices leading to lower energy consumption usually implies it provides. Therefore, to calculate consumer's surplus that most of the consumption at high prices is by those requires an estimate of demand for lighting, entertainment, individuals with higher incomes or with a stronger desire communications, or other services closely linked to energy. for energy. For an individual, the lower consumption at For lighting, the demand is quantified as kilolumen hours higher prices usually means that the individual will tend and for entertainment, it is radio or television listening to substitute for other goods that provide a less costly or viewing hours. Thus, it is possible to obtain a measure means to maximize his or her satisfaction or well being. of consumer's surplus by using the price and quantity of In either case, if the market price of energy is below what kilolumen or radio listening hours for households using some people would be willing to pay for it, these people kerosene, batteries, or electricity from a grid system.51 This would experience a gain in their well being. This gain is annex assumes that the benefit measured by consumer's the principal argument for using consumer's surplus as a surplus is a satisfactory measure of the benefit of policy benefit measure (see Figure A3.2). that brings about lower energy prices and considers key Ifthe price ofenergyfellfrom P toP , those individuals challenges in this approach. 0 1 (or a particular individual) who would be willing to pay P 0 in order to consume Q of energy, can now consume the Q 0 1 of energy at the lower price. The original "Q consumers" (or Estimating the Demand Curve 0 in the case of an individual, the "Q consumption") realize 0 Consumer's surplus greatly depends on the shape of the a benefit of P ­ P for each of the units of Q consumed. 0 1 0 demand curve, particularly between the original Q , P This benefit would be realized for any consumption less 0 0 combination and the new Q , P combination. If the curve than Q corresponding to any price higher than P , such 1 1 as for those willing to pay P* for Q* units of energy. Thus, 1 1 were not a simple straight line but instead bowed or bent toward the origin of the graph, such as indicated by the when one considers all consumption­price combinations dotted line, the consumer's surplus would be smaller, as shown by the more lightly shaded area. Although the consumer's surplus could be far smaller than that Figure A3.2 Figure A3.2 measured in the World Bank's ESMAP energy studies, the investigators chose not to assume that it is smaller.52 Consumer's Surplus Instead, the studies based the demand curve (and the resulting measure of consumer's surplus) on actual price­quantity observations drawn from household surveys. For combinations not observed, there was a simple linear extrapolation between points that were observed. Of course, the extrapolated points may not fall on the "true" (but unobservable) demand curve. However, since the extrapolations rely solely on what can be observed, the investigators believe that this is a more honest approach than simply assuming some sort of curvaturebetweentheobservablepoints.Suchassumptions are necessarily arbitrary and could produce a wide range of results. Besides the problem of estimating the correct demand curve, application of the consumer's surplus Source: Peskin (2006); and World Bank (2002c). method has raised a number of questions and criticisms.53 51 See World Bank, Rural Electrification in the Philippines: Measuring the Social and Economic Benefits, ESMAP Report 225/02, Washington, DC, 2002. 52 Ibid. 53 Peskin, 2­4. 112 5580-CHAnnex3.pdf 112 5580-CHAnnex3.pdf 112 3/11/09 10:57:10 AM 3/11/09 10:57:10 AM References Asaduzzaman, M., and A. Latif. 2005. "Energy for Rural -------- . 2005. "Unlocking the Potential: National Strategy Households: Towards a Rural Energy Strategy in for Accelerated Poverty Reduction." Dhaka: General Bangladesh." Bangladesh Institute of Development Economics Division, Planning Commission. Studies, Dhaka. Hossain, M. M. Golam. 2003. "Improved Cookstove and Barkat, A., M. Rahman, S. Zaman, A. Podder, S. Halim, Biogas Programs in Bangladesh." Energy for Sustainable N. Ratna, M. Majid, A. Maksud, A. Karim, and S. Islam. Development 7(2): 97­100. 2002. "Economic and Social Impact Evaluation Study of the Rural Electrification Program in Bangladesh." Hutton,G.,andE.Rehfuess.2006."GuidelinesforConducting Report to National Rural Electric Cooperative Cost-Benefit Analysis of Household Energy and Health Association (NRECA) International, Dhaka. Interventions to Improve Health." Paper prepared for World Health Organization, Geneva. BBS (Bangladesh Bureau of Statistics). 