Burkina Faso - Urban household energy strategy

Report No. 134/91 / /  JOINT UNDPIWORLD BANK ENERGY SECTOR MANAGEMENT ASSISTANCE PROGRAMME (ESMAP) PURPOSE The Joint UNDP/World Bank Energy Sector Management Assistance Programme (ESMAP) was launched in 1983 to complement the Energy Assessment Progamme which had been established three years earlier. An international Commission was convened in 1990 to address the creation of ESMAP's role in the Nineties. It concluded that the Programme had a crucial part to play over the next decade in assisting the developing countries to better manage their energy sectors given that the supply of energy at reasonable prices is a critical determinant of the pace and magnitude of the growth process. The Commission's recommendations received broad endorsement at the November 1990 ESMAP Annual Meeting. Today, ESMAP is carrying out energy assessments, preinvestment and prefeasibility activities and is providing institutional and policy advice. The program aims to strengthen the impact of bilateral and multilateral resources and private sector investment through providing technical assistance to the energy sector of developing countries. The findings and recommendations emerging from ESMAP activities provide governrments, donors, and potential investors with the information needed to identify economically and enviromnentally sound energy projects and to accelerate their preparation and implementation. ESMAP's operational activities are managed by two Divisions within the Industry and Energy Department at the World Bank and an ESMAP Secretariat. o The Programme's activities are governed by the ESMAP Consultative Group which consists of its co- sponsors, the UNDP and the World Bank, the govermnents which provide financial support and representatives of the recipients of its assistance. The Chairman of the Group is the World Bank's Vice President, Sector Policy and Research. He is assisted by a Secretariat headed by the Group's Executive Secretary who is also responsible for relations with the donors and securing funding for the Programme's activities. The Secretariat also gives support and advice to a Technical Advisory Group of independent energy experts which meets periodically to review and scrutinize the Programme's strategic agenda, its work program and other issues related to ESMAP's functioning. * The ESMAP Strategy and Programs Division is responsible for advising on which countries should receive ESMAP assistance, preparing relevant ESMAP programs of technical assistance to these countries and supports the Secretariat on funding issues. It also carries out broadly based studies such as energy assessments. X The ESMAP Operations Division is responsible for the detailed design and implementation of tasks consisting mainly of sub-sectoral strategy formulation, preinvestment work, institutional studies, technical assistance and training within the framework of overall ESMAP country assistance programs. FUNDING The ESMAP represents a cooperative international effort supported by the World Bank, the United Nations Development Programme and other United Nations agencies, the European Community, Organization of American States (OAS), Latin American Energy Organization (OLADE), and a number of countries including Australia, Belgium, Canada, Denmark, Germany, Finland, France, Iceland, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Portugal, Sweden, Switzerland, the United Kingdom and the United States. FURTHER INFORMATION For further information or copies of completed ESMAP reports, contact: Office of the Director OR The Executive Secretary Industry and Energy Department ESMAP Consultative Group The World Bank The World Bank 1818 H Street N.W. 1818 H Street, N.W. Washington, D.C. 20433 Washington, D.C. 20433 U.S.A. U.S.A.  BURKINA FASO URBAN HOUSEHOLD ENEW:GY STRATEGY JUNE 1991  EXCHANGE RATE US$ = CFA Francs 300 CONVERSION is'ACrORS USED General Units 1 ton crude oil equivalent (TOE) = approximately 42 GJ 1 kWh = 3.6 MI Fuel Tom3 MVJkg Charcoal 29.0 LPG - 45.7 Kerosene 0.79 43.5 Fuelwood Volume and Weight Units 1.0 m3 3.2 steres 1.0 stare 0.31 mi3 1.0 m3 0.80 tons (density) 1.0 stre 250.0 kg 4.0 steres 1.0 ton 1.0 stere 36.0 standard bundles Standing Volume of Commercial Fuelwood For unmanaged natural savannah forest, 60% of the total volume is in commercial fuelwood species of usable diameter. For rmanaged natural savannah forest, 80% of the total volume is assumed to be in commercial species of usable diameter.  ABBREVIATIONS AND ACRONYMS CAPRO Ministry of Commerce DSVF MET Extension Service ESMAP Energy Sector Management Assistance Program FAO Food and Agriculture Organization IBE Burkinabe Institute for Energy MAE Ministry of Agriculture MEFSN Family Ministry MESRS Ministry of Higher Education and Scientific Research MET Ministry of En" '.onment and Tourism SONAHBY Burkinabe Oil Company TOE Ton of oil equivalent  TABLE OF CONTENTS EXE CUTIVDESU MMARY ................................. i I. INTRODUC1lON ........................1 H. BiACKGROUNID ....................................... ... 3 A. Ec3onomic and Demographic Trends 4 B. Energy and the Environment .. 4 Energy Balance .. 4 Household Energ Demand .. Biomass Resources .. 9 Managing the Fuel Transition ..10 C. Government Household Energy Strategy ............ 11 Household Energy: A National Program .. 11 Institutions and Policy Framework ..12 III. DEMAND MANAGEMENT: DISSEMINATION Or IMPROVED STOVES . 14 A. Program Context ............. .............. 14 B. The National Improved Stove Program ........................... 15 Stove Models .............. .............. 15 C Stove Production and Distribution ............ ................... 17 The Incentive Environment and Stove Production ........ ............ 18 Marketing of Improved Stoves .................................. 18 D. Improved Stove Program: Results ............................... 21 IV. PROMOTION OF SUBSTITUTE FUELS: LPG AND KEROSENE AND CONSUMPTION OF ELECIRIClTY ............................... 26 A. PolicyContext ........................................... 26 Choice of Substitute Fuels: LPG, Kerosene and Electricity ............. 27 B. Substitute Fuels: Infrastructure and Security of Supply ....... .......... 28 Stove Equipment ........................................... 30 C. Marketing LPG and Kerosene ................................... 31 Structure of LPG and Kerosene Prices ....... .................... 32 D. LPG and Kerosene Use: Program Results ......................... 34 E. Inter-fuel Substitution: Medium and Longer Term Prospects .... ........ 36 Switching to Fuelwood Substitutes: Economic Implications ............. 38 V. MANAGING HOUSEHOLD ENERGY DEMAND: THE IMPACT OF PRICING .OL.CIES ......................... 39 A. Policy Context ........................... 39 B. PricingPolicy: Issuesand Options ......................... 39 Government Revenues ............ ............. 40 Relative Fuel Prices ......................... 41  Polcy Trde-o ... ........................................ . 43 C. The Impact of Economic Pricing on lnter-Fuel Substitution . ............ 44 VI. SUPPLY MANAGEMENTOFFUELWOOD ......................... 46 A. he Importance of Foresty Management ......... .. ............... 46 The urban Supply System ....... ............................... 47 Pricing .................................................... 49 B. A Strategy for Natural Forest Management ......... .. .............. 52 Urban Fuelwood Supply Stategies ............................... 54 Me Economics of Natural Forest Management ......... ............ 56 VII. AN URBAN HOUSEHOLD ENERGY STRATEGY: REINFORCING GOV'ERNEMEPENT POUCY .... 63 A. Introduction ................................................ 63 A Revitaied Govemnment Strategy ............. ................. 63 B. Urban Woodfuel Conservation Program .......... .. ................ 68 Short-term Measures ................... ...................... 68 C. Inter-fuelSubstitution Program .................................. 69 Short-term Measures ................... ...................... 