Uganda - Issues and options in the energy sector

Report No. 4453-UG Uganda: Issues and Options in the Energy Sector July 1983 Report of the Joint UNDP/World Bank Energy Sector Assessment Program This document has a restricted distribution. Its contents may not be disclosed without authorization from the Government. the UNDP or the World Bank.  JOINT UNDIP/WIORLD TANK EETNER-Y SECTOR ASSESSM4ENT PFRODGsRKL- REPORTS ALREADY ISSUED Country D-te No- Indo77esia Nom$e 1381 35433-IND Mauri t 11us Deceer 1.93 I 3510-MAS Kenya May 38 0 0 -KE Sri Lanka Nay 3323794-CE Zmbabe _r,e 37665-ZTM Haiti June 1.3,32 3672-HA Papua Newv,7 uu:nea June 19 8 2 3882-PNG Burundi 3une l32 3778-BU Rwanda June 1982 3779-RW Malawi Au-gust 1.8 32 3903-:HAL Bangladesh October 1982 38 73-BD Zambia January 1383 4110-ZA Turkey March I-383 3877-TU Bolivia April 1933 4213-BO Fiji June 1983 4462-FIJ Solomon islands June 1 383 4404-SOL Senegal July 1983 4182-SEL  FOR OFFICIAL USE ONLY Report No. 4453-UG UGANDA ISSUES AND OPTIONS IN THE ENERGY SECTOR July 1983 This is one of a series of reports of the Joint UNDP/World Bank Energy Sector Assessment Program. Finance for this work has been provided, in part, by the UNDP Energy Account, and the work has been carried out by the World Bank. This report has a restricted distribution. Its contents may not be disclosed without authorization from the Government, the UNDP or the World Bank.   TITLE : UGANDA: ISSUES AND OPTIONS IN THE ENERGY SECTOR COUNTRY : UGANDA REGION : EASTERN AFRICA REPORT TYPE CLASSIFICATION MM/YY LANGUAGE 4453-UG EA Official Use 07/83 English ABSTRACT: In Uganda, per capita consumption of energy, especially commer- cial energy, is extremely low at the present time. However, there has been a recovery in economic performance during the past two years and, if this is sustained, it will lead to a significant increase in energy demand over the remainder of this decade. Equally, the availability of energy will be an impor- tant determinant of the pace at which economic recovery can proceed. This Report focuses on the implications of this out- look for energy supply options, pricing policies, institutions and manpower, investment priorities and resource requirements. Major recommendations of the Report relate to: (a) implementation of a number of relatively simple and cheap conservation measures in the industrial and transport sectors; (b) rehabilitation of Owen Falls Power Station and the prepara- tion of a medium-term power development program; (c) reductions in supply costs for petroleum imports through diversification of supply sources, utilization of alterna- tive transport modes and routes, and stocking policies; (d) preparation of a petroleum exploration promotion package to determine the interest of oil companies in further exploration and development; (e) alleviation of growing woodfuel shortages through programs for agro-forestry, urban and industrial plantations, and forest management; (f) price adjustments for electricity, petroleum and woodfuels, in order to encourage energy efficiency and finance future investments in the sector; (g) a phased program of institutional development, including establishment of a new Energy Department in MPPT and an Inter-Ministerial Policy Committee; and (h) consideration of an investment program for the energy sector of close to US$400 million over the next eight years.   ABBREVIATIONS AND ACRONYMS Weights and Measures GWh gigawatt hour ha hectare kg kilogram km kilometer kV kilovolt kVA kilovolt-ampere kWh kilowatt hour 1 liter m3 cubic meter m3/s cubic meter per second MVA megavolt-ampere MW megawatt TOE ton of oil equivalent Organizations GSMD Geological Survey and Mines Department MPED Ministry of Planning and Economic Development MPPT Ministry of Power, Posts and Telecommunications NTC National Tobacco Corporation UCTU Uganda Cooperative Transport Union UEB Uganda Electricity Board URC Uganda Railways Corporation UTC Uganda Transport Company EAC East African Community EEC European Economic Community EPD Economic Analysis and Projections Department (World Bank) IDA International Development Agency (World Bank) ODA Overseas Development Administration (United Kingdom) UNDP United Nations Development Program This Report is based on the findings of an energy assessment mission which visited Uganda in October-November 1982. The mission members were: Mark Baird (Mission Leader/Macro-Economist), John Besant-Jones (Energy Economist), Argun Ceyhan (Power Engineer), Kurt Loos (Industry/ Petroleum Specialist), Turgut Ogmen (Petroleum Marketing Specialist), Keith Openshaw (Woodfuel/Renewable Energy Specialist), and Brian Robinson (Efficiency Specialist). Julian Bharier, Chief of the UNDP/World Bank Energy Assessment Division, joined the mission for the wrap-up meetings. A draft of the Report was reviewed with the Government in June 1983. The principal authors of the Report are John Besant-Jones and Mark Baird.  CURRENCY AND FUEL EQUIVALENTS Currency Since August 1982, Uganda has operated a dual exchange rate system. The prevailing rates at the end of 1982 were: USh 100 = US$1 at window one USh 240 = US$1 at window two. For economic analysis, a shadow exchange rate of USh 200 = US$1 is used in this Report. The black market rate, relevant for unofficial trade, is estimated to be about USh 300 US$1 at the present time. Major FueLs a/ TOE; b/ m3 per metric ton per metric ton Petroleum products Aviation fuel 1.01 1.29 Gasoline 1.03 1.33 Kerosene 1.01 1.22 Auto. diesel 0.995 1.18 Industrial diesel 0.995 1.09 Fuel oil 0.956 1.06 L. P. gas 1.07 Woodfuels Fuelwood (air-dried) 0.375 1.43 c/ Charcoal 0.773 kWh per TOE Electricity Thermal-equivalent (supply) 4,000 Hydro (end use) 12,000 a/ Conversion factors are approximations only. b/ One ton of oil equivalent (TOE) = 10.2 million kcal = 42.7 million kilojoules. c/ Solid.  Table of Contents Page No. SUMMARY AND CONCLUSIONS ...................... i I. ENERGY IN THE ECONOMY: AN OVERVIEW ........................... 1 A. The Ugandan Economy ....................................... 1 B. The Energy Sector ......................................... 5 II. ENERGY DEMAND: OUTLOOK FOR THE 1980s ......................... 25 A. Scope for Energy Conservation ............................. 25 B. Scope for Energy Substitution ............................. 33 C. Economic Projections and Energy Demand .................... 37 D. Major Recommendations on Energy Efficiency ................ 45 III. ENERGY SUPPLY: OPTIONS FOR DEVELOPMENT ....................... 49 A. Hydroelectric Resources ................................... 49 B. Petroleum Products ........................................ 54 C. Fuelwood and Charcoal ..................................... 62 D. Other Indigenous Energy Sources ........................... 70 E. Major Recommendations on Supply Options ................... 73 IV. ENERGY PRICING POLICIES ....................................... 75 A. Electricity Tariffs ....................................... 75 B. Petroleum Prices .......................................... 80 C. Woodfuel Prices ........................................... 89 D. End-Use Energy Costs ...................................... 91 E. Major Recommendations on Pricing Policies ................. 94 V. INSTITUTIONS AND MANPOWER ..................................... 95 A. Sector Coordination and Planning .......................... 96 B. Subsector Management and Development ...................... 101 C. Training and Technical Assistance ......................... 106 D. Major Recommendations on Institutions and Manpower ........ 109 VI. INVESTMENT PRIORITIES AND RESOURCE REQUIREMENTS ............... 111 A. The Government's Plans .................................... 111 B. Investment Priorities ..................................... 114 C. External Financing ........................................ 118 D. Major Recommendations on Investment Priorities and Resource Recommendations ................................ 121  Annexes Page No. I. Basic Data on the Energy Sector in Ugan,da .... ............. 125 II. Projections: Approach and Assumptions .................... 134 III. Industrial Specific Energy Consumption .................... 142 IV. Hydroelectric Potential and Development ................... 143 V. Derivation of Petroleum Import Costs ...................... 152 VI. Production and Revenue from a Commercial Firewood Plantation ....................................... 159 VII. Electricity Tariffs in Uganda ............................. 160 VIII. Average Cost of Power from New Hydroelectric Capacity ..... 162 IX. Proposed Terms of Reference for Power Studies .... .......... 163 X. Proposed Terms of Reference for Short-Term Technical Assistance in Energy Efficiency ........................... 170 Map IBRD 16933R Uganda: Power System Developments as of December 31, 1982  Tables in the Main Text Page No. Chapter I 1.1 Economic Indicators .................................. 3 1.2 Trends in Energy Consumption ......................... 6 1.3 Retail Energy Prices .................... 14 1.4 Energy and the Balance of Payments ..... .............. 17 1.5 Estimated Energy Balance for 1970 ..... ............... 22 1.6 Estimated Energy Balance for 1980 ..... ............... 23 Chapter II 2.1 Projected Trends in Energy Consumption .... ........... 41 2.2 Alternative Projections of Energy Consumption ........ 42 2.3 Projected Energy Balance for 1985 ..... ............... 46 2.4 Projected Energy Balance for 1990 ..... ............... 47 2.5 Illustrative Energy Balance for 2000 .... ............. 48 Chapter III 3.1 Major Sites with Hydroelectric Potential .... ......... 50 3.2 Cost of Petroleum Products from Various Supply Sources 55 3.3 Transport Costs for Petroleum Products Through Kenya 58 3.4 Costs for Petroleum Products Supplied Through Kenya and Tanzania ....... ............................. 60 3.5 Economic Returns to Fuelwood Plantation .... .......... 68 3.6 Economic Value of Fuelwood ........................... 68 Chapter IV 4.1 Indicative Estimates of Long-Run Marginal Cost of Power in Uganda .......... ........................ 78 4.2 Retail Price Formula for Petroleum Products .... ....... 81 4.3 Net Taxation on Petroleum Products ..... ............... 83 4.4 Official Selling Prices of Crude ...... ................ 85 4.5 Oil Company Margins in 1982 ....... .................... 85 4.6 Import-Parity Prices for Petroleum Products .... ....... 87 4.7 Import-Parity and Retail Prices for Petroleum Products 88 4.8 Costs for Fuelwood and Charcoal Marketed in Kampala ... 90 4.9 End-Use Energy Costs in Urban Areas .................. . 92  Page No. Chapter V 5.1 MPPT's Proposal for an Energy Department .... .......... 98 5.2 Mission's Proposal for Future Organization of the Energy Sector ........ ............................. 99 5.3 Immediate Technical Assistance Priorities .... ......... 107 Chapter VI 6.1 Government's Recovery Progrem for the Energy Sector ... 112 6.2 Public Investment Priorities for the Energy Sector .... 115 6.3 External Assistance for the Energy Sector .... ......... 119  SUMMARY AND CONCLUSIONS Introduction i. During 1980, per capita energy consumption in Uganda is estimated to have been 0.3'i TOE, of which only 0.06 TOE was commercial. This level of commercial energy consumption, while exceptionally low by world stand- ards, is comparable to estimates for some other low-income countries in Sub-Saharan Africa. In Uganda's case, the low level of commercial energy consumption reflects not only the country's low per capita income, but also the dominance of the subsistence sector and the significant decline in the industrial and transport sectors during the 1970s. For the same reasons, energy consumption is concentrated in the household sector and supplied primarily from woodfuels. Electricity, which is the major focus of public involvement in the energy sector, has never supplied more than 4% of com- mercial energy consumption. In terms of primary supply, 84% of Uganda's commercial energy was domestically produced in 1980, with the balance coming from imported petroleum products. About 7% of the net supply of commercial energy was consumed outside Uganda, including electricity exports to Kenya and unofficial sales (smuggling) of petroleum products to most neighboring countries. ii. Since mid 1981, the Government has introduced a number of major policy reforms, including devaluation of the Ugandan shilling and related price adjustments. These measures have contributed towards a recovery in economic performance over the past two years. The impact of this recovery on commercial energy consumption has so far been constrained by higher energy prices and the shortage of foreign exchange with which to import petroleum products. As a result, the declining trend of commercial energy consumption, which has been evident since 1970, continued through 1982. However, even allowing for further improvements in energy efficiency, this trend is now expected to be reversed. Given the many uncertainties in the Ugandan situation, a range of illustrative energy demand projections are presented in Chapter II of this Report. Under the base case, commercial energy demand rises by 6.3% per annum through 1985 and 4.7% per annum from 1985 to 1990. Even with more pessimistic assumptions on the pace of reco- very (the low case), the growth rate of commercial energy demand averages 4.8% per annum over the remainder of the 1980s. To meet this demand is the basic challenge facing the energy sector today. This Report focusses on the major issues this raises for sector planning and policies. The Mission's findings and conclusions are summarized below. Brief summaries of the Mission's major recommendations are also provided at the end of Chapters II through VI. Energy