2005. Report of the Agricultural Sample Survey of Bangladesh, vol. 1. Dhaka: Hutton, G., E. Rehfuess, F. Tediosi, and S. Weiss. 2006. Bangladesh Bureau of Statistics. "Evaluation of the Costs and Benefits of Household Energy and Health Interventions at Global and Dasgupta, S., M. Huq, M. Khaliquzzaman, K. Pandey, and Regional Levels." Paper prepared for World Health D. Wheeler. 2004. Who Suffers from Indoor Air Pollution?: Organization, Geneva. Evidence from Bangladesh. World Bank Policy Research Working Paper 3428. Washington, DC: World Bank. IEA (International Energy Agency). 2003. Key World Energy Statistics 2003. Paris. DataInternational. 2004."EnergyUsagebyRuralEnterprises of Bangladesh." Background Report, Dhaka. Islam, M. N. 1980. "Village Resources Survey for the Assessment of Alternative Energy Technology." Report Ghimire, Prakash C. 2005. "Final Report on Technical submitted to International Development Research Study of Biogas Plants Installed in Bangladesh." Centre, Bangladesh University of Engineering and Report submitted to National Programme on Domestic Technology, Dhaka. Biogas in Bangladesh (partnership of Netherlands -------- Development Organization and IDCOL), Dhaka. . 1986. Rural Energy Survey in the Third World: A Critical Review of Issues and Methods. Ottawa: International GOB (Government of the People's Republic of Bangladesh). Development Research Centre. 1987. Bangladesh Energy Planning Project: Final Report, -------- vols. I­VII. Prepared by Sir Halcrow and Partners, . 2001. "Energy Strategy for Rural Bangladesh." Motor Columbus Consulting Engineering, Inc., Background paper prepared for World Bank, Dhaka. Petronus Consultants in Association with TSL and Khalequzzaman, M. 2005. "Assessment of Rural Energy PSL, Dhaka. Delivery Mechanisms in Bangladesh." Background -------- . 1996. "National Energy Policy," Ministry of Energy study prepared for World Bank Energy Sector and Mineral Resources. Bangladesh Gazette (January), Management Assistance Program (ESMAP), Dhaka. Washington, DC. 113 5580-CHRef.pdf 113 5580-CHRef.pdf 113 3/11/09 10:59:08 AM 3/11/09 10:59:08 AM Special Report Restoring Balance: Bangladesh's Rural Energy Realities Kulkarni, V., Douglas Barnes, and Sandro Parodi. 2007. Pitt, Mark M., Mark R. Rosenzweig, and Md. Nazmul "Rural Electrification and School Attendance in Hassan. 2005. Sharing the Burden of Disease: Gender, Nicaragua and Peru." Draft paper, World Bank, the Household Division of Labour and the Health Effects Washington, DC. of Indoor Air Pollution. CID Working Paper No. 1119. Boston: Center for International Development, Leach, Gerald. 1987. "Household Energy in South Asia." Harvard University. Biomass 12: 155­184. United Nations. 2005. Energy Services for the Millennium McKenzie, Wood. 2006. "Gas Sector Master Plan and DevelopmentGoals.JointpublicationofUNMillennium Strategy for Bangladesh." Interim report prepared for Project, United Nations Development Programme Petrobangla, Dhaka. (UNDP),WorldBank,andEnergySectorManagement Assistance Program (ESMAP). Washington, DC and Nexant. 2006. "Power Development Program, Draft Final New York: World Bank/ESMAP and UNDP. Report: Master Plan Update." Prepared for Asian Development Bank and Ministry of Power, Energy, World Bank. 2002a. Energy Strategy for Rural India: Evidence and Mineral Resources. from Six States. ESMAP Formal Report 258/02. Washington, DC. Openshaw, Keith. 2004. "Bangladesh: Biomass Energy Supply." Background study prepared for World Bank -------- . 2002b. India: Household Energy, Indoor Air Pollution, Energy Sector Management Assistance Program and Health. ESMAP Formal Report 261/02. (ESMAP), Washington, DC. Washington, DC. O'Sullivan, Kyran, and Douglas F. Barnes. 2006. Energy -------- . 2006. World Development Report 2007: Development and Policies and Multitopic Household Surveys: Guidelines for the Next Generation. Washington, DC. Questionnaire Design in Living Standards Measurement Studies. Energy and Mining Sector Board, Discussion WHO (World Health Organization). 2006. Fuel for Life: Paper No. 17. Washington, DC: World Bank. Household Energy and Health. Geneva. 