70 D. Natural Woodlands Management ............... .................. 70 Short-term Measues ................... ...................... 71 ANNEXES I Fuelwood Demand Forecasts ...................................... 73 II Factors Influencing Urban Household Energy Demand ................... 78 Hi Annual Financial Cost of Cooking by Type of Fuel, Family of Six Persons, Ouagadougou, 1987 ............ ............... 84 IV lbe Security of Supply for Kerosene and for LPG .......... ............. 85 V The Price Structures for LPG and Kerosene ............... ............ 87 VI Key Assumptions of LPG and Kerosene Penetration ......... ............ 89 VII The Economic Value of Fuelwood Substitution ............ ............. 91 VIII Denifing a Sustainable Pricing Policy ................................. 95 IX Legl Tebs .................................. 97 X Urban Fuekvood Supply Strategies ................................. 100 A. Ouagadougou ............ ...................... 100 B. Koudougou ................. ................. 105 C. Bobo-Dioulasso ...... 108 D. ouahigouya ................ .................. 112  Xl The Outline of a Basic Management Plan ............................ 142 XII Ile Economics of Natural Forest Management ........ ................ 143 xm Project Proposals: Detailed Description ............. ................ 144 XIV Tle Impact on Fuelwood Demand of Demand Side Interventions ........ .. 153 XV Economic Value of Improved Stove Program ............ .............. 154 TABLES 2.1 Final Energy Consumption By Sector, 1987 ............... . 4 22 Rural and Urban Household/Informal Sector Energy Consumption, 1987 ............................ 5 2.3 Fuel Use for Cooking in Urban Population ............................ 6 2.4 Househol Size and Average Consumption in LPG Consuming Households .................................... 8 2.5 Penetration of Electricity in Urban Households, 1987 ....... .............. 8 2.6 Informal Sector Fuehwood Consumption in Ouagadougou ....... ........... 9 3.1 Market Prospects for Improved Stoves Door-to-Door Sales and Acceptability Tests ............................... 16 3.2 The Burkido Stove: BasicData ..................................... 16 3.3 Price Structure of Improved Metal Stoves ............ ................. 17 3.4 Campaign Costs and Sales ........................................ 20 3.5 Improved Stoves: Rate of Market Penetration by City ......... ........... 22 3.6 Benefits and Costs of Urban Improved Stove Progam .................... 25 4.1 Cost of Investment in 3 kg and 12 kg LPG Stoves in FCFA ...... .......... 30 4.2 Price Structures of LPG and Kerosene .............. ................. 33 5.1 Taxation of Household Fuel 1987 . ................................. 41 6.1 Official and Actual Structure of Fuelwood Prices ........ ................ 50 62 Proposed Fuelwood Price Structure .................................. 51 6.3 Main Cover Type Classification Used with the Urban Fuelwood Supply Zone Maps .57 6.4 Estimated Annual Productivity of Natural Forests of Selected Priority Areas of Fuehwood Supply Zones .58 6.5 Population Projections and Household Woodfuel Demand Estimates for Ouagadougou ........................................ 59 6.6 Population Projections and Household Woodjuel Estimates for Koudougou .... 60 6.7 Population Projections and Household Woodfuel Estimates for Bobo-Dioulasso .61 6.8 Population Projections and Household Woodfuel Estimates for Ouahigouya .62  A2.1 Energy Saving Behaviour in Urban Households ......................... 79 A22 Expenditures on Fue}wood in Percent of Household Income ............... 82 A7.1 Results of Sustainable Fuelwood Substitution ........................... 92 A10.1 Standing Fuelwood Estimates for Priority Areas within the Ouagadougou Fuelwood Supply Zone .116 A10.2 Estimated Annual Productivity of Priority Areas within the Ouagadougou Fuelwood Supply Zone .117 A103 Standing Fuelwood Estimates for Priority Areas within the Koudougou Fuehwood Supply Zone .118 A10.4 Estimated Annual Productivity of Priority Areas within the Koudougou Fuehwood Supply Zone .119 A10.5 Standing Fuehvood Estimates for Priority Areas within the Bobo-Diouilasso Fuelwood Supply Zone .120 A10.6 Estimated Annual Productivity of Priority Areas within the Bobo-Dioulasso Fuelwood Supply Zone .121 A10.7 Standing Fu_.--ood E-siiinates for Priority Areas within the Ouahigouya Fuelwood Supply Zone .122 A10.8 Estimated Annual Productivity. of Priority Areas within the Ouahigouya Fuetwood Supply Zone ....... ............... 123 FIGURES 2.1 Ouagadougou Household Fuelwood Consumption by End-Use, 1987 ..... ..... 6 3.1 Market Penetration of Improved Stoves by Income Group ...... ........... 23 3.2 Level of Household Income & Penetration of Different Types of Improved Stoves ................. ..................... 23 4.1 Penetration of LPG Use by Income Group Urban Households in 1987 .... .... 35 4.2 LPG and Kerosene Use by Income Group ............. ................ 37 4.3 LPG Penetration Ratios and Consumption Levels Needed for Sustainable Demand .38 5.1 Relative Economic and Financial Cost of Cooking by Type or Fuel and Stove .............. ............................... 42 6.1 Impact of Demand Interventions ................................... 46 7.1 Coverage of Urban Demand Growth ................................. 67 A1.1 Fuetwood Demand/Supply Balance .................................. 73 A1.2 Rural and Urban Fuelwood ............. ........................... 73 A2.1 Household Size and Energy Consumption ............................. 78 A2.2 Frequency of Different lTpes of Meals .............................. 80 A2.3 Impact of Household Income on Energy Consumptions ..... ............. 81 A7.1 Taxation and Consumer Welfare ................................... 91 A8.1 Relationship between the Rate of Penetration of Substitute Fuels and the Relative Price of Substitute Fuels to Woodfuels .96 A8.2 The mix of Decreasing Subsidies on Modem Fuels and Increasing Taxes on Woodfuels that is needed in Order to Obtain the Desired Ratio of Relative Prices in the Year 2000 .96  1. This report addresses the accelerating demand for woodfuels as an energy source for Burkina Faso's urban households and its impact on the country's natural resource endowment. It sets out an urban household energy strategy, anchored in Government's on-going efforts, that aims to slow current fuelwood consumption and promote a broader scale substitution of alternative fuels than is presently the case. This strategy is the result of a coliaborative venture, begun in 1987, between the Government of Burkina Faso and the joint UNDP/World Bank Energy Sector Management AssiAance Program (ESMAP). It was co-financed by the Governments of Norway and the Netherlands. Energ and Environment 2. Energy policy is a critical instrument of environmental management for Burkina Faso, which has no known hydrocarbon resources. Fuelwood is the principal source of energy and the household sector its most important fuelwood consumer, accounting for 3 million tons per annum, 98% of household primary energy demand, and 89% of fnal energy demand. The informal sector represented as much as 20% of total urban woodfuel consumption, where demand is concentrated in a few energy intensive activities: the informal sector beer brewers, consisting largely of women, account for about 12% of total urban fuelwood consumption and about 60% of informal sector fuehvood consumption. Biomass is the most prevalent energy source even in industry, where consumption is three times as high as that of oil-based products. 