Efficiency iii. Conservation. The efficiency of energy use in Uganda is extreme- ly low. Some degree of improvement in energy efficiency will follow auto- matically, as rehabilitation proceeds and capacity utilization increases. However, additional savings should also be achieved by implementing rela- tively simple and cheap energy conservation measures, especially in the industrial and transport sectors. These conservation measures would have an immediate impact on the energy situation, by constraining the growth of  -~~~ ii petroleum import requirements and the demand oni generation capacity at Owen Falls Power Station. InI the industrial sector, the Mission recommends that: (a) fuel-firing techniques in boilers and furnaces should be improved through operator training and provision of appropriate monitoring instruments (paragraph 2.6); (b) power-factor correction equipment should be installed and proper- ly utilized (paragraph 2.7); (c) various investments in energy-saving equipment should be under- taken, including recuperators on furnaces and kilns, effluent heat recovery plant and recovery of condensate (paragraph 2.8); (d) new and more efficient boilers, already imported, should be installed as soon as possible (paragraph 2.8); and (e) general energy management and housekeeping techniques should be improved (paragraph 2.9). Similar energy savings can be made in agro-industrial plants, through im- proving boiler efficiency in tea-drying factories (paragraph 2.10), improv- ing kiln design and burning efficiency in the brick and kiln industry (paragraph 2.11), improving barn design and increasing barn size for tobac- co curing (paragraph 2.13), and possibly using brick hearths for fish smok- ing (paragraph 2.14). iv. The Mission estimates that diesel fuel consumption in rail and road transport could be reduced by 5% to 10% with appropriate conservation measures. These include: (a) better maintenance procedures, workshop facilities and availabil- ity of spare parts; (b) installation of diesel-testing equipment (some of which has al- ready been imported) and proper tuning of engines; and (c) training of drivers and mechanics (paragraphs 2.16 and 2.18). Improvements in railway track conditions and grades, as well as road reha- bilitation, would also contribute towards better fuel efficiency in the transport sector (paragraphs 2.14 and 2.1.8). v. Energy efficiency in households and commercial establishments, especially for woodfuel stoves, is also low. The most common stoves at present are the fuelwood-burning three-stone i-ire and the charcoal-burning metal stove (sigiri). Higher efficiencies could possibly be achieved with burnt-clay stoves. However, for these stoves to be acceptable to the con- sumer, they must be portable, easy to usea and low cost. Particular atten- tion needs to be given to the development of simple and non-customized  - iii - designs which utilize local materials, and to the practical problems of stove manufacture and marketing (paragraph 2.21). A number of countries in the region have recently started programs to improve stove efficiency and Uganda should follow these programs closely. To date, progress has been limited and even the potential for achieving higher levels of stove effi- ciency in everyday use remains unclear. Nevertheless, further development and promotion of better stove designs is justified by the potential longer- term impact on energy demand, especially for woodfuels (paragraph 2.20). Further savings of wood raw material could be realized by improved charcoal production techniques. Courses should be organized in earth kiln produc- tion, and brick and metal kiln production should be revived and applied where appropriate (paragraph 2.23). Alternative fuel options should also be investigated, including briquetting of charcoal powder/fines, sawdust and crop residues, and compressed and pre-dried wood (paragrah 2.21). vi. Present losses on power transmission and distribution offer con- siderable scope for energy conservation. Urgent rehabilitation works are scheduled with ODA and IDA assistance. However, additional rehabilitation will be required to reduce distribution losses to acceptable levels, and the extent of such works are included in the terms of reference for the proposed study into uprating and rehabilitating Owen Falls Power Station. Improvements in power factors, especially for major industrial consumers, will also help to reduce distribution losses (paragraph 2.25). vii. Substitution. The feasibility and economic viability of propos- als for fuel substitution in industry should be studied on a case-by-case basis, taking into account projected production levels and the associated costs of energy investments. The Mission's preliminary findings suggest that there are several industrial and agro-industrial plants in Uganda for which there is a strong economic justification, as well as financial incen- tive, for fuel substitution. The most promising option is the substitution of fuel oil by woodfuels, for example in the cement industry and for tea drying and coffee roasting. However, a basic prerequisite would be a guar- anteed source of woodfuel supply, either from the establishment of planta- tions or designation of specific forest areas for commercial management (paragraph 2.26). Other promising forms of energy substitution include switching to heavier grades of fuel oil (paragraph 2.30) and the use of in-house agricultural wastes such as bagasse and coffee husks (paragraph 2.31). In general, the Mission doubts that major boiler conversions to electricity would be economically justified before the addition of power generation capacity from a new hydro station, which would be in 1991 at the earliest (paragraph 2.32). Similarly, electrification of the Ugandan rail- way system is unlikely to be justified in the near future because of low traffic densities (paragraph 2.26). However, the Government's plans to extend the power transmission network to isolated local centers, presently supplied by diesel operators or which are without electricity supply, are considered justified. This extension will substantially reduce the costs of supply and provide an additional energy option for households, agro- industries and water pumping stations (paragraph 2.34).  - iv - Energy Supply Options viii. Hydroelectricity. Uganda is well endowed with hydroelectric potential, especially along the Victoria Nile. According to the base-case demand projection, additional generating capacity will be required by 1986 to meet Ugandan demand and existing export contracts. The only capacity which could be added in time to avoid significant shortfalls in supply is the rehabilitation and uprating of Owen Falls Station, which could add about 60 MW in rated capacity. Already, the timetable for project prepara- tion and implementation is tight. It is therefore essential that the pro- posed feasibility study is started immed:Lately (paragraph 3.6). The next addition to capacity to meet projected demand will be required towards the end of the decade. Since the period required for project preparation will be about two years and for construction about five years, planning for the next station must be started now. In this context, an immediate priority is preparation of a least-cost power development program up to the year 2000. The program should be based on a detailed demand forecast taking into account the prospects for economic recovery, fuel substitution possi- bilities, additional exports and extension of the transmission system (paragraph 3.7). Proposed terms of reference for these two studies were drafted and discussed by the Mission with MPED, UEB and ODA (Annex IX). ODA has subsequently agreed to finance these two studies. ix. Petroleum. At the present time, Uganda imports all of its re- quirements of petroleum products. There are signs that Uganda could have some petroleum resources in the Lake Albert area and further work to assess the extent and economic viability of these resources is justified. Techni- cal assistance is already being provided by CFTC and IDA to help prepare petroleum legislation, undertake airmag and gravity surveys (jointly with other countries in the region) and complete a petroleum exploration pack- age. This package can then be used to determine the interest of oil compa- nies in further exploration and developmernt (paragraphs 3.26 to 3.29). x. Development of domestic petroleum resources is a long-term and still uncertain prospect. At least for the 1980s, Uganda will remain dependent on imported petroleum products and the main objective should be to minimize the costs, particularly in foreign exchange, of these sup- plies. Significant progress in this regard has already been made during 1983. Even so, the Government still needs to take a more active role in monitoring petroleum imports, based on an informed understanding of chang- ing market and regional conditions, to ensure that an appropriate mix of supply options is maintained (paragraph 3.12). The Mission's recommenda- tions are as follows: (a) diversification of supply sources. The cheapest source of petro- leum products for Uganda is direct imports from the Middle East. The next best option is usually processing crude through the Mombasa Refinery, although purchases on the Kenyan market occa- sionally become more economic (paragraph 3.13). During 1983, the oil companies have stopped buying products on the Kenyan market  -v - and some have started importing directly from the Middle East. This shift is proving beneficial to Uganda. Further adjustments in supply sources should be made, as and when appropriate, to take advantage of changing cost relationships (paragraph 3.16); (b) reduction in transport costs to the Ugandan border. Presently less than 10% of petroleum products are transported from Mombasa to the Ugandan border by rail. However, this is by far the most economic option and the Ugandan Government should do all it can to ensure that better use is made of the railway. This will require the allocation of rolling stock for transporting petro- leum products in Kenya and Uganda, and some rehabilitation of railway infrastructure. But the more binding constraint is the influence of trucking vested interests on the allocation of freight traffic, and this will be difficult to break (paragraph 3.20); (c) development of alternative supply routes. The Government is pre- sently arranging with Shell to carry out a trial run for import- ing petroleum products through Dar es Salaam and on by railroad to Mwanza and thence by lake ferry to Jinja. It would appear that the costs of this route are similar to those of using road transport through Kenya. The Mission supports this initiative (paragraph 3.23); and (d) maintenance of a product stockpile. Uganda should maintain a "minimum safety stockpile" of petroleum products (paragraph 3.24). There would appear to be no general shortage of storage capacity at the present time. The actual level of stocks has also been increased in recent months from one to four weeks' consumption. This level of stocks would seem to be adequate to provide protection against a sudden disruption in supplies (paragraph :3.25). xi. Woodfuels. There is evidence that Uganda's wood capital is being eroded. In certain areas, where population densities are high or the available land is being converted to agricultural uses, the accessibility of woodfuels has already been seriously affected and there is a danger that local wood supplies will soon be depleted. The Mission therefore recom- mends that the Forestry Department undertakes an inventory of tree stocks on a district-by district basis. This inventory should be complemented by a survey of consumption trends and projections for all wood products to help pinpoint the areas of immediate shortages. Measures can then be pur- sued to help remedy this situation (paragraph 3.32). In particular, the Government should: (a) encourage farmers to introduce trees into the cropping system. Appropriate agro-forestry practices should be taught in agricul- tural and forestry colleges and at Makerere University. Exten- sion workers should be trained in agro-forestry and farm tree  -- vi - management so that they can provide practical advice to farmers. At present, the extension system in Uganda is basically non- operational and it will have to be substantially strengthened and reorganized to become effective (paragraph 3.35); (b) assist in the planning aLnd execution of urban and peri-urban plantations. Legislation may be required to permit plantation development, if there are not already forest reserves in the vicinity. Private individLuals may invest in woodfuel plantations as the most profitable use of land at present urban prices for fuelwood and charcoal. The Government could encourage this in- vestment by providing inputs (for a charge) and management assistance (paragraph 3.37); (c) assist industries to establish industrial plantations and wood- lots as an alternative energy source. This could be done by pro- viding technical help and leasirg out areas of forest reserves as has been done in the case of the tobaczco industry. If necessary, the Government could also provide inputs and financial support (paragraph 3.38); (d) expand forest plantations to provide the raw material for sawn- wood and panel products and, most importantly, for woodfuels and poles. The output per unit area could be increased up to five- fold by converting from natural forests to plantations. This should at least be done around the perimeters, to prevent en- croachment (paragraph 3.39); (e) improve the management of natural forests and woodlands, by cutting out unwanted trees for charcoal production or fuelwood and controlling the legal and illegal use of commercially valua- ble trees. Line planting should be reintroduced to boost natural regeneration and more forest arteas should be included within the management plans (paragraph 3.4C); and (f) give special emphasis to the needs of low-potential agricultural areas, such as Karamoja, where there is an acute shortage of wood and ecological conditions are fragile. It is essential to estab- lish a protective barrier of trees to prevent desert creep and with browse or feed trees it may even be possible to sustain a higher animal populatiorn as weLl (paragraph 3.41). The estab- lishment of trees is the most difficult problem and micro- catchment techniques with solar elect:rical fences should be tried (paragraph 3.42). Sufficient seeds, cuttings and seedlings must be available in the right place at the right time to undertake an expanded planting program as pro- posed above. Related back-up requirements include trials on tree species, the acquisition and distribution of indigenous and exotic tree seeds, the establishment of seed stores and nurseries, and the development of seed orchards and a tree-breeding program (paragraph 3.43).  - vii - xii. Other indigenous energy sources. In western Uganda there are a number of areas which show promising geothermal potential. A UNDP study, completed in 1971, concluded that further surveys and exploration of this potential was justified. However, the Mission does not consider this an immediate priority, given the present resource constraints and the avail- ability of sufficient hydroelectric capacity to meet the country's power requirements for many years to came (paragraph 3.50). The Mission has also looked at new renewable energy options and concludes that these are unlike- ly to have a significant impact on Uganda's overall energy situation during the 1980s or even over the longer term. However, some could make a contri- bution in specific applications (e.g., solar drying and water heating), in certain industries (e.g., use of bagasse by the sugar industry) and in isolated areas (e.g., biogas). Further development of these renewable energy options is justified. The Government should consider providing special tax incentives to encourage use of solar heaters in new and exist- ing buildings, especially in areas not served by UEB's transmission network (paragraphs 3.51 to 3.58). Energy Prices and Pricing Policy xiii. Retail prices for all energy sources in Uganda declined in real terms during the 1970s. This led to waste and inefficient use, reduced the sector's contribution to public revenues, and encouraged smuggling of petroleum products to neighboring countries. Since 1981, the Government has increased petroleum prices substantially, at least in line with the devaluation of the Ugandan shilling and by as much as 2,200% in the case of kerosene. However, the potential for profitable smuggling has not been fully eliminated and there are still significant price distortions between products. Comparable adjustments have not been made in domestic electri- city tariffs and they are now well below the long-run marginal cost of supply. Woodfuel prices, which are not officially controlled, have risen (in real terms) over the past two years, reflecting the increased prices of alternative fuels, higher transport costs, security bottlenecks and the steady depletion of the most accessible and economic forest resources. xiv. The estimated economic costs of available energy sources for selected uses are given in Table 4.9. Electricity produced from hydro- capacity is the least-cost form of energy for lighting in urban areas. In other uses, the comparisons between electricity and woodfuels are very much dependent upon the economic value of land used for wood production and transport and distribution costs. For the major population centers in southern Uganda, where the surrounding land is suitable for production of medium-value crops and where supplies from more-distant low-value land involve substantial transport costs, electricity probably remains the most economic energy source for household cooking. For urban areas in the northern part of the country, and in most rural areas, woodfuels would become a more attractive option. Similarly, woodfuels are also more attractive in non-household uses, where transport costs are generally lower and the end-use conversion efficiency higher than in household use. How- ever, in all uses, both electricity and woodfuels are substantially more economic than petroleum products.  - viii - xv. Electricity tariffs. Uganda's present electricity tariff struc- ture and levels are based on historical factors, and do not adequately re- flect the costs of installing and operating additional generating capacity to meet projected demand. Domestic tariffs have been increased by 240% over the past three years in order to maintain the financial viability of UEB. However, a more general review of tariffs according to economic prin- ciples is required and should be undertaken immediately after the prepara- tion of the long-term power development: program (paragraph 4.2). The Mission estimates that the long-run marginal cost of power in Uganda is at least US cents 5/kWh, or USh 5 to 10/kWh depending on the exchange rate used. By comparison, the average yield from the present tariffs during 1983 will be about USh l/kWh. The Government. should raise tariffs to a level that corresponds to the economic cost of power as soon as is practic- able, and at least by 1991 when the next hydroelectric station will be required. The burden on the poorest households could be alleviated through introduction of a "lifeline" low tariff rate for essential uses, particu- larly lighting (paragraph 4.11). The Government should also adjust the tariff structure by eliminating the element of regressive tariff rates for additional blocks of consumption (paragraph 4.3) and increasing maximum demand charges to encourage improvements in power factors (paragraph 4.5). xvi. The rates for the sale of power to Kenya were last raised in 1980 and provide an average yield on the first 30 MW of only US cents 0.66/kwh. In the past, when Uganda had substantial surplus capacity, sales to Kenya at low rates were justified as extra income from power supplied at zero marginal cost. However, once Uganda's surplus capacity is fully absorbed by domestic demand, power supplied to Kenya will have an opportunity cost equal to the long-run marginal cost of installing new capacity discussed above. Even with these costs fully reflected in the export tariffs, Ugandan power would remain a low-cost source of supply for Kenya (paragraph 4.13). xvii. Petroleum prices. For petroleum products the most accurate indi- cation of economic cost is the import-parity price for supplies obtained from the Middle East market valued at the shadow exchange rate (USh 200 = US$ 1). Even after the latest increases, the retail prices of fuel oil, kerosene and auto. diesel are substantially lower than import-parity prices; the retail prices of gasoline are higher than import-parity prices (Table 4.7). These distortions arise pr:Lmarily from three basic factors: (a) the overvalued "window one" exchange rate used in the price formula for petroleum products; (b) the cross-subsidization of products at the Mombasa Refinery; and (c) differences in effective tax rates. The Mission there- fore recommends that the Government continues to adjust petroleum prices whenever there is a major change in the exchange rate. Under present con- ditions, it might be preferable to fix the exchange rate used in the price formula above the prevailing "window one" rate to anticipate future depre- ciation of the Ugandan shilling and thereby avoid more frequent price increases (paragraph 4.16). The Government should also realign the taxa- tion structure to offset the cross-subsidization of products. This implies  - ix - relatively higher tax rates on fuel oil in particular, but also on kerosene and auto. diesel; tax rates on gasoline could possibly be reduced, espe- cially if the Ugandan shilling depreciates further. Taxes on petroleum products make a major contribution to government revenue. Therefore as the official exchange rate tends towards the shadow exchange rate, it may be necessary on revenue grounds to raise retail prices above import parity (as has already occurred for gasoline). This would also help discourage smug- gling and the uneconomic use of petroleum products within Uganda (paragraph 4.24). xviii. Under the pricing formula, the oil companies receive a margin (for operating expenses, capital costs and profits) computed as 22% of the value of gasoline, kerosene and auto. diesel delivered to Kampala (exclud- ing customs duty). The oil companies generate revenue from other products as well and are able to allocate profits between Kenyan and Ugandan affili- ates, especially for products purchased on the Kenyan market. Therefore, the Ugandan Government is not able to assess the actual level of profit- ability of the oil companies operations in the country, nor is it possible to ascertain precisely the use of the foreign exchange allowance. The use of a fixed margin for operating expenses, independent of crude prices and import volumes, would also seem inappropriate. In addition, the Government should take measures to ensure that the component of the margin currently allowed for "rehabilitation and development of assets in Uganda" is used for the purpose intended or is reduced accordingly. More generally, the Mission recommends that the Government reviews the formula for reimburse- ment of all operating and capital expenditures with the oil companies (paragraph 4.21). xix. Woodfuel prices. In the woodfuels subsector, the Government's major concern is the high level of retail prices, especially for house- holds, in Kampala. Obviously, woodfuel prices will come down as transport and security around Kampala is improved and this is a priority of the Government. The official stumpage fee, which is presently only USh 200/m3, should also be reviewed. At present, the fee does offer an attractive financial return on fuelwood production from land with low economic poten- tial. However, large amounts of such land are unlikely to be available near to urban areas such as Kampala. There is therefore a strong case for raising the stumpage fee to reflect more fully the economic value of wood as a fuel. An initial adjustment to at least USh 1,000/m3 for forests serving urban areas would seem justified. This is unlikely to have any significant impact on retail prices, as the stumpage fee is a relatively small component of total costs and the increase would be partially absorbed by a reduction in the "windfall" profits of distributors. Provided the Government uses the revenue generated to strengthen its tree planting and management program, the overall impact on supplies will be positive over the longer term. In the final analysis, the only effective way to reduce retail prices for woodfuels in Kampala is to improve supply (paragraph 4.27).  x Institutions and Manpower xx. The energy sector in Uganda suffers from severe manpower and resource constraints. In many respects, these are economy-wide problems. However, they have been compounded in the energy sector by a weak and poor- ly coordinated institutional structure. In particular, no one institution (or clear hierarchy of institutions) has been given primary responsibility for energy planning and policy formulation. Irndeed, as yet, there has been very little recognition that energy concerns warrant a sectoral approach. As a result, day-to-day management is left to a plethora of separate enti- ties, while higher-level policy decisions have been taken without due regard for their impact on sector performance and development. The Mission considers that it would be premature at this time to try and tackle these problems by establishing a separate Ministry Df Energy. Instead, a more modest and phased program of institutional development is proposed (para- graph 5.4). xxi. Sector coordination and planning. Broadly speaking, the Mission proposes a three-tiered organizational structure (Table 5.2): day-to-day management would be left to the subsector institutions as at present, sector planning and policy formulation would be coordinated by an Energy Department in MPPT, and higher-level policy meLking would be the responsi- bility of an Inter-Ministerial Policy Committee. Initially, the Energy Department would comprise an Energy Secretariat and a small Energy Efficiency Section only. The Secret.ariat would support the Inter- Ministerial Policy Committee and formulate proposals for Cabinet consider- ation. It would also begin work on monitoring energy trends and help deve- lop an integrated approach to energy issues. The Energy Efficiency Section would be responsible for coordinating the activities of the many public and private organizations involved in energy-efficiency activities. The Sec- tion should also propose, through the Energy Secretariat, appropriate poli- cies for promoting more efficient use of energy in the economy. Eventual- ly, the Energy Department could be expanded to include separate sections responsible for coordinating subsector activities. However, manpower and budgetary resources for this are simply not available at the present time. Instead, it is proposed that the Government identify suitable staff and begin training them, within the existing institutional structure, to form the nucleus of an expanded Energy Department al: some future date. In the meantime, these staff can make a contribution through their normal line responsibilities as well as in liaison with the Energy Secretariat (para- graph 5.5). xxii. Subsector management and development. The institutional struc- ture for energy management