114 5580-CHRef.pdf 114 5580-CHRef.pdf 114 3/11/09 10:59:08 AM 3/11/09 10:59:08 AM Glossary aman rice planted in July­September and harvested in November­December aus rainfed summer rice planted in March­April biri tobacco boro irrigated winter rice planted in December­February and harvested in April­ June gur molasses hachak (see petromax) kupi/cherag traditional kerosene lamp motive power energy source used for powering machinery paan betel quid petromax pressurized kerosene lamp thana/upazila rural subdistrict 115 5580-CHGloss.pdf 115 5580-CHGloss.pdf 115 3/11/09 10:58:28 AM 3/11/09 10:58:28 AM 5580-CHGloss.pdf 116 5580-CHGloss.pdf 116 3/11/09 10:58:28 AM 3/11/09 10:58:28 AM Special Report Series RENEWABLE ENERGY THEMATIC AREA East Asia and Pacific Region (EAP) Sustainable and Efficient Energy Use to Alleviate Indoor Air Pollution in Poor Rural Areas in China (002/07) Latin America and the Caribbean Region (LCR) Nicaragua Policy & Strategy for the Promotion of Renewable Energy Resources (003/07) Global (GLB) Considering Trade Policies for Liquid Biofuels (004/07) ENERGY POVERTY THEMATIC AREA Global (GLB) Risk Assessment Methods for Power Utility Planning (001/07) South Asia Region (SAR) Restoring Balance: Bangladesh's Rural Energy Realities (006/09) ENERGY SECURITY THEMATIC AREA Global (GLB) Coping with High Oil Price Volatility (005/08) 117 5580-CHList.pdf 117 5580-CHList.pdf 117 3/11/09 10:58:47 AM 3/11/09 10:58:47 AM 5580-CHList.pdf 118 5580-CHList.pdf 118 3/11/09 10:58:47 AM 3/11/09 10:58:47 AM Energy Sector Management Assistance Program (ESMAP) Purpose The Energy Sector Management Assistance Program is a global knowledge and technical assistance partnership administered by the World Bank and sponsored by bilateral official donors since 1983. ESMAP's mission is to assist clients from low-income, emerging, and transition economies to secure energy requirements for equitable economic growth and poverty reduction in an environmentally sustainable way. ESMAP follows a three-pronged approach to achieve its mission: think tank/horizon-scanning, operational leveraging, and knowledge clearinghouse (knowledge generation and dissemination, training and learning events, workshops and seminars, conferences and roundtables, website, newsletter, and publications) functions. ESMAP activities are executed by its clients and/or by World Bank staff. ESMAP's work focuses on three global thematic energy challenges: · Expanding energy access for poverty reduction; · Enhancing energy efficiency for energy secure economic growth, and · Deploying renewable energy systems for a low carbon global economy. Governance and Operations ESMAP is governed and funded by a Consultative Group (CG) composed of representatives of Australia, Austria, Denmark, France, Germany, Iceland, the Netherlands, Norway, Sweden, the United Kingdom, and the World Bank. The ESMAP CG is chaired by a World Bank Vice President and advised by a Technical Advisory Group of independent energy experts that reviews the Program's strategic agenda, work plan, and achievements. ESMAP relies on a cadre of engineers, energy planners, and economists from the World Bank, and from the energy and development community at large, to conduct its activities. Further Information For further information or copies of project reports, please visit www.esmap.org. ESMAP can also be reached by email at esmap@worldbank.org or by mail at: ESMAP c/o Energy, Transport, and Water Department The World Bank Group 1818 H Street, NW Washington, DC 20433, USA Tel.: 202-473-4594; Fax: 202-522-3018 ENERGY AND POVERTY Just less than one-half of the people in devel- oping countries have no access to electricity and a similar number are reliant on biomass energy for cooking and heating. As a conse- quence, they are deprived of the means of moving out of poverty. Greater access to mod- Energy Sector Management Assistance Program ern energy services can improve poor people's 1818 H Street, NW Washington, DC 20433 USA income through enhancement of productive Tel: 202-458-2321 use of energy and it can also increase their Fax: 202-522-3018 quality of life by providing quality lighting, Internet: www.esmap.org Email: esmap@worldbank.org communication, and other important services. ESMAP has the goal of substantially improv- ing energy use by poor people through ad- dressing the widespread problems of the household energy. This is done through high quality analytical work on energy access, promoting an increase in the quality and number of projects dealing with energy and poverty issues by international donors, and by disseminating successful approaches to the international development community.