3. Burkina's overall energy consumption in 1987 of 1.45 million tons oil equivalent is one of the world's lowest, expressed in terms of per capita energy consumption (160 kg of oil equivalent); in terms of per capita consumption of oil products (17 ag); or in terms of per capita consumption of electricity (14 kWh). Yet even the satisfaction of this low level of demand for oil products represents a very real challenge to policymakers. In 1987, the cost of importing 130,000 tons of oil products amounted to US$ 47 million, or 16% of export income and 6.5% of total imports of goods and services. In light of these costs and competing development priorities for scarce resources, the transition from biomass to modern fuels, which is in process, requires careful management. Indeed, the key energy issue facing the country is how to achieve an ecologically sustainable balance between fuelwood demand and fuelwood supply with minimal consequences on oil imports. 4. An array of obstacles constrain Burkina's development prospects, with widespread poverty and rapid demographic growth posing the greatest challenge. They are the principal causes of the persistent erosion of natural resources. A total population of 9.8 million in 1989 is projected to double over the next twenty years and reach 13.8 million by the year 2000. Although the urban population represented no more than 11% of the total population in 1989, the urban sector is a significant feature of national economic growth, accounting for an average 68% of GDP (constant 1979 prices) from 1982-1987. Urban economic growth - and its claim on natural resources to fuel that growth - is expected to be further enhanced in the future. The urban population is growing at about 8% a year, more than double the overall rate of population growth. By the end of the century, the urban share will climb to 21% of the total population of 13.8 million, or 2.9 million.  5. Woodfuels will c-ntinue to be the main energy source for the majority of Burkina Faso's population in the future. National woodfuel consumption will rise from 3 million tons in 1988 to 4.5 million tons in 2000 and urban fuelwood consumption from .33 million tons to 0.90 million tons. At the same time, this report estimates that the forest cover is presently decreasing at about 60,000 hectares per year: land clearing for agricultural development accounts for about 40,000 hectares, while the remainder is the result of bushfires, drought and the cutting of wood for fuelwood consumption in mostly unmanaged natural forests. 6. Total natural wood production is currently estimated to be about 10 million m3 (6- 7 million tons) per year. The Ministry of Environment and Tourism (MET) estimates that, of the total woodfuel demand of 3 million tons per annum, 2.5 million tons are available for fuelwood consumption in the sense of being accessible to the local population and corresponding to local demand. In 1987, the deficit of 0.5 million tons, representing the difference between demand and presently sustainable supply, was met by clearing ef the forest cover (cutting of green or remaining dead wood). How much of this is a by-product of agricultural land clearing and l,ow much is due to the rapid growth in the urban demand for biomass is unknown. Government PoliEc 7. Government acknowledged the impo.-tance of natural resource management to economic development in a far reaching forest protection program, ALes Trois Luttese, which was set in motion in 1985 to reduce the impact of bush fires, uncontrolled grazing and overcutting of fuelwood on its natural environment. The Government implemented a household energy strategy with complementary actions in the areas of (a) demand reduction, (b) fuel switching and (c) fuelwood supply management. This strategy can be characterized as follows: (i) The demand managment effort was anchored in the dissemination of improved woodstoves and the promotion of LPG and kerosene. The supply manag nt initiatives centered around a change in the fuelwood price structure in favor of fuelwood producer and the creation of woodcutters cooperatives. (ii) Pricing poliQc is an important policy instrument of this program: For budgetary reasons, no subsidies are given to the fuelwood substitute fuels LPG and kerosene. Instead, the Government increased the prices of woodfuels by raising both cutting fees and producer prices in the official price structure and by reorganizing the transport sector. (iii) To achieve policy objectives, the program relies heavily on the public sector with a high degree of Government involvement not only in supportive research and training but also in operational activities and a reliance on political rather than commercial messages in promotion campaigns.  8. nThe 1990 targets for household energy policy in the 1986-1990 Development Plan were to: (a) have two-thirds of all households using an improved fuelwood stove, (b) have 32,000 households using an LPG stove and to reach an annual LPG consumption of 4,800 tons, (c) market LPG and kerosene stoves for the urban population and (d) protect and regenerate the wood resources. Important lessons have been learned by this program, which is one of Sub-Saharan Africa's most ambitious efforts. Over 100,000 improved stoves were constructed by the rural population and public awareness of the importance of environmental conservation was raised. At the same time, however, the preservation of the forest cover is not an immediate concern of poor rural households and, as a consequence, they did not attach high priority to these energy saving measures. In addition, the relatively good access to fuelwood, as perceived by many women who collect it, provides little incentive to reduce consumption, further undermining the objectives of the program. The lackluster penetration of improved stoves in rural households is ample evidence; a 70% penetration rate of improved stoves drops to less than half of that level after the first year. 9. On the stove productIon side, the Burkinabe Institute for Energy (IBE) has made good progress in developing a wide array of different stove models for household and informal sector use. Nonetheless, efforts to mobilize local artisans to produce improved stoves have not had the desired results, due to excessive Government interference in distribution and marketing activities and modest demand for the product. In 1987, about one-third of all urban households were using an improved woodstove (50% Plan fulfillment), although it was but one of several cooking devices in use. Almost 60% of urban households continue to use the traditional 3-stone stove exclusively. The pace of sals of Improved woodstoves (6,000-7,000 per annum) since 1987 has been below that needed for etove replacement and to keep up with the growth of the urban population. The annual formation of about 12,000 new urban households, representing 23,000 additional tons of firewood per annum, wiped out the already meager fuelwood savings (15,000 tons per annum) claimed by the program. The response of very poor Burkinabe families to the improved stove program has been especially disappointing: Of households wit}l monthly incomes below 20,000 CFAF, only 18% had acquired an improved 3-stone stove and only 5% an improved metal stove. 10. Efforts on the supply side in favor of reforestation produced modest results. Industrial plantations have not proved cost-effective for fuetwood production. Similarly, organization and implementation of village afforestation schemes have been extremely difficult. Thorny issues including those of land tenure, traditional patterns of cultivation, village politics, internal migration trends, to mention but a few, further complicated an extremely complex issue. The introduction of techniques to manage natural forests and degraded woodlands promises to be the most cost effective means of promoting sustainable production methods. These efforts are, however, still in their early stages. Government has paved the way for this initiative by adopting a new price structure for fuelwood, which provides more adequate margins for fuelwood produceirs, and by encouraging the establishment of woodcutters' cooperatives. These cooperatives represent an important first step in rationalizing the fuelwood supply chain but sustainable management of forest and tree resources will require that these woodcutter?' cooperatives evolve into woodland management entities with clearly defined rights and obligations vis-a-vis land tenure; pricing; cost recovery; multi-purpose land uses, etc. A recent FAO-assisted pilot project in natural forest  - iv - management in the national forest at Nazinon, within the Ouagadougou fuelwood supply zone, is beginning to provide positive results. It wilt, hwever, require testing on a much larger scale and a careful evaluation of the incentive environment needed to mobilize across-the-board popular participation that will be essential to the success of this program. 