and development at tlhe subsector level is large- ly in place and has performed remarkably well given country conditions. However, the Mission has identified the following weaknesses which need to be addressed on a priority basis: (a) UEB should strengthen its development planning capacity. At the moment, UEB is well staffed with operational personnel and engineers but has no economists to help analyze the viability of future investment proposals (paragraph 5.7);  - xi - (b) UEB's billings have often been delayed by computer malfunctions and breakdowns. As similar problems are faced by other para- statal bodies, the Government should consider establishing a central computer center, possibly managed and serviced by the supplier (paragraph 5.8); (c) there is an urgent need to strengthen the Government's capacity to monitor developments in the petroleum subsector and to nego- tiate effectively with the oil companies. The Bank of Uganda has recently established a petroleum desk and this should be strengthened to assume primary responsibility for overseeing petroleum importing and marketing (paragraph 5.9); (d) there has been inadequate attention given to woodfuel issues in the past. Until a separate woodfuels section can be established in the Energy Department, the Forest Department should assume responsibility for coordinating woodfuel production activities. Activities relating to the utilization of woodfuels, including charcoal production techniques and stove designs, should be handled separately by the proposed Energy Efficiency Section in the Energy Department, the National Research Council and Makerere University (paragraph 5.10); (e) GSMD is already well staffed with technical personnel to take responsibility for promoting and regulating petroleum explora- tion, but requires two or three legal and commercial staff to help draft and negotiate prospective agreements with foreign oil companies (paragraph 5.12); and (f) the National Research Council should take the lead in coordinat- ing the development of new renewable sources of energy. Much research in this area has already been done both in Uganda and under similar conditions in neighboring countries. The priority now is to evaluate this research, identify appropriate areas for development and propagate the use of practical techniques which are economic and acceptable to consumers. Support for public and private concerns interested in marketing the necessary equipment and other promotional activities (e.g., demonstration centers, tax incentives for conversion and building code provisions) should be considered (paragraph 5.13). xxiii. In terms of numbers, the manpower requirements related to the Mission's proposals are deliberately modest. Initially the only require- ments are for two additional economists to form the Energy Secretariat, two engineers and two economists for the Energy Efficiency Section, two econo- mists to strengthen the development planning capacity of UEB, and two or three legal and commercial staff in GSMD to deal with petroleum exploration promotion. The real priority now is not so much more manpower (which is not readily available in any case), but better utilization of the available manpower. This touches on issues of working conditions and wage levels  - xii - which go beyond the scope of this Report. However, it also involves re- training of staff who have not used their skills effectively in recent years, development of some new and specialized skills not previously em- phasized (e.g., energy planners, agro-foresitry experts and petroleum specialists), reorientation of training programs to focus on energy issues, and judicious use of external technical assistance (paragraph 5.14). Priority tasks identified in this Report, which could be assisted by tech- nical assistance, are summarized in Table 5.3. Follow-up support and longer-term technical assistance will also be required in a number of these areas (e.g., training). Identification of these requirements, and prepara- tion of appropriate action programs, should be an integral part of the initial technical assistance requirements (paragraph 5.16). Investment Priorities and Resource Requireiments xxiv. Substantial energy investments will be required over the next decade to rehabilitate existing assets and provide for expansion. Some of these investments (e.g., in petroleum exploration, woodfuel plantations, and the development of more efficient woodfuel stoves and new renewable energy sources) can be undertaken by the private sector, and the Government should encourage this. However, many of the investments will have to be made by the public sector. Priorities identified by the Mission for public investment during the remainder of the 1980s are summarized in Table 6.1 (all costs at 1982 prices). These include: (a) electricity. The immediate priority in the power subsector is to rehabilitate and upgrade the existing generating, transmission and distribution systems, to meet the expected growth in demand through 1990. The total cost of these investments is estimated at US$37 million. An additional US$14 million is provided for extending the transmission system. It is also important to begin planning now for construction of a second power project to meet projected demand during the 1990s. The location, scale and timing of this project will be determined by the proposed power development study. However, it seems likely that, without addi- tional exports, a second project of 60 MW (51 MW of firm capa- city) would be adequate to meet domestic demand and existing export contracts through 1995. This would cost at least US$72 million. A much larger station of up to 240 MW (160 MW of firm capacity) could be considered if additional export contracts for 120 MW could be signed in advance. However, this would cost an additional US$177 million, equivalent: to 45% of the 1983-90 investment program. Such a larger project would only be viable if other countries in the region benefitting from this investment repay Uganda through economic tariffs for the electricity pur- chased (paragraph 6.6). Furthermore, the Government should only consider this investment if suitable financing can be found to avoid significant net foreign exchange outflows in any year, even during construction (paragraph 6.10);  - xiii - (b) petroleum. The only major public investments anticipated in the petroleum sector during the 1980s are for preparing an explora- tion promotion package. This will cost about US$2 million and be complete by 1985. It is anticipated that any additional costs incurred by the Government would be fully covered by prospecting fees and royalties paid by the oil companies. If the package fails to attract oil company interest, the Government should not commit its own resources for further petroleum exploration, at least during the 1980s (paragraph 6.7); (c) woodfuels. The major priority in the woodfuels subsector is to strengthen the Forestry Department's planting and management pro- gram. The resource requirements for this program over the remainder of the 1980s are estimated at US$20 million (paragraph 6.8); and (d) energy efficiency. The Mission has identified a number of con- servation measures in the industry and transport sectors which would contribute to a general improvement in energy efficiency. The related investments (largely in the parastatal sector) would cost an estimated US$8 million over the next two years. No explicit provision has been made for industrial conversions as these generally require further study on a case-by-case basis (paragraph 6.9). xxv. Including the larger option for the second power project, the total resource requirements of this public investment program over the next eight years is close to US$400 million (at 1982 prices), of which 85% is in foreign exchange. This is equivalent to 15% of the projected investment in the economy. On an annual basis, the foreign exchange cost works out to about US$40 million, or 5% of the projected imports of goods and services over this period. This is considered manageable, provided the bulk of investment costs is covered by concessional financing. In fact, the real impact of these investments on the balance of payments will be substantial- ly less, because of the direct foreign exchange earnings/savings they generate. Similarly, the various price adjustments for electricity and petroleum products proposed by the Mission would reduce, if not eliminate, the need for investment resources from the government budget (paragraph 6.11). Even so, relative to what has been done in the past and the exist- ing pipeline of aid commitments for the energy sector, this is an ambitious program. If the necessary resources cannot be mobilized, the Government will have to accept some reduction in coverage and implementation delays. It is equally important that both the Government and donors make best use of available resources by giving priority to the most pressing requirements of the sector, rejecting low-priority investments (even if this means sacrificing tied external assistance) and avoiding duplication of external financing efforts (paragraph 6.13). As allowance has already been made for investments from private sources, and given Uganda's need for concessional assistance, most of the external financing will have to come from official sources. Large investments, such as for the second power station, will have to be financed by a consortium of donors, possibly supplemented with judicious use of suppliers' credits (paragraph 6.12).  - xiv - Regional Cooperation xxvi. The Mission has identified a number of issues directly related to the energy sector in Uganda which have a wider regional significance. These issues include: (a) the potential development of hydroelectric capacity in Uganda for exports to other countries in the region; (b) the operations of the Mombasa Refinery and its impact on petro- leum supplies and costs; (c) the use of alternative transport modes and routes for transport- ing petroleum products to Uganda; and (d) the exploration and possible future development of petroleum resources. There would seem to be economic advantages for both Uganda and other coun- tries in adopting a regional approach to these issues. Urgent joint action is required to reduce short-term economic costs (e.g., petroleum transpor- tation) and to provide a basis for longer-term planning (e.g., electricity exports). For Uganda, the Mission recommends that the proposed Inter- Ministerial Policy Committee takes the lead in regional discussions relat- ing to the energy sector (paragraph 5.5).  I. ENERGY IN THE ECONOMY: AN OVERVIEW 1.1 The Ugandan economy suffered a serious decline during the 1970s, due in large part to economic mismanagement. As for most other developing economies, Uganda's problems were compounded by a series of external shocks, including the sharp increase in petroleum prices after 1973. But, in general, the energy sector in Uganda has tended to mirror (rather than cause) developments elsewhere in the economy. For the 1970s, this implied falling demand, inadequate maintenance and low investment, distorted prices and a virtual absence of sector policy-making and planning. Now, following the introduction of major policy reforms in 1981 and 1982, there is encour- aging evidence of an economic revival in Uganda. If this recent progress can be sustained, it will have significant implications for the future growth of energy consumption. Equally, the availability of energy will be an important determinant of the pace at which economic recovery can pro- ceed. Related issues are discussed in the following chapters of this Report. However, as background, it is first useful to review briefly the recent performance of the economy, its impact on energy demand, and the present structure of the energy sector in Uganda. A. The Ugandan Economy 1/ 1.2 Uganda has an enviable reserve of natural wealth, with highly favorable soils and climate for agricultural production, a rich mineral base to support the industrial sector, and ample water and forest resources for energy development. Prior to the decline of the 1970s, the country's development was also supported by a well-developed transport system and an exceptional supply of skilled labor. The economy is dominated by agricul- ture, which provides livelihood to about 90% of the population and has sup- plied almost all of Uganda's exports in recent years. Although the indus- trial sector has always been small, it did in the past make a valuable con- tribution towards supplying the domestic market with basic goods and, in some instances (e.g., textiles and copper), produced a surplus for export. Economic activity is concentrated in the "fertile crescent" extending to the west from Lake Victoria. This productive belt includes the capital city of Kampala, the main industrial center of Jinja and the coffee/banana growing areas of the Buganda region. To the north and east, however, the country is less developed. In some areas (e.g., Karamoja) severe malnutri- tion has occurred in recent years, due to drought, poor internal food dis- tribution and local security problems. 1.3 At Independence, in 1962, Uganda was one of the strongest and most promising economies in Sub-Saharan Africa. The economy continued to perform well throughout the 1960s. However, after the coup in 1971, the 1/ For a more detailed discussion of recent economic developments in Uganda see Prof. Dudley Seers, et al.: The Rehabilitation of the Economy of Uganda, a report by a Commonwealth Team of Experts (June 1979), and the World Bank: Uganda, Country Economic Memorandum (1982).  situation quickly deteriorated. Under the military regime, many of the country's best skilled personnel left the country, the parastatal sector became bloated with the addition of many abandoned or confiscated indus- tries, and professional standards within the administration were eroded. On top of these largely self-imposed problexms, the Ugandan economy was shaken by a series of external shocks: (a) the sharp rise in petroleum prices after 1973 put additional pressure on Uganda's already weak balance of payments and in- creased production costs, especially in the industrial and tran- sport sectors; (b) with the breakup of the East African Community (EAC) in 1977, Uganda lost virtually all railway rolling stock and aircraft, as well as her part ownership rights to railway, port and telecommu- nications facilities in Kenya and Tanzania. This not only neces- sitated heavy investment by Uganda, but also led to substantial disruptions and cost increases on international traffic move- ments; and (c) the damage and looting which occurred during the 1978-79 war further eroded the country's productive capacity, especially around Kampala and to the south-west. 1.4 The impact of these domestic and external developments on the economy was severe (see Table 1.1). GDP at 1966 prices basically stagnated from 1970 to 1978 (implying falling per capita incomes2/), with particu- larly sharp falls recorded in the valute added of industry and monetary agriculture. Increasingly, Ugandans reverted to subsistence activities for survival. The savings rate fell sharply, and with limited external capital inflows, investment had to be cut back. Few development projects were started during this period, and little effort was made to replace obsolete plant and equipment. Even more importantly, little attention was paid to the maintenance of existing infrastructure and productive assets. The country's capacity to import essential items was steadily eroded, while the Government's budget deficits grew out of control. This in turn fuelled price increases, with inflation averaging more than 70% per annum from 1976 to 1979. For those dependent on officially-set prices and wages, such as export crop producers and civil servants, rea]. incentives fell dramatical- ly. It is not surprising, therefore, that corruption became commonplace and a large share of economic activity was diverted to the pervasive paral- lel market system, commonly referred to as "magendo." 1.5 By April L979, when the military regime was overthrown, the Ugandan economy was in ruins. While there was at that time high hopes for rapid economic recovery, initial progress was constrained by the unsettled 2/ Uganda's population is estimated to have grown by 2.8% per annum during the 1970s, reaching 13 million by 1981.  - 3 - Table 1.1: ECONOMIC INDICATORS a/ 1965-70 1970-75 1975-80 1980-82 A. GDP growth rates (% per annum in 1966 prices) Monetary economy 5.3 -1.5 -5.0 4.7 - Agriculture 6.0 -0.9 -5.0 6.2 - Industry b/ 5.7 -5.4 -16.3 2.6 - Transport & Communication 7.0 2.8 -16.5 6.6 - Other 4.3 -1.3 -1.4 3.9 Subsistence economy 3.9 3.6 -2.4 7.4 Total GDP 4.8 0.1 -4.0 5.7 B. Investment rate c/ 14.6 10.2 6.0 7.0 (% of GDP) C. Prices (% change per annum) Domestic inflation d/ 3.8 17.7 67.3 73.2 Merchandise exports e/ 3.8 f/ 5.6 19.8 -19.1 Merchandise imports e/ 0.5 T/ 18.2 14.5 -2.6 External terms of trade e/ 2.8 fi -10.4 2.3 -16.9 D. Trade volumes (% change per annum) Merchandise exports 2.6 f/ -6.3 -10.2 22.7 Merchandise imports 2.7 T/ -13.6 -1.5 -2.3 E. Current account deficit jI -57.2 h/ 141.9 234.7 223.5 (US$ million at 1982 prices at end of period) a/ Growth rates are trends, calculated by the least-squares method. b/ Mining, manufacturing and agro-processing. c/ Period averages, excluding the base year. d/ Based on the GDP deflator. eI In US dollar terms. T/ For 1966-70 only. g/ Converted to 1982 prices using the import price index. h/ Surplus. Sources: (1) Ministry of Planning and Economic Development. (2) EAC and Uganda Customs, Annual Trade Reports. (3) Bank of Uganda. (4) World Bank estimates.  - 4 - political situation, administrative and manpower constraints, and a severe shortage of foreign exchange. From the Ugandan side, little progress was made on policy or institutional reforms (during 1979 and 1980. No decisive action was taken to remove price distortions and the exchange rate remained seriously overvalued, with the parallel market offering as much as ten times the official rate. Budgetary performance continued to deteriorate, as the revenue base contracted and efforts to re-establish an effective system of expenditure control met with only limited success. In the exter- nal sector, export recovery was inhibited by inadequate incentives and the impact of processing and transport bottlenecks. The purchasing power of Uganda's exports was further eroded by adverse trends in the external terms of trade, which deteriorated by more than 50. from 1978 to 1981. There was, in fact, an encouraging response from the international community in terms of new aid commitments, but these could not be effectively utilized due to poor aid coordination within Uganda and the unfavorable country con- ditions. As a result of these developments, the volume of imports in 1981 was almost 30% lower than in 1978 and only one-third of the peak levels attained in the early 1970s. 1.6 In mid 1981, the Government introduced a major package of policy reforms and has followed this with further measures over the past two years. This program, supported by stand-by arrangements with the IMF, is intended to stabilize the economy and eventually revive investment and pro- duction through restoring a measure of confidence in the currency, reducing price distortions and improving fiscal aLnd monetary discipline. The most decisive actions have been to devalue the Ugandan shilling by more than 90% since mid 1981,3/ and to raise substantially i;he official prices for major export crops, petroleum products and public utilities (including power). The initial impact of these measures on econcmic performance has been en- couraging. Led by higher coffee sales, export earnings rose by more than 33% (in US dollar terms) during 1982. There has also been a significant increase in output from a number of key industries, including cement, textiles, sugar and cigarettes. As a riesult, overall GDP is estimated to have grown on average by 5.7% per annum in 1981 and 1982. Finally, the impact of the program on consumer prices has been moderated somewhat by the fact that many goods were already being sold at "inflated" magendo prices. Because of this, the overall inflation rate is estimated to have fallen from around 100% per annum in 1979-81 to 50% in 1982. 3/ The exchange rate was devalued from USh 8 = US$1 to USh 80 = US$1 to the US dollar in June 1981 and had depreciated further to USh 100 = US$1 by the end of 1982. In addition, the Bank of Uganda has now opened a second window where foreign exchELnge is more freely traded in an auction system. The rate established at the second window steadily fell from USh 300 = US$1 in August to USh 240 = US$1 by the end of 1982. It is intended that ithe differential between the two exchange rates will continue to be narrowed and eventually eliminated.  - 5 - B. The Energy Sector Energy Consumption and Supply 4/ 1.7 Energy balances for 1970 and 1980 are provided in Tables 1.5 and 1.6 at the end of this Chapter. During 1980,5/ per capita energy consump- tion in Uganda is estimated to have been 0.35 TOE, of which only 0.06 TOE was commercial. This level of commercial energy consumption, while excep- tionally low by world standards, is comparable to estimates for some other low-income countries in Sub-Saharan Africa.6/ In Uganda's case, the low level of commercial energy consumption reflects not only the country's low per capita income, but also the dominance of the subsistence sector (which is not a factor in commercial energy consumption) and the significant decline of the industrial and transport sectors during the 1970s. For the same reasons, energy use is concentrated in the household sector (80% of total energy and 37% of commercial energy in 1980) and supplied primarily from woodfuels (95% of total energy and 71% of commercial energy in 1980). 4/ Energy sources covered in this Chapter are petroleum, electricity and woodfuels. In addition, crop residues and dung are estimated to have provided 110,O000 TOE of energy in 1980. No information is available on energy from human and animal power. Commercial energy is defined as all energy traded outside the subsistence sector. 51 1980 is used as an appropriate reference point to illustrate the impact of the economic decline during the 1970s on the energy sector. 6/ Data for a sample of other countries in Sub-Saharan Africa are as follows (in TOE per capita): Commercial Total Year energy a/ energy Kenya 1979 0.12 0.43 Tanzania 1979 0.03 Malawi 1979 0.04 Zambia 1979 0.51 Zimbabwe 1980 0.50 0.72 Ethiopia 1979 0.01 a/ These estimates generally exclude the commercial use of woodfuels and therefore are not strictly comparable to the estimates for Uganda used in this Report. Excluding woodfuels, the estimated consumption of commercial energy in Uganda for 1980 is 0.02 TOE per capita. Sources: (1) World Bank/UNDP, Energy Assessment Reports. (2) World Bank, World Development Report 1982.  -6- Table 1.2: TRENDS IN ENERGY CONSUMPTION a/ Growth rate (% per annum) Composition (% of total) b/ 1965-70 1970-75 1975-80 1980-82 c/ 1965 1970 1975 1980 1982 c/ A. Commercial energy 6.9 -2.5 -4.1 -2.4 100.0 100.0 100.0 100.0 100.0 By source - Petroleum 16.3 -4.4 -9.6 -14.1 25.9 39.3 35.0 26.8 20.8 - Electricity 4.9 -2.5 -9.5 5.2 3.5 3.1 3.3 2.6 3.2 - Fuelwood 1.6 -3.9 -3.8 1.2 66.0 51.1 47.9 48.2 51.9 - Charcoal 15.0 13.1 6.4 1.7 4.5 6.5 13.7 22.3 24.0 By end use - Household 3.5 -0.1 -2.7 1.7 35.7 30.4 34.6 37.2 40.3 - Commerce 3.0 3.0 1.9 2.7 14.1 11.7 15.6 21.0 14.8 - Industry 8.3 -5.6 -7.0 -2.3 30.6 32.5 26.4 21.7 23.1 - Transport 12.6 -5.0 -8.4 -16.1 19.8 25.5 23.4 20.2 21.8 B. Non-commercial energy d/ 1.3 1.7 1.5 1.9 100.0 100.0 100.0 100.0 100.0 By end use - Household 1.0 1.3 1.0 1.6 94.6 93.0 91.3 89.2 88.6 - Commerce 7.4 7.1 6.3 4.9 4.4 5.9 7.6 9.6 10.1 - Industry 2.4 2.1 3.0 2.2 1.0 1.1 1.1 1.2 1.2 C. Total energy 2.6 0.7 0.4 1.2 100.0 100.0 100.0 100.0 100.0 By source - Petroleum 16.3 -4.4 -9.6 -14.1 5.5 10.3 7.8 4.8 3.4 - Electricity 4.9 -2.5 -9.6 5.2 0.8 0.8 0.7 0.5 0.5 - Fuelwood 1.4 0.9 0.9 1.8 92.8 87.2 88.4 90.8 92.0 - Charcoal 15.0 13.1 6.4 1.7 1.0 1.7 3.1 4.0 4.0 By end use - Household 1.2 1.2 0.7 1.6 82.0 76.6 78.6 79.9 80.6 - Commerce 5.4 5.5 4.7 4.2 6.5 7.4 9.4 11.6 12.3 - Industry 7.7 -4.8 -5.4 -1.4 7.3 9.3 6.8 4.9 4.6 - Transport 12.6 -5.0 -8.4 -16.1 4.2 6.7 5.2 3.6 2.5 D. Per capita energy consumption 457.0 433.0 393.0 352.8 341.6 (kg of oil equivalent) Commercial 95.3 113.4 87.4 62.9 56.6 Non-commercial 361.7 319.6 305.6 289.9 285.0 E. Average elasticities Commercial energy with respect to monetary GDP 1.3 1.7 0.8 -0.5 Total energy with respect to total GDP 0.5 e/ -0.1 0.2 a/ This table is derived from domestic consumption data in TOE, adjusted for transformation and losses. All exports (official or unofficial) are excluded. Growth rates are trends, calculated by the least- squares method. b/ Percentages show composition of commercial energy for Section A, non-commercial energy for Section B and total energy for Section C. c/ Estimates. T/ All fuelwood. e/ GDP stagnated during this period. Therefore, no meaningful elasticity can be calculated. Source: Annex I.  Electricity, which is the major focus of government and parastatal involve- ment in the energy sector, has never supplied more than 4% of commercial energy consumption. In terms of primary supply, 84% of Uganda's commercial energy was domestically produced in 1980, with the balance coming from imported petroleum products.7/ About 7% of the net supply of commercial energy was consumed outside Uganda, including electricity exports to Kenya and unofficial sales (smuggling) of petroleum products to most neighboring countries. 1.8 The major trends in energy consumption since 1965 are summarized in Table 1.2. The following points should be noted: (a) total energy consumption in Uganda stagnated during the 1970s and, although there has been some growth over the past two years (1.2% per annum), per capita consumption continues to fall. These trends are consistent with the economic developments dis- cussed above;8/ (b) non-commercial energy consumption has grown very steadily since 1965 at an average rate of 1%-2% per annum. (This increase was sustained throughout the 1970s, despite the decline in subsist- ence agricultural production after 1975). Above average growth occurred in the non-commercial consumption of the commerce and industrial sectors, as farmers became increasingly dependent on their own supplies (e.g., brewing and brick-making). In the household sector, however, non-commercial consumption of wood- fuels has declined on a per capita basis, reflecting the growing scarcity of wood supplies and partial switching to crop residues and dung as energy sources; and (c) consumption of commercial energy, on the other hand, has been much more volatile, rising more strongly (at 6.9% per annum) 7/ Apart from very small amounts of fuel oil used to generate electricity and the conversion of wood to charcoal, all fuels are consumed in their primary form. Conversion losses on electricity simply reflect the dif- ference between thermal-equivalent and actual hydro generation. Note that the energy balances treat all petroleum products as imported at the Ugandan border. In practice, however, about 50% of the primary supply for petroleum products is made up of crude oil purchased for Uganda by oil companies and refined in Mombasa; fuel oil produced in excess of Ugandan requirements is exported. This stage of the energy supply sequence is omitted from these balances for simplification. 