11. On the positive side, pricing policy showed its strength. Following a reupnlzatlon of the fhelwood supply sre in 1985, fuelwood pries had climbed by 60% in Ouagadougou with household demand dropping to a level that was, on average, 9% lower than average fuelwood demand in the three other major cities. 12. The Government, through SONABHY, the national oil company, has made important investments in inErastructure by, inter setting up LPG bottling plants in Ouagadougou and in Bobo-Dioulasso. Notwithstanding these efforts, lnter-fuel substitution activities have been hampered by the high cost of moderns fuels, the inadequacy of stove equipment, the widespread skepticism about their safety and convenience, and lack of coordination between SONABHY, the importer and bottlingoperator; the private oil companies, responsible for distribution; and IBE, responsible for development of appropriate kerosene and LPG stoves. Both kerosene and LPG find their use mainly as supplementary sources of energy for cooking. In 1987, approximately 9% of households in the four major cities had acquired an LPG stove (one-third Plan fulfillment) and total demand for LP(G had increased from 745 tons in 1985 to 1,476 tons in 1987. Since then, demand has stagnated; LPG demand in Burkinabe households is no higher than 700 tons, equivalent to a replacement of no more than 4,000-5,000 tons of fuelwood per annum. In 1987, kerosene stoves were found in about 4% of urban households and urban household kerosene consumption amounted to approximately 3,500 tons, of which 94% was used for lighting, 2% for cooking and water heating, and 4% for other uses. Need for New Directions 13. It is difficult to identify any dynamic forces that, during the 1990s, could reverse the stagnation of Government's energy savings and fuelwood substitution program. The expected rise in per capita income is too low, by itself, to accelerate fuelwood substitution. Moreover, the standing volume of wood is still sufficiently large to meet the rising demand for fuelwood with little or no upward pressure on prices. Without a fine-tuning of current policies-shifting emphasis and expanding their scope, it is unlikely that progress will be achieved beyond present levels. Govemment's household energy strategy has the necessary components - promotion of fuel savings, fuelwood substitution, fuelwood supply management. Critical to the success of Government efforts over the longer term is an assessment of (a) the effectiveness of the mode of implementation of the components, (b) the adequacy of the legal and institutional framework and (c) the rationale of Government pricing policies. 14. Even with forestry management plans and with current population projections, a more widespread shift to substitute fuels in the urban centers will become a necessity early into the next century. At that time, urban fuelwood demand will have reached its sustainable plateau and LPG and kerosene will have to meet the household energy requirements of the total increase in urban population. To do so, LPG and kerosene consumers must use these fuels as their primary  .V. energy source for cooking and heating. During the 1990's, inter-fuel substitution initiatives must prepare for the future by acquainting households that do not use modem fuels with their advantages. The number of LPG and kerosene consuming households which, in the year 2000, will total approximately 115,000, is projected to increase to 650,000 by the year 2010. Depending on the contribution of kerosene, LPG household consumption could be in the range of 7,200-8,500 tons by the year 2000, and should reach somewhere between 60,000-115,000 tons by the year 2010. While it is recognized that the switch to substitute fuels will be expensive, the price of substitution, which is inevitable in light of current population trends, will be even higher if Government continues on its present course. Under any scenario, and given the country's development constraints, the reality is that, early in the next century, energy for all will be more expensive. Key Policy Recommendations and Conclusions of the ESMAP/MET Study Redefining Government's Role. 15. The high degree of direct Government involvement in the household energy sector during the 1980s helped to achieve some important short term results, although still far short of the program's objectives. Yet, the present stagnation demonstrates that long-term progress will depend on a redefmition of the balanc between the use of direct Government intervention and the use of policy tools that use the market mechanism to achieve the desired objectives. Two principles should shape this balance: (i) Government intervention is justified only where market failures block the achievement of program objectives; (ii) Where such circumstances exist, and where legal and/or institutional obstacles are the principal impediments, it should be determined whether their removal would be more cost-efficient than direct Government intervention. As a general rule, the Government should be responsible for guiding policy and shaping public awareness, supporting energy R&D, securing funding for information and education campaigns and monitoring progress. Private entities and NGO's should have primary responsibility for implementing the components of the strategy. Shift of Attention to Urban Fuelwood Needs 16. In the past, Government's efforts have been heavily anchored in promoting improved stoves in the rural areas; in 1987, emphasis began to shift to the urban sector. The justification for the rural improved stove program lies in the fact that present rural fuelwood consumption is eight times as high as urban fuelwood consumption and will still be four times as high in the year 2000. However, because the rural population tends to meet its fuelwood requirements from dead wood, the impact on the environment is not as large as that due to clearfelling for agricultural development. The rapidly growing urban sector poses a far more serious problem for the management of these natural resources and must become the prime focus of future efforts. While up to the year 2000, the rural sector will account for 60% of the increase in fuelwood demand, the pace of urban growth will change this scenario thereafter: towards 2010, the urban sector will account for almost 80% of the increase in fuetwood consumption. "Tus, while present efforts in the rural sector should be continued, any increase in offo should be directed towards the uLrban areas only.  - vi - 17. The concentrated urban fuelwood demand contributes directly to environmental degradation, since a rising share of this natural resource is mined exclusively for the urban market. Because these mined areas are not used for agricultural purposes and are unprotected from wind and rain, soil degradation is most severe and the regenerative capacity of the tree cover is lost. The impact of this demand is already visible around the larger urban centers where soil erosion and deforestation have affected their immediate hinterlands. Actions to restrain urban fuelwood demand must, therefore, receive clear priority and need to be directed, in the short term, to Ouagadougou on the central plateau, where the majority of urban residents live. Fuelwood Saving 18. The highest immediate operational priority is to reinforce the fuelwood savings program, by rationalizing the urban improved stoves program and by extending its scope to include changes in consumer behavior. The benefit-cost ratio of interventions m this area is substantial arnd some 20 to 25% of potential demand can be cut. To achieve an 80% urban penetration rate by the year 2000, the current approach needs to shift towards a self-sustaining, non-subsidized, private sector-based activity, that uses existing commercial channels for production and marketing. Tle changes are needed: (a) The MET should withdraw from direct operational involvement and discontinue public sector sales of improved stoves; (b) The METs role in stove promotion should be to: (i) identify and fund supportive actions in R&D in collaboration with the IBE, (ii) prepare general promotional material, (iii) assist in identifying and promoting *best practices" and (iv) monitor results. (c) The MET should establish targets for the level of stove penetration rates for each of the five major cities and secure financing for operations. Local NGO's with relevant experience should be identified for implementation. (d) The ad-hoc promotion campaigns should be replaced by year round promotion efforts. (e) Promotional pricing of stoves during sales campaigns should be abandoned. (f) A wider range of improved stove models should be marketed to capture more energy-conscious consumers, through the introduction of multi-pot stoves, a more vigorous promotion of ceramic stoves and of more expensive, attractive stoves stoves. (g) The needs of the lowest income groups for improved fuelwood stoves need special attention to secure a broader penetration rate. Such measures could include the continued promotion, even in the urban centers, of the "improved 3-stone" stove.  vii - (h) Production of improved stoves should continue to be based on the informal sector, since it is the lowest cost producer. To improve the incentive environment, producers should be allowed to set their own margins. (i) Door-to-door sales of improved stoves should be promoted. 19. The informal setor must become a priority target for energy saving campaigns because of (i) the high concentration of energy consumption in a few energy intensive activities, and (ii) the commercial orientation of the fuel users. Initially, action should be focused on the informal beer ("dolo") producers, the "dolotieres" who account for 12% of urban and 60% of informal sector fuelwood consumption. IBE has developed an improved stove for "dolo" production, which, according to test results, can reduce fuelwood consumption by 30%. Once user acceptance of the stove has been established in field tests, it should be produced and promoted by an intensive campaign aimed at an 80% penetration rate within three years. 20. Promotion of energy saing habit. The potential for energy savings through changes in cooking practices is as large as that achievable through the introduction of improved stoves - some 10 - 20% of domestic fuel consumption. The MET, in collaboration with the IBE and the Ministry of Social Affairs, should (a) implement information campaigns directed at households to promote energy saving habits, e.g. use of a lid; protection from wind; reducing the amount of fuelwood after lighting; etc. and (b) in collaboration with the Ministry of Agriculture and the Ministry of Social Affairs, promote new food habits that are healthy, provide market outlets for domestically produced crops, and are less energy intensive than the traditional Burkinabe cuisine. Fuelwood Substitution 21. The promotion of substitute fuels has the residual role of reducing the excess consumption that remains even where implementation of the proposed fuelwood saving and fuelwood supply management programs is deemed successful. Since electricity is not used for cooking, the increased consumption of electricity will have no impact on fuelwood demand. LPG and kerosene are, therefore, the prime substitute fuels and steps need to be taken to revitalize the LPG and kerosene promotion program. In particular, the following measures are needed: (a) improve the cooperation among the key actors; (b) increase the number of sales points for LPG; (c) introduce more technically adequate and socially acceptable stoves on the market; and (d) improve the price competitiveness of the substitute fuels. 22. The establishment of a meaningful cooperation between SONABHY (the operator of the bottling plant), the private oil companies (the wholesalers and retailers) and IBE (the stove developer) is the immediate priority for re-invigorating the fuelwood substitution program. Renewed cooperation must be based on two basic understandings: (a) effective promotion of LPG and of kerosene is achieved through the establishment of a wholesale and distribution network, and (b) wholesalers must have responsibility for its development. The present institutional framework and the price structures have to be reevaluated from this point of view.  eviUii 23. The present promotion system, whereby SONABHY administers a fund, replenished by a contribution in the LPG price structure, to finance oil company proposals for promotion campaigns is not working. The established organizational framework for LPG (inlIa_ the introduction of the 'unitary buttle concept") does not give the private oil companies incentives to implement promotion campaigns. Instead, SONABHY should take up direct responsibiity for the organization of promotion campaigns. 24. Recently developed prototypes for low-cost LPG stoves developed by the IBE should be aggressively market tested and promoted. In addition, the ongoing development work by [BE on the development of a low-cost kerosene stove, adapted to Burkinabe cooking requirements should be given high priority to broaden the substitution possibilities. 25. The LPG stove models on the Burkinabe market show considerable differences with regard to their conversion efficiency. IBE should make attempts to improve the eficiency of the low-performance stoves through changes in design. Inefficient stove models should be phased out from the commercial market in consultation with the private oil companies. Finally, introduction of an appropriate low-cost kerosene stove on the market should be considered as a high priority for the broadening of substitution policy. Pricing Policy 26. Since the Government avoids subsidies and tries to let the price of fuelwood reflect producer costs, consumers- by and large, are provided with the correct relative economic prices. Nonetheless, by the mid-1990s, Government will need to increase the real price of taxation on fuelwood to prepare for the large scale switch to petroleum products that will be required. Because LPG and kerosene are potential mass consumption products and could become a serious drain on Government revenues, subsidies on LPG and kerosene must be avoided. 27. The Government has been less successful in providing the suppliers with the correct signals. For some actors in the price structures, e.g. the wholesalers, the margins allocated to them do not correspond to the functions they should perform in promoting the attainment of the Government's policy goals, specifically, the creation of an effective distribution system. On the other hand the margins for the retailers are overly generous resulting in the retailing activity being retained by the wholesalers. The anomalies that are identified in the relevant chapters of this Report should be corrected. 28. Concerning petroleum products, the 'price stabilization fund" should protect the consumer against short-term price volatilities, but adjust to long-term changes in prices. At the present low level of international oil prices, the contributions to the fund have the character of taxation. However, if international oil prices rise again, there is a risk that the fund once more will become a drain on Government revenues as political pressures prevent a realignment of national prices to new realities. Since the state oil company SONABHY now has the monopoly for oil imports, the fund can be replaced by an annual realignment of the import-price element in the price structure in accordance with expectations concerning international price movements in the coming year. If prices turn out lower, SONABHY wili make an extra profit, if they turn out higher, SONABHY will incur a loss. Since in the end, profits and losses are borne by the Government budget, the result is similar to the operation of the stabilization fund. The consumer is protected  - Lx - against random short-term fluctuations in the price, but adjustment to long term changes is smoother, Fuelwood Supply Manaagement 29. Successful demand-side interventions could reduce the projected fuelwood consumption (about 900,000 tons) by 145,000 tons in the year 2000. Given projected demand trends, natural forest management will need to cover the major share of the increase in fuelwood demand, supplying over 313,000 tons to accomodate a tripling of urban demand between now and the year 2000. Whereas demand-side interventions dominated the previous Plan period and will continue to be important, future policy must, give greater emphasis to forestry management. Because commercial fuelwood reforestation schemes are not economically viable, there are two main avenues to increase supply: (a) closely coordinating energy policy and agricultural development policy to ensure that a maximum of the wood resources cleared for agricultural expansion are recovered for energy purposes and are not burned standing; and (b) increasing the productivity of the natural forests through the introduction of natural forest management. 30. Natural forest management is the most cost-effective way of increasing the sustainable production of fuelwood. The economics are less favourable than for the fuelwood savings program, but the magnitude of the intervention is substantial: Around 70% of the potential increase in urban household energy demand up to the year 2000 has to be covered by increases in the supply of woodfuels. To realize this increase in supply without negative environmental consequences, the rural population must be given a vested economic interest in the protection of the forest resources. The central issue in the management of the natural forestry resources is well- known - because of undefined land tenure rights: (a) no vested interest exist to protect the resource base against overexploitation and (b) under free market operations, the producer price of fuelwood will not reflect the full cost of replacement. This situation leads to overconsumption and underproduction of fuelwood simultaneously. 31. Government has taken a number of steps to pruvide the needed conditions to promote natural forest management. The Village Land Management Program is examining the incentive environment needed to mobilize the will and commitment of the rural population to protect its natural resources. Within this framework, the woodcutters' cooperatives must evolve towards woodland management entities and tenure arrangements clearly defined so that members not only cut the wood but manage the resource for the longer term. The present myriad of legal decrees, several of which are contradictory, seriously hinder prospects for putting sustainable forestry management plans in place. It is understood that Burkina's Environmental Action Plan is examining this issue with a view to revising the appropriate legal texts. The Government, with donor assistance, has begun to test alternative forestry management techniques and, under the ESMAP activity, has undertaken preliminary mapping of existing and potential areas of fuelwood supply within the supply zones of Ouagadougou, Bobo-Dioulasso, Ouahigouya, and Koudougou. Other measures, described in this Report, include: (a) Completing the detailed mapping of natural woodland areas within the urban supply zones, cataloguing their presention condition and evaluating their potential productivity under management conditions;  v % . (b) Preparing detailed management plans for areas identified for natural forest management, stipulating, inter-aia the technical parameters of the forestry plan; (c) Organizing the woodcutters' cooperatives into woodland management entities, and defining arrangements concerning land tenure, cost recovery, multi-purpose use of the designated management zones and other rights and obligations of the cooperatives and the Government needed to reach management targets; (d) Introducing a system of differentiated cutting permit fees for managed and unmanaged zones and for wood coming from agricultural land clearing operations; fuelwood coming from unmanaged zones will carry a higher fee as will fuelwood coming from areas closer to the city; (e) Creating an Urban Fuelwood Working Group, which would, inier a. liaise with the Ministry of Agriculture concerning fuelwood recovery on land clearing operations; (f) Undertaking pilot projects in the different urban supply zones to test management techniques and cooperative management arrangements; and (g) Putting in place monitoring arrangements to evaluate the results of the pilot efforts. A Proposed Urban Household Energ Investment Program 32. This Report sets out and describes a program of investment opportunities that would assist Government to meet the above-described program objectives. It is recommended that the program be considered as an integral part of the National Environment Action Plan. For the year 2000, this program would aim for: (a) 80% or 220,000 urban households to use improved biomass stoves to cover most of their cooking needs; (b) 100% of the informal sector beer brewers to use the Buriddo stove; (c) 41% or 115,000 urban households to use LPG and kerosene stoves to cover their auxillary cooking needs; and (d) at least 50% of commercial fuelwood to be produced by woodland management cooperatives. 33. Reaching the strategic objectives, will require Government to undertake (a) short- term actions at zero or little cost that will begin the process of policy realignment and (b) investment activities to implement the re-oriented urban household energy strategy with the help of external finance. These investment opportunities, which would reinforce on-going household energy programs, are set out in the table below:  I -xi Project Title Total Estimated WpleaontotIon Fundfng Cost in USO Period Source Fuel Consgrvtion Proara Urban Fuetwood Savings Project 1,500000 2 years To be identified Institutions and Training Needs Assessment 100,000 12 months To be idmntiffed Inter-Fuel Substitution Proarar LPG Infrastructure Strengthening Project 700.0Q0 12 months To be fdentified LPG Stove Developnent and Karketing Project 100,000 10 months To be fdentified Kerosene Promotion Options 40,000 6 wseks To be Identified Natural Woodtands NanmgeMent Ilproving inter-Agency Coordination workshop 45.000 6 weeks To be identiffed Village-based M g t of Old Fsllo* and Degraded Woodlands Project S,000,000 S yers To be identified 34. The strategy and investment program set out in this report are consistent with the energy and environmental recommendations of the Long Term Perspectives Study for Africa (LTPS). The LTPS highlights the need to address the household energy demand of the majority of the population of Sub-Saharan Africa that currently uses fuelwood and can be expected to do so over the longer term. It emphasizes the need both to broaden the use of improved fuelwood stoves as well as to encourage the shift to substitute fuels. The LTPS underscores the importance of identifying and testing different forestry management alternatives to protect the ecological base and to create an enabling environment that will encourage broadbased popular participation needed to reach this objective.  I. INTRODUCrHON 1.1 In April 1985, Burkina Faso launched a far-reaching forest protection program, "Les Trois Luttes", to reduce the impact of bush fires, uncontrolled grazing and overcutting of fuelwood on its natural environment. At the same time, the Government undertook a fuelwood conservation campaign anchoreed in a large-scale dissemination of improved stoves and the promotion of LPG as a fuel substitute. 