8/ The general fit between economic and energy trends can be deduced from Tables 1.1 and 1.2. Average elasticities are also shown in Table 1.2. However, these are not considered to be very meaningful under Ugandan conditions, where many trends have been stagnant or declining, and where relationships have been affected by price and institutional distortions.  - 8 - than monetary GDP during the late 1960s and falling more rapidly (at 3.3% per annum) during the 1970s; This trend is especially marked for electricity and petroleum-9/ In part, this reflects the changing sectoral composition of commercial energy consump- tion, with the shares of trarsport (the major consumer of petro- leum) and industry (l-he major consumer of electricity) rising during the late 1960s and falling during the 1970s. However, similar trends are also evident within the consumption of each consuming sector. For the 1970s, this suggests that: (i) within some sectors, activities usiqn petroleum and electricity as ener- gy sources declined more rapidly (e.g., the virtual collapse of large-scale industries within the industrial sector); and (ii) there was substitution towards woodfuels, which are not so di- rectly dependent upon foreign exchange availability or central distribution (e.g., substi tution oE charcoal for kerosene in the household sector). 1.9 Electricity. Basic data on the electricity subsector are given in Tables A1.1 to A1.3 of Anneox I. Uganda has abundant hydroelectric po- tential estimated at about 2,000 MW with an annual firm generation capabi- lity of 10,000 GWh. To date, the only major utilization of this potential is at the Owen Fall]s Station, situated just downstream from the Victoria source of the Nile. This Station was first commissioned in 1954 and ex- panded to its present capacity cf 150 MW by 1968. The balance of Uganda's public electric-generating capacity ccxmes fr om nine small diesel stations (4.4 MW) and one small hydro station (:L.0 MW) in the northern and western regions.10/ The operating performance of the Owen Falls Station has been good, due in large part to the abundant storage, fairly stable head and pure water provided by Lake Victoria. However, none of the ten units has been overhauled since installation, and with one unit (15 MW) usually held for reserve, the firm capacity is now probably down to about 105 MW. Simi- larly, four of the nine diesel stations (as well as the small hydro station at Kabale) are alreadv inoperative due to security problems, fuel shortages and lack of spare parts. Even so, despite a low average system load factor 9/ Petroleum consumption was constrained during 1982 due to supply bottle- necks. Otherwise, it is probable that commercial energy consumption would have increased during, 1980-82, consistent with the recovery in economic activity. 10/ In addition, there are some hydro and diesel stations installed, owned and operated by industries. No inventor-y of these is available. In general, however, they are very small and most have been inoperative for a number of -years.  - 9 - (65% in 1982) and sales to Kenya,11/ the maximum demand on the system has remained below capacity, falling from a peak of 132 MW in 1968 to only 100 MW in 1982. 1.10 The country's major transmission system (132 kV and 66 kV) ex- tends across the southern part of the country, to cover Masaka, Kampala and Jinja to the west of Owen Falls and Tororo to the east (see the map at the end of this Report). At the Tororo substation, the line connects with the Kenyan system and to a northern line running up to Lira. Distribution in towns is mainly by 33 kV and 11 kV lines. As the system is supplied from one source, any fault at Owen Falls Station results in total loss of supply throughout the country and to Kenya. This has happened many times, al- though the duration has generally been less than two hours. As some of the transmission lines are very long and pass through unpopulated areas, they are difficult to maintain. The distribution networks of the main towns, such as Kampala and Tororo, are reaching their capacities and voltage drops occur frequently. Local overloading occurs due to increased domestic demand, including illegal connections. As many of the substations and cables in the distribution system are now 30 or more years old, with only limited maintenance in recent years, there is a serious backlog of replace- ment requirements. System transmission and distribution losses averaged about 13% of total sales during the 1970s (see Table A1.2), although they rose to about 20% in 1979 and 1980 due to the general disruptions in the country. Since losses on bulk supplies to Kenya are low, a more reliable measure is losses as a proportion of Ugandan sales, which were around 25% during the 1970s and rose to 46% in 1980. Subsequently, losses as a per- centage of both total sales and Ugandan sales have fallen back to the levels of the mid 1970s. Even so, these energy losses remain disturbingly high and have not created serious problems for UEB only because they have been generated at zero marginal cost from surplus hydroelectric capacity. Some measures to reduce these losses are now underway, including emergency works under the Government's Recovery Program, and rehabilitation of UEB's transmission and distribution system is scheduled to be studied shortly (paragraph 3.6 and Table 6.3). 1.11 UEB estimates that about 5 million people (40% of the population) live in areas served by the Owen Falls system and isolated plants. How- ever, the number of connections by end 1981 was only 93,156, of which 71% were for domestic consumers. Assuming an average household size of six, the number of people having access to electricity is estimated to be only 0.4 million, 8% of the population living in areas supplied by electricity and only 3% of the total population. Per capita consumption of electricity 11/ Uganda entered into a 50-year agreement in 1955 to supply power to Kenya. The initial agreement provided for 45 MW of firm power, but this was reduced in 1964 to 30 MW and Kenya was committed to maintain- ing a minimum load factor of 90%. During 1981, maximum Kenyan demand was 30 MW during the day and 75 MW at night. Energy sales to Kenya reached a peak level of 303 GWh in 1973, but have averaged only 211 GWh per annum over the past five years.  - 10 -- in 1982 was only 21 kWh, one of the lowest levels in the world.12/ How- ever, this reflects not only the limited access to electricity but also the impact of declining economic activity over the past decade. In the late 1960s, for example, per capita consumption of electricity was more than double present levels. Total consumption of electricity fell by more than 40% during the 1970s, with industrial use (which had accounted for two- thirds of electricity consumption in 1969) down by more than 60%.13/ This is indicative of the virtual collapse of industrial activity during this period, especially in the large-scale sector. During 1980, domestic con- sumption of electricity was actually lower than exports to Kenya. Subse- quently, there has 'been a recovery in electricity consumption. This was concentrated in the household and commerce sectors during 1981. During 1982, consumption in both of these sectors fell, probably due to the impact of higher tariffs, but this was offset by a 34% increase in industrial con- sumption. 1.12 Petroleum. Uganda imports all of its petroleum requirements, either by refining crude through the Mombasa refinery or by direct purchas- es on the Kenyan market. The products are transported to Nairobi by pipe- line and on to Kampala by rail or road (more than 90% were transported by road in 1982 compared to only 20% in the early 1970s). The six oil compa- nies operating in Uganda maintain their own storage facilities as well as retail outlets. In line with the pattern of economic activity, distribu- tion and consumption of the petroleum products is concentrated around Kampala (which accounts for 60% of all gasoline and auto. diesel sales) and Jinja (which accounts for 80% of fuel oil sales). The complex import and marketing arrangements for petroleum products, as well as alternative supply options, are discussed further in Chapter III, Section B and Chapter IV, Section B. 12/ Data for a sample of other countries 'n Sub-Saharan Africa are as follows: Population Per capita Year with access consumption (%) (kWh/year) Kenya 1980 6 92 Tanzania 1980 4 44 Malawi 1980 2 64 Zambia 1980 10 172 Zimbabwe 1980 16 928 Ethiopia 1980 2 22 Sources: (1) World Bank/UNDP, Energy Assessment Reports. (2) Mission estimates. 13/ This is exclusive of a 14% decline in industrial consumption during 1969, resulting from a switch from electricity to fuel oil for raising steam, primarily at Nyanza Textiles.  - 11 - 1.13 The volume of petroleum imports, excluding non-energy products, peaked at 431,000 tons in 1970. However, during the 1970s, as the economy declined and foreign exchange became increasingly scarce, the volume was sharply curtailed, falling below 250,000 tons per annum in 1978-80. Fur- ther, an increasing proportion of these imports was smuggled out of the country. In 1980, for example, it is estimated that up to 40% of packed kerosene imports were smuggled to Rwanda and Zaire and 20% of auto. diesel imports back into Kenya. As a result, domestic consumption of petroleum products during 1978-80 was probably only about half the peak level of 1971. Throughout this period transport remained the largest end user of petroleum, accounting for about two-thirds of total consumption. This was despite a severe reduction in imports of aviation fuel, caused by the virtual cessation of international air services to Uganda. Industrial con- sumption of petroleum products (mainly fuel oil) fell rapidly throughout the 1970s, and by the end of the decade it was only about one-third.of its 1970 peak level. Domestic consumption (mainly kerosene) took longer to respond to the deteriorating economic conditions, but it too fell after 1975. 1.14 Following the major retail price adjustments introduced in 1981, there was a noticeable decline in both domestic consumption and smuggling of petroleum products. This trend was compounded by supply constraints during 1982, caused by the shortage of foreign exchange and refinery/pipe- line problems in Kenya. Therefore, although demand pressures re-emerged during the year, this could not be matched by increased imports. This led to a serious drain on Uganda's already minimal petroleum stocks, which fell to less than five days of consumption for premium gasoline, auto. diesel and kerosene by the end of the year (see Table A1.6). However, following recent changes in supply arrangements (see Chapter III, Section B), the stock position has now substantially improved, reaching an average of four weeks' consumption by June 1983. 1.15 Woodfuels. The available information on woodfuels supply and consumption is very incomplete, due in part to the decentralized organiza- tion of the subsector and in part to the little attention given to wood- fuels by government organizations in the past. For this Report, rough estimates have been prepared from a variety of sources: partial and out- dated surveys, random samples around the Kampala area, and relevant experi- ence of other countries in the region (see Tables A1.7 and A1.8 of Annex I). Fortunately, the picture that emerges is relatively straightforward and generally consistent with developments elsewhere in the economy. How- ever, the conclusions can only be treated as preliminary and much remains to be done before a reliable data base on woodfuels is completed. 1.16 Uganda's forests and woodlands are estimated to cover 2.76 mil- lion hectares, or 12% of the country's total area (including inland waters). However, an inventory of all wooded areas, including those out- side the forests and woodlands, is needed to get an accurate assessment of the country's woodfuel potential (paragraph 3.32). According to the rough estimates presented in Table A1.7, the total growing stock of Uganda is about 450 million tons, providing a sustainable annual yield of 14.1 million tons. Allowing for non-energy uses of wood (for poles, sawnwood,  - 12 -- panels and paper products), the sustainable annual yield available for fuel is about 10.9 million tons or 4.1 million TOE.14/ Less than half of this available energy comes from forests and woodlands and almost one third from arable land and pastures. By comparison, present wood energy use -- including conversion losses in charcoal production -- is estimated at 5.0 million TOE (see Table Al.8), implying that the country's wood capital is being eroded.15/ In certain areas, where population densities are high or the available land is being converted to agricultural uses, the accessibil- ity of woodfuels has already been seriously affected and there is a danger that local wood supplies will soon be depleted. 