1.2 The UNDP-World Bank Energy Assessment for Burkina Faso ('Burkina: Issues and Options in the Energy Sector/January 1986) affirmed Government's policy direction and highlighted the household sector as its largest energy user. While not the only nor the primary cause of deforestation, the demand for fuelwood was recognized as a source of mounting pressure on an already fragile ecological base. The Assessment cautioned that urbanization, in an overwhelmingly agricultural economy, was a factor in the depletion of standing wood stocks. The report stressed the need to develop an urban household energy management strategy to reinforce Government's on-going efforts in the rural areas. 1.3 At the request of the Government of Burkina Faso, and with financial assistance from the Government of Norway and later from the Government of the Netherlands, the UNDP- World Bank Energy Sector Management Assistance Program (ESMAP) undertook preparation of an urban household energy strategy and action program beginning in 1986. The objective was to design a strategy and companion investment program to meet the urban demand for energy at the lowest cost, while protecting the country's renewable resources and limiting imports of oil products and equipment to the extent possible. 1.4 The program was executed in two phases: Phase I, from October-December 1986, carried out an overview of the urban household energy situation; reviewed on-going household energy conservation and substitution strategies and projects, and prepared a detailed project document to define priority actions to be implemented under Phase II. Phase II fieldwork began in August, 1987 and was completed in December, 1989. The findings of Phase II constitute the subject of this Report.1/ Throughout the course of this study, ESMAP and its consultants worked closely with the Ministry of Environment and Tourism (MET) and other Burkinabe institutions. 1.5 The principal outputs of this activity include: (a) Creation of a Data Base an ernergy survey of over 1100 households in five major cities. (b) Demand Studies (i) pilot promotion campaigns of improved wood stoves in Ouahigouya and Koudougou; k/ The lpon was wniten by Carolyn Tager (Task Manager), with Wolfgang Mosten (Senior Ene&j Economist) and JosdpWneA#paillange (Consultan). Mr. SamirAmous (ESMAP Consultat) managed thefield wotv in Bwu*na Faso. Seeatial asistance and report prodwtion sevices were pvwded by Sophie Walop.  -2- (ii) testing of door-to-door sales of improved wood stoves; stove acceptability test in Ouagadougou. (c) Sdis (i) testing of stove efficiency (improved wood, kerosene and LPG stoves); (ii) a study of the production and marketing structure for improved stoves; (Uii) a household petroleum product distnbution study-, (iv) a forestry management stucdy and preparation of forestry management strategies for Burkina's 4 major urban centers.  - 3 - II. BACKGROUND A. Economic and Demographic Trends 2.1 With a GDP per capita of US$ 297 in 1987, Burkina Faso is one of the world's poorest countries, its development constrained by an array of natural obstacles. Agriculture is its most important economic activity, employing almost 90% of the population, accounting for one-third of GDP and virtually all of its exports. Yet, the natural resources on which agricultural output depends are meager and increasingly vulnerable to the effects of rapid population growth. Soils are generally shallow, poorly structured, and easily degraded by traditional methods of cultivation. Rainfall is scarce and highly variable, both among regions and from year to year. The country has located important mineral deposits, but its landlocked position and poorly maintained transportation network make the cost of their exploitation relatively high. 2.2 An annual population growth of approximately 3.3 percent is regarded by many development planners as the country's most serious challenge to promoting the sustainable use of its natural resources. The uneven distribution of this population is also a constraint and is felt in declining fallow periods, accelerating deforestation, and deteriorating soil fertility, especially on the central plateau. A small but rapidly increasing share of the total population lives in urban areas, accounting for approximately 11% of the total population of 9.8 million in 1989. The present urban growth rate of about 8% a year is more than double the overall population growth rate and shows no signs of slowing down. By the end of the century, total population is projected to be at least 12.5 million, with the urban share reaching 21% of the total, or 2.4 million. Ouagadougou, the capital and largest city, accounted for about 50% of the urban population in 1985 and its annual growth rate averaged 9.5% between 1975-1985. Bobo-Dioulasso, next with approximately 213,000 inhabitants, accounted for 25% of the urban population, growing annually at 7%. These two cities are expected to at least triple in size within 15 years, fueled by the high natural population growth and by uncontrolled internal migration. 23 The share of economic activity located in urban areas is striking given the still low level of urbanization. Over the period 1982-1987, an average 68% of GDP (constant 1979 prices) was accounted for by urban-based activities (28% by industrial and 40% by service-related activities), almost exactly the reverse of the situation in 1960. Despite notable strides in boosting agricultural output during 1982-1987 alongside slower growth in the manufacturing and tertiaty sectors, the importance of the urban sector to overall economic growth will not diminish. Indeed, as national development policy shifts towards creating conditions more favorable to private sector growth, improving public resource management and removing physical and institutional constraints, the economic role of the urban sector will be further enhanced. The downside of this scenario is an annual increase in urban residents of between 80,000 and 90,000 each year and the pressures of these numbers on already deteriorated living and working conditions. Tle impact of this explosive growth is visible in the hinterlands of the major urban centers, especially Ouagadougou, where soil degradation and erosion are the most severe, the combined effect of land dearing for agriculture and the accelerating demand for fuelwood for energy.  -4 - B. Energy and the Environment 2.4 The Government is deeply concerned about the accelerating depletion of its natural resources and the need to move both boldly and quickly. It is, therefore, not surprising that the inplementation of a program to reverse the ecological decline has ranked high on the country's development agenda since the early 1980's. This commitment was an explicit objective of the 1986- 1990 Five Year Plan, togther with (1) controlling population growth and migration, (2) achieving food security, (3) improving the management of the economy and (4) improving the role of women in development. A broad range of international and bilateral donors and NGOs endorsed these priorities through a series of programs that began as early as 1985. More recently, the Government appealed directly to IDA for assistance to elaborate a more comprehensive and coordinated national environmental plan and investment package. Energy Balance 2.5 Energy policy is a critical instrument of environmental management for Burkina Faso as fuelwood is the principal source of energy. The household sector is the most important fuelwood consumer, accounting for 3 million tons per annum and for 98% of household primary energy demand. Energy consumption by sector and fuel type is depicted in Table 2.1. Reflecting Burkina's low level of industrialization, the household sector (including informal sector productive activities) accounted for 88% of national final energy consumption in 1987. Transport followed with 6%, industry/agriculture 4% and the public sector 1%. Table 2.1: Final Energy Consumption by Sector, 1987 (000 of toe) Sector Biomass Oil Products Electricity TOTAL Households 1256 13 12 1281 Transport 85 85 Industry 39 12 21 72 Public Sector - 9 2 11 TOTAL 1295 119 35 1449 Source: ESMAP estimates. Electricity consumption reflects SONABEL sales only, as extent of autoproduction of electricity is unknown, although probably very minor. 2.6 Consumption of biomass fuels accounts for 89%, oil products for 8%, and electricity for 2% of final energy consumption. Biomass is consumed primarily in the form of fuelwood (91%); next is agricultural residues (5%); bagasse (3%) and charcoal (1%). It is important to note that, although the household sector is the dominant consumer of biomass fuels, biomass is the most prevalent energy source even in industry, where its consumption is three times as high as the consumption of oil products. 2.7 Burkina's overall energy consumption in 1987 of 1.45 million tons oil equivalent (toe) is one of the world's lowest, expressed in terms of per capita energy consumption (160 kg of oil  -5- equivalent); in terms of per capita consumption of oil products (17 kg); or in terms of per capita consumption of electricity (14 kWh). Yet even the satisfaction of this low level of demand for oil products represents a very real challenge to policymakers. In 1987, the cost of importing 130,000 tons of oil products 2/ amounted to US$ 47 million, equivalent to 16% of export income and to 6.5% of total imports of goods and services. Given these high costs and competing development priorities, the transition from biomass to modern fuels must be carefully planned. Indeed, given Burkina's well documented development constraints and prospects for future economic growth, the key energy issue facing the country is how to achieve an ecologically sustainable balance between fuelwood demand and fuelwood supply with minimal consequences on oil imports. Househol Me= Demand 2.8 While urban household energy demand is growing more rapidly, household energy consumption in Burkina Faso is dominated by the rural sector. Rural household energy consumption is about 8 times higher than that of the urban sector as shown in Table 2.2 below. Tabte 2.2: Rurat and Urban Household/Informl Sector Energy Consumption, 1987 #/ (ton, electricity a GkH) Fuetwood Charcoal Agr.Res. LPG Kerosene Electr. Rural Household 2,700,000 - 200,000 - 7,600 n.s. Urban Household 252,000 12,000 - 660 3,600 35 Urban Inf.Se. 41,000 2,000 - n.a. n.a. n.s. Total 2,993,000 14,000 200,000 660 t1,200 35 g/ Consumption of electricitya extrapolation of 1986 Sonabel figures Other consurption per capita per year: Rural consuwption of fuelwood * 365 kg; Urban household consumption a results from ESMAP/MET8s surveys of consuw,tion in the four major cities, other cities a 235 kg of fuelwood + 11 kg of charcoal; urban informal sector a extrapolation to all cities of ESMAP Ouegadouou surveys. If it is assumed that the factors influencing household energy consumption are held constant over the next decade, that annual population growth is 3.3% and urban population growth is 8%, rural biomass consumption will amount to 3.8 million tons fuelwood equivalent 2/ and urban consumption of fuelwood to 0.9 million tons in the year 2000. 4/ After the year 2000, however, the picture 2/ 1987 consumpdon was higher 153,00) tons. i/ With an estimated eneiy content of Z500 Kcal/k& the consmtion of 200,0 rtons of agcultwni residues in 1987 and 250,000 tons in the year 2000 comsponds to a "saving of 130,00 tons and 160,000 tons of fuelwood respectively. Yi C!amoal is denvedfronm two sources (a) as a residue fomm the buning of fuelwood (mostly supplied frm the infonnal setor beer bmwe) and (b) fivm the commercial carbonihzaon of fuelwood. If it is assumed that 15% of the enesgy content of uwd fuiewood in infommat sector beer poduction is retained n marketed Wsidue charoat, the 1987 suppty fim this sowu amounted to 2500 tons. In the year 2000, the supy would amount to 6,0A0 tons (increase equivalent to wban popution gowth). Based on these assump 'qns, conmea pvducwdi of chamcoal can thus be esimated at aund 8,a0o tons in 1987 and would Ke to about 26,000 tons in the year 20ltO At a carbonization rate of 20o, this tmnslates into a use of 40,000 tons of fuelwood in pumary poducion in 1987 and 130,000 tons in the year 2000.  -6- changes under the impact of continued high urban population growth: about 80% of the potential increase in fuehvood demand between 2000 and 2010 will come from the urban sector (see Annex I). 2.9 As depicted in Table 2.3 below, the structure of fuelwood consumption for cooking in Burkina's major urban centers reflects the population's rural roots, its low level of income, and the high cost of modern fuels. Only a minority of urban households uses modern fuels to cover some or all of their energy needs. Table 2.3: Fuel Use for Cooking (in Urban Population) Fuetwood 91X Charcoal 64X LPG 7X Kerosene a/ 3X Vegetal residues 1X a/ Use for lighting not included Source: ESIAP 2.10 The 1987 ESMAP/MET household energy survey showed that households cook an average of 1.5 hot meals per day and that this use accounts for about 3/4's of urban household energy demand in Ouagadougou as well as in Ouahigouya (the most rural of the four largest cities). The structure of household energy use in Ouagadougou is shown in Figure 2.1: Figure 2.1: Ouagadougou Household Fuelwood Consumption by End-Use, 1987 R2EL.O Ct ONSWT I ON BY ENO USE j 5 ok P." Sorce: P T H Source: ESHAP/MET Househotd Energy Survey 1987  -7- 2.11 Urban households consume on average 644 grams of fuelwood and 30 grams of charcoal per capita/day, or a total of 693 grams per capita/day of fuelwood equivalents (calorific basis). Sasonal differences in consumption between the cold (December to February) and the warm period (the remainder of the year) are relatively modest, varying between 3% in Ouahigouya and 13% in Bobo-Dioulasso. Regional differences in fuelwood and charcoal consumption are higher: Consumption is lowest in Ouagadougou with 658 grams/per capita/day and highest (+23%) in Ouahigouya with 838 grams of fuelwood equivalents per person and per day. Preparation of medicine uses less energy in Ouagadougou than in Ouahigouya, because of a higher consumption of modern forms of medicine. The relatively high availability of fuelwood and the possibility for self-collection are other factors that induce a high consumption of fuelwood in Ouahigouya. 2.12 Factors influencingnconsumption. The three important socio-economic factors that influence energy consumption are household size, income elasticity and price elasticity. The implications of these factors are analyzed in Annex II. Because of economies of scale in energy consumption, smaller households have higher per capita consumption of household energy than larger households. Thus, a family of four will typically consume twice as much as a family of 13. Household energy consumption increases with income, but the effect in the case of fuels other than electricity is modest, owing to the basic needs character of household energy demand. Similarly, the price elasticity of household energy demand is low: a doubling of price will only decrease demand by about 15%. 2.13 LPG and kerosene consumption. LPG consumption is entirely an urban phenomenon, limited to 9% of urban households who use it principaHly as a secondary fuel. Consumption expanded during the mid-1980's, but has since stagnated because most of the households in the 'captive market"--the upper income groups-have already switched to LPG. Kerosene is widely used in both urban and rural households, but its use is limited almost exclusively to lighting. 2.14 LPG-consuming households differ from the average fuelwood-consuming households in three aspects: these households are richer; they are smaller (see Table 2.4) and have a preference for meals that are less energy intensive than the traditionai Burkinabe meal: Whereas 1t,", (Burkina's food staple made of millet and sauce) makes up 60% of the daily diet in fuelwood- consuming households, it represents only 12% in LPG-consuming households. Per capita consumption of household energy in LPG-consuming households is, as a consequence, lower than in other households, although the opposite would have been predicted on the basis of their size and income.