1.17 Total end-use consumption of woodfuels has grown steadily by 1%- 2% per annum over t:he past decade and a half. This trend, however, is dominated by the non-commercial use of woodfuels in rural areas, which has declined in per capita terms due to the growing scarcity of wood supplies and partial switching to crop residues and dung as energy sources. In the commercial sector, somewhat different trends are evident: (a) per capita commercial consumption of woodfuels in urban areas continued to rise through 1975, with a significant switch from fuelwood to charcoal (which resulted in an even greater increase in wood energy used). Subsequently, the growth rate of consump- tion has slowed. Since 1980, there has also been some switching back to fuelwood, as charcoal supplies have been affected by transport and security bottlenecks; and (b) per capita commercial consumpt:ion of woodfuels in rural areas has fallen consistently since 1965, with an especially sharp decline during the 1970s. This reflects a number of developments: (i) the general decline in agricultural production and supporting industries; (ii) some improvement in the efficiency of woodfuel use, in response to growing supply constraints and rising costs; and (iii) conversion of the Tororo cement factory from charcoal to fuel oil in the early 1970s. As in urban areas, there was a noticeable switch from fuelwood to charcoal consumption during the 1970s. Energy Prices 1.18 General price inflation in Uganda (EIs measured by the GDP defla- tor) averaged 41% per annum during the 1970s, with a marked acceleration after 1973. However, energy prices rose much less rapidly, by 15%-25% per 14/ Actual sustainable removals will be somewhat less, as some forest areas are protected and cannot be exploited. 15/ The margin of error on these estimates is of course large. However, this general conclusion is also supported by independent surveys con- ducted by the Forestry Department, which show a 5% decline over the past 15 years in the area covered by forests and woodlands.  - 13 - annum for most petroleum products and woodfuels and by only 6% per annum for electricity (see Table 1.3). For petroleum products, this also implies a major distortion between domestic prices and import costs, which rose on average by close to 30% per annum in US dollar terms during the 1970s, and would have increased substantially more in Ugandan shilling terms if the exchange rate had been adjusted to reflect the scarcity value of foreign exchange. The consequences of these trends were predictable: (a) the real cost of energy from all sources declined, encouraging waste and inefficient use.16/ In the case of petroleum prod- ucts, this was offset to a large extent by the impact of the balance of payments situation on supplies, which led to enforced shortages and rationing.17/ But, in the case of woodfuels, total consumption continued to rise, leading to erosion of the country's wood capital as noted above; (b) the price of electricity, relative to other energy sources, declined. To some extent, this trend was justified by the very low marginal costs of producing power from installed hydroelec- tric capacity relative to the sharp increase in world petroleum prices and the increasing scarcity of foreign exchange. But, in practice, the impact on fuel substitution seems to have been min- imal, due to the overriding impact of industrial decline on elec- tricity consumption and the unfavorable climate for investment in conversion to alternative energy sources; (c) petroleum products in Uganda became substantially cheaper than in neighboring countries, especially when traded at black market exchange rates (which offered up to ten times more Ugandan shillings per US dollar than official rates by 1980). This in- evitably encouraged the growth of smuggling noted above; (d) the government revenue generated by customs duties and sales taxes on petroleum products was steadily eroded (in real terms), due to the narrowing differential between import costs and retail 16/ There were many other factors which also contributed to inefficient use of energy (see Chapter II, Section A). In the industrial sector, for example, efficiency fell due to low plant throughput, poor plant condition and inadequate energy management. In the transport sector, energy efficiency is constrained by insufficient maintenance, shortages of spares and the lack of suitable diesel-testing equipment. 17/ It is possible, under these conditions, that secondary markets devel- oped for some petroleum products, with prices substantially above offi- cial levels. However, no clear evidence on this is available. For those having privileged access to petroleum products, the most valuable uses were probably in own-consumption or in direct smuggling to neigh- boring countries, and this probably limited the development of internal secondary markets.  Table 1.3: RETAIL ENERGY PRICES Comparative Prices for 1982 -- (US$/TOE) b/----- Average Average Kenya 1970 a/ Change 1980 Change 1982 Uganda Retail Prices Retail (X per annum) (X per annum) A c B d/ Prices e/ A. Petroleum prices f/ (in USh/liter) Aviation fuel .. (.) 4.3 (303.5) 70.0 863.0 450.1 Gasoline - premium 1.4 (23.0) 7.4 (349.0) 150.0 1,861.7 970.9 813.4 - regular 1.3 (23.6) 7.2 (340.7) 140.0 1,737.6 906.2 764.6 Auto. diesel 1.2 (17.8) 4.3 (357.5) 90.0 1,020.3 532.1 511.7 Fuel oil .. (.) 3.4 (283.5) 50.0 506.5 264.1 Kerosene 0.8 (21.6) 3.8 (360.0) 80.0 926.1 483.0 370.9 B. Electricity revenue (in USh/MWh) On Uganda sales 216.9 (6.2) 397.5 (44.4) 829.0 105.3 49.8 693.6 On Kenya sales 36.7 (6.5) 69.2 (290.6) 1,056.0 84.5 C. Fuelwood prices./ (in USh ton) Rural - household 42.0 (16.9) 200.0 (217.0) 2,010.0 56.7 26.8 7.1 - non-household 13.0 (22.6) 100.0 (213.0) 980.0 27.7 13.1 3.9 1 Urban - household 80.0 (23.9) 680.0 (216.7) 6,820.0 192.4 90.9 14.2( - non-household 25.0 (21.1) 170.0 (218.1) 1,720.0 48.5 22.9 6.8 4 D. Charcoal prices (in USh/ ton) Rural - household 220.0 (25-9) 2,200.0 (126.1) 11,250.0 154.0 72.8 48.2 - non-household 180.0 (25.9) 1,800.0 (123.6) 9,000.0 123.2 58.2 37.2 Urban - household 280.0 (25.9) 2,800.0 (175.0) 21170-0 289.8 136.9 70.9 - non-household 200.0 (25.9) 2,000.0 (191.0) 16,940.0 231.9 109.6 53.5 E. Memo item Exchange rate (USh/US$) - End of year 7.14 (0-6) 7.57 (271.2) 104.30 h/ - Annual average 7.14 (0-4) 7.42 (256.9) 94.51 h/ a/ 1972 for petroleum prices. b/ Unadjusted for end-use efficiency (compare to Table 4.9). c/ At the official "first window" exchange rate, except for revenue on electricity sales to Kenya which is converted at annual average exchange rate of UShl5O - US$1 (the rate at which it was earned). d/ At a shadow exchange rate of USh 200 - US$1. On the black market, the exchange rate is probably closer to USh 300 - US$1 at the present time. e/ At the official exchange rate of KSh 10.9 - US$1 on average for 1982 and KSh 12.7 - US$1 at the end of 1982 (for petroleum products). f/ At end of year. Prices are at the pump for gasoline, auto. diesel and kerosene, and ex-depot for other products. g/ Estimated free market prices. h/ At the "first window". Sources: (1) Annex I. (2) Mission estimates.  - 15 - prices, the valuation of products at the official exchange rate, and the loss of official sales to smuggling. 1.19 Over the past two years, the Government has made commendable efforts to correct distortions in energy prices, as an integral part of its financial programs with the IMF. As shown in Table 1.3, retail prices of petroleum products have been increased at least in line with the devalua- tion of the Ugandan shilling and by as much as 2,200% for kerosene. Dis- tortions still exist in the relative prices of petroleum products, primari- ly due to the lower tax burden on kerosene, auto. diesel and fuel oil vis-a-vis gasoline and this probably has encouraged some unwarranted fuel substitutions (e.g., blending of kerosene into gasoline). There is also still scope for profitable smuggling of products to neighboring coun- tries.18/ Nevertheless, there is little doubt that the overall discrepancy between petroleum retail prices and import costs (the latter have been relatively stable over the past two years) has been substantially reduced. Unfortunately, comparable adjustments have not been made in electricity prices. Although domestic tariffs have been increased by 240% since mid 1980, electricity in Uganda still costs less than US cent 1 per kWh (at the "first window" exchange rate), one of the lowest rates in the world. Even allowing for Uganda's comparative advantage in hydroelectric generation, the long-run marginal cost of supply is estimated to be close to US cents 5 per kWh (see Chapter III, Section A), or five-times present tariff levels. Woodfuel prices, which are not officially controlled, have continued to rise, reflecting the increased prices of alternative fuels (especially kerosene and fuel oil), higher transport costs, security bottlenecks and the steady depletion of the most accessible and economic forest resources. However, the price adjustment for woodfuel has been more moderate than for petroleum products, in part due to improvements in consumption efficiency but probably more so to the isolation (in distance and income) of many rural consuming areas from alternative commercial fuel options. In urban areas, charcoal prices have risen sharply in recent months, as the supply has been disrupted by transport and security bottlenecks. This in turn has encouraged some switching back to fuelwood consumption (a reversal of the 1970s trend), with a consequent saving in the utilization of wood raw mate- rial. An analysis of present energy prices, and the Mission's recommenda- tions for energy pricing policies, are presented in Chapter IV. 18/ As shown in Table 1.3, Ugandan retail prices for petroleum products are generally double those in Kenya, when converted at the official "first window" exchange rate. Using a shadow exchange rate of USh 200 = US$1, the discrepancy is substantially reduced but not eliminated. But, on the black markel: (which still offers at least three times the "first window" rate for foreign exchange), Ugandan prices are lower. Compared to Rwanda, Ugandan kerosene and auto. diesel prices are still relative- ly cheap, even at the shadow exchange rate.  - 16 - Impact on Balance of Payments 1.20 Uganda's external energy balance is dominated by imports of pet- roleum products, the cost of which rose from US$14 million in 1970 to a peak of US$124 million by 1980, an average increase of 24% per annum (see Table 1.4).19/ This increase was due entirely to higher prices; the volume of petroleum imports actually fell by more than 40% (in terms of TOE) during the 1970s. The actual foreign exchange cost would have been less if smuggling losses, which probably accounted for 10%-20% of all pet- roleum imports during 1980, could have been curtailed. The foreign ex- change earned from electricity sales to Kenya has been relatively small due to the exceptionally low tariff levels. As a result, the external energy balance rose from only 5% of merchandise exports in 1970 to 21% in 1975 and 38% in 1980 (when it also contributed almost one-half of Uganda's overall current account deficit). 1.21 Subsequent to 1980 there has been some improvement in this situa- tion, due to the stabilization of petroleum import costs and the impact of higher retail prices on domestic consumption and smuggling of petroleum products. However, there are already signs that additional petroleum im- ports will be required soon to support the ongoing economic recovery. Al- though more foreign exchange could possibly be realized from a renegotia- tion of the electricity export contract with Kenya, the amounts involved are unlikely to have any significant impact in the context of the overall balance of payments. Similarly, Uganda will not be able to substantially increase the level of electricity exports to Kenya and other neighboring countries until a second station is operational. This will be in 1991 at the earliest (see Chapter III, Section A). For the immediate future, some relief might be provlded by lower world market petroleum prices, but this remains an uncertain prospect and one which, in any event, is beyond the control of Uganda. Actionable options include: (a) improved arrangements for the purchase and transport of petroleum products (see Chapter III, Section B),; (b) more efficient use of petroleum products, especially in the tran- sport and industrial sectors (see Chapter II, Section A); and possibly (c) industrial substitution away from petroleum towards woodfuels and possibly, in the 1990s, electricity (see Chapter II, Section B). 19/ The value of petroleum imports after 1971 is based on the foreign ex- change released to oil companies by the Government of Uganda. In prac- tice, this includes payments for dividend remittances, services and oil-related imports, as well as for petroleum products. At present, it is estimated that petroleum import payments account for approximately 80% of the foreign exchange released to oil companies, but there is no precise way to disaggregate the payments on a year-to-year basis.  - 17 - Table 1.4: ENERGY AND THE BALANCE OF PAYMENTS (in US$ million) 1970 1975 1980 1982 A. Exports Energy 1.6 1.6 2.7 1.8 - Electricity (1.3) (1.1) (2.7) (1.8) - Petroleum re-exports (0.3) (0.5) ( .) (.) Total goods 261.6 237.2 319.1 314.4 Total goods and services 297.1 251.9 334.4 331.0 B. Imports Energy 14.4 50.7 124.0 100.0 - Petroleum products (14.3) (50.7)a/ (124.0)a/ (100.0)a/ - Coal and coke (0.1) (-) (-v) (..) Total goods 204.9 262.4 503.7 435.5 Total goods and services 271.3 320.6 582.2 554.5 C. Net balance Energy -12.8 -49.1 -121.3 -98.2 Total goods 56.7 -25.2 -184.6 -121.1 Total goods and services 25.8 -68.7 -247.8 -223.5 Current account 13.3 -76.1 -247.5 -223.5 D. Memo items Net energy deficit as % of: Merchandise exports 4.9 20.7 38.0 31.2 Merchandise imports 6.2 18.7 24.1 22.5 Current account deficit - 64.5 49.0 43.9 a/ Covers all foreign exchange released to oil companies, for dividend remittances, services and oil-related imports, as well as for petroleum products. Sources: (1) Annex I. (2) EAC and Uganda Customs, Annual Trade Reports. (3) Mission estimates.  - 18 At some later date, Uganda may also find and develop its own oil resources, which potentially could have a major impact on its import requirements (see Chapter III, Section B). But even so, it seems almost certain that Uganda will remain a substantial net importer olf energy throughout the 1980s. The challenge will be to minimize the impact of these imports on the balance of payments, and to avoid the dislocations which have occurred in the recent past due to supply disruptions. Sector Organization and Institutions 1.22 The energy sector has suffered from the general institutional collapse which characterized Uganda during the 1970s. However, these prob- lems have been compounded by the absence of any one institution (or clear heirarchy of institutions) to take the lead in coordinating energy policies and investments. Indeed, as yet, there has been very little recognition in Uganda that energy concerns warrant a sectoral approach. Instead, sub- sector issues have generally been left to the operational organizations in- volved, such as the Uganda Electricity ]3oard, the Forestry Department and the oil companies. At the policy level, many decisions affecting the ener- gy sector are made by the Bank of Uganda (through its control over foreign exchange) and by the Ministry of Finance (though its responsibilities for budgetary allocations and taxation policies), without due regard for the impact on sector performance and development. The Ministry of Planning and Economic Development (MPED) has formal responsibility for considering ener- gy within the context of inter-sectoral priorities and resource require- ments, but as yet it has no staff working solely on energy-related matters. Similarly, the Ministry of Power, Posts and Telecommunications has formal responsibility for the power subsector and has proposed the establishment of an Energy Department to assume broader responsibilities for the sector as a whole. But with limited staff and resources, the Ministry's impact to date has been minimal. Moreover, responsibility for the development of petroleum and geothermal resources remairns apart with the Geological Survey and Mines Department. This weak and poorly-coordinated institutional structure is identified as one of the basic constraints on the development of the energy sector in Uganda. Recommendations for improving the sector's organization are presented in Chapter V of this Report. 1.23 Ministry of Power, Posts and Telecommunications (MPPT). MPPT was established in 1981, and has responsibilities for posts and telecommunica- tions (which are beyond the scope of this Report) as well as for power. According to MPPT, the responsibility for power is defined loosely to in- clude all energy sources, not just electricity, although this is not made explicit in any legislation or regulations. In practice, MPPT does not exercise any significant regulatory control over energy subsectors other than electricity, and there the function is largely performed indirectly through UEB. The Ministry is critically understaffed, with only four pro- fessionals (the Undersecretary, one economist and two statisticians) pres- ently assigned to the power section; five other approved positions are  - 19 - vacant.20/ Recruitment of qualified staff is extremely difficult due to the low wages offered by the civil service. For example, a starting grad- uate would presently earn only USh 2,515 (US$25 at the "first window" ex- change rate) per month, compared to salaries of up to USh 4,500 (US$45) per month offered by UEB, which also provides additional transport and housing benefits for its staff. More lucrative employment opportunities are avail- able in the private sector and overseas. An additional constraint on MPPT is its meager allocation of resources. The Ministry's total budget was increased from USh 10.8 million (actual expenditure of USh 13.7 million) in 1981/82 to USh 16.5 million in 1982/83. Given the sharp increase in costs over recent years, this budget is considered inadequate to meet the Ministry's recurrent expenditure requirements, let alone provide for devel- opment expenditure. In practice, however, the Ministry's financial posi- tion has become even more constrained, with cash releases by the Ministry of Finance limited to only USh 1.5 million during the first quarter of 1982/83. At this level, the Ministry can only meet minimal operating ex- penditures, primarily wages and salaries; the vehicle fleet is not being maintained or operated, official travel has largely been stopped, and staff have virtually no materials for day-to-day work. Under these conditions, it is not surprising that staff morale and motivation are extremely low. 1.24 Forestry Department. The Forestry Department is part of the Ministry of Agriculture and Forestry. Its primary responsibilities are to preserve and manage the country's existing forest resources, to increase reserves as needed to meet domestic and export demand, and to promote effi- ciency in the use of forest products and environmental conservation. In practice, very little attention has been paid to woodfuel issues in fores- try development; woodfuels from non-forest resources have been ignored. The Department's Forest Research Institute does, however, provide two to three-week training courses in charcoal making (for licensed producers) and also in wood preservation. The Forestry Department is reasonably well staffed, with 329 positions presently filled. Each year, about ten grad- uates are employed from Makerere University's degree course in forestry, and the Department's own Forestry College provides additional training for forest rangers and foresters. All of these training programs suffer from a lack of laboratory equipment and books. The Forestry Department's recur- rent budget was increased from USh 121 million in 1981/82 to USh 182 mil- lion in 1982/83; this is totally inadequate in relation to present costs. The capital budget -- which includes plantation expansion, equipment and vehicles -- was actually cut from USh 150 million in 1981/82 to USh 52 mil- lion in 1982/83. As a result, it is not surprising that the Department is unable to manage Uganda's 2.76 million hectares of forests effectively. In particular, staff are unable to travel to monitor consumption trends (in- cluding woodfuels), to control illegal cutting and excessive exploitation, and to promote new plantings. 20/ Total Ministry staff is presently 50, of which 14 are University grad- uates. In addition, the Ministry has six vacancies.  - 20 - 1.25 Geological Survey and Mines Department (GSMD). GSMD, part of the Ministry of Lands, Minerals and Water Resources, is responsible for activi- ties connected with prospecting and mining of all minerals, petroleum and geothermal resources. The Department is adequately staffed, with 5 mine inspectors, 33 geologists, hydrogeologists, geophysicists and mineral pros- pectors, and 17 chemists and other laboratory assistants. However, GSMD's work is seriously constrained by the loss or t.heft of laboratory and office equipment during the 1970s. Fortunately, most of the records remain in- tact. GSMD has already commissioned several surveys of the country's pet- roleum and geothermal potential but, as yet, no major exploration or devel- opment has taken place. In the case of petroleum exploration, GSMD is not well placed to draft or negotiate prospecting agreements with foreign oil companies, due to the absence of suitable petroleum legislation and a shortage of legal and commercial staff. 1.26 Uganda Electricity Board (UEB). UEB is responsible for the pro- duction, transmission and distribution of public electric supply in Uganda. Although MPPT has formal jurisdiction over UEB, this power is not strongly exercised and UEB operates with a considerable degree of autonomy. Despite the difficult country conditions of the past decade, UEB has remained a well-organized and relatively effective institution (unlike most other parastatals). This reflects its strong management and its ability to at- tract and retain well-qualified technical personnel.21/ At present, UEB has a total staff of 2,757, of which 109 are professionally qualified and 994 are unskilled; apart from the Chairman/Managing Director, all staff are Ugandan. There are no significant manpower shortages. The only potential organizational weakness would seem to be in the area of development plan- ning. Although this has not been of major significance in the past, it will become more critical in the years aLhead, as the country's excess gen- erating capacity is utilized and further expansion is required. 1.27 Under the Electricity Act of 1964, the tariffs to be charged for electricity service by UEB are to be adequate to cover operating costs (ex- cluding depreciation), debt service and normal capital expenditure. This was the case through 1973, when UEB was earnjing a rate of return on fixed assets of around 12% (see Table A1.3 of Annex I). However, in subsequent years -- with only marginal tariff adjustments, difficulties with billing and payment collection and declining demand -- UEB's financial position de- teriorated. In 1978 and 1979, UEB's accounts showed a loss. More recently the situation has improved once again, due to a series of tariff adjust- ments since 1980 (totaling about 240%) and additional revenue from sales to Kenya.22/ Even so, UEB has difficulty in providing sufficient funds to 21/ This is reflected in the fact that 68% of the professionals and 58% of the semi-skilled technicians have wiorked for UEB for more than ten years. 22/ The tariff on sales to Kenya was increased in 1980 and UEB has been able to convert the foreign exchange earnings at the higher "second window" exchange rate since August 1982.  - 21 - meet the costs of imported parts and materials for routine maintenance and is unable to make any significant contribution for capital development. This is of some concern at a time when rehabilitation and expansion of Uganda's generation capacity and transmission system is required. 1.28 Oil Companies. The importation and marketing of petroleum prod- ucts in Uganda is handled by six oil companies. Their historical market shares are as follows: Shell 30%, Total 19%, Caltex 15%, Esso 15%, Agip 14% and Mobil 8%. The Ugandan Government has an equal (50:50) shareholding in three of the companies: Shell, Total and Agip. This participation is formally exercised through appointment of the Board Chairmen and up to half of the company directors. However, in practice, control of oil company op- erations remains with the non-government partners, who appoint the managing directors and provide management through "Consultancy Services Agreements". The oil companies employ about 600 persons in Uganda, including up to 20 expatriates. All of the managing directors and finance managers, and most of the operations managers, are expatriates; Ugandan nationals hold about 15 managerial positions. Salary levels are among the highest in Uganda, and the oil companies have no difficulty attracting qualified staff.  Mable 1.5: E5TnOTED ElElEY BALANZ MR 1970 (in '000 TOE) a/ SUPPLY CONSUMPTION Total Supply Trarsmission & Net Supply T_tal Eniergy Primary After Distribution Available for Igaida Source Produrtion Imports Supply Tramfonmation Losses Consumption Household Ccmmnrce Industry Tramport Costumption Exports A. CDmercial energy Aviation fuel - 77 77 77 - 77 _ _ _ 77 77 - Gasoline - 119 119 119 - 119 - - - 119 119 - Kercsene - 41 41 41 - 41 41 - - 41 - Auto. diesel - 87 87 87 - 87 - - - 87 87 - lIndustrial diesel - 10 10 10 - 10 - - 10 - 10 - FI1 oil - 100 100 99 - 99 - _ 99 _ 99 - L. P. gas _ 2 2| 2 _ 2 2 - _ _ 2 Electricity - hydro 183 - 183 61 6 55) 6 6 23 - 35 21 - themal - - - 1 - 1) Fbilwood 867 - 867 567 _ 567 252 101 214 567 _ Charcol - - - 72 - 72 36 23 13 - 72 - Total 1,050 436 1,486 1,136 6 1,130 337 130 359 2a3 1,1C 21 Conwersion losses 1 - electricity - - 122 - fuelwood - - -| 228 Total canurcial eneiEy 1,050 436 1,486 1,486 1,130 337 130 359 283 1,109 21 B. N1n-coammrcial energ, Fuilwood 3,126 - 3,126 3,126 3,126 2,908 184 34 - 3,126 - C. lbtal energ 4,176 436 4,612 4,612 4,256 3,245 314 393 283 4,235 21 a/ See front of report for conersions to TCE. Source: Annex I.  Table 1.6: ESlJAIH) EER1Y BAIANCE F)R 1980 (in '000 TOE) a/ SUPPLY CONSUMPTION Total Supply Transmission & Net Supply Total Eneigy Primry After Distribution Available for LUarda Source Production ImpDrts Supply Trarsformation lases Coamuption Hkusebol Camerce Industry Tiansport Comuuption Exports A. Omnercial energy Aviation fuel - 13 131 13 - 13 - - - 13 13 GEsoline - 89 891 89 - 89 - - - 89 89 Kerosene - 48 48 48 - 48 29 - - - 29 19 c/ Auto. diesel - 72 72 72 - 72 - - - 58 58 14 Indiustrial diesel _ 1 1| 1 1 1 - 1 - Fuel il - 23 23 22 - 22 - - 22 - 22 - L. P. s - 1 Ij 1 - 1 1 - - - 1 - El-ctricity - hydro 159 - 159 53 9 b/ 44) 7 6 8 _ 21 24 -thenul - - - 1 - 1) Fuelooi 1,121 - 1,121 383 - 383 146 96 141 - 383 - Charcml - - - | 177 - 177 112 65 - - 177 - Total 1,280 247 1,527 860 9 851 295 167 172 160 794 57 Cowersion losses 1 - electricity 106 - fnelwax J 561 Total commrcial energy 1,280 247 1,527 1,527 851 295 167 172 160 794 57 B. No-commercial energy Fuelvooc 3,662 - 3,662 j 3,662 3,662 3,267 351 44 - 3,662 - C. Total -er 4,942 247 5,189 5,189 4,513 3,562 518 216 160 4,456 57 a/ See front of report for conversions to TCE. b Inchlues illegal consuption. c/ Lkificial exports (i.e., smgglitg). Source: Mission estinates.  - 24 -  - 25 - II. ENERGY DEMAND: OUTLOOK FOR THE 1980S 2.1 A realistic projection of energy demand is an important input for planning sector policies and investments. In Uganda, however, the task of preparing energy demand projections is complicated by two basic problems: (a) the paucity of reliable data on the present energy situation, and the distoritions introduced by the rundown state of the economy (see Chapter I and Annex I); and (b) the highly unpredictable nature of future developments in the economy, and their likely impact on energy demand. The consequent uncertainties are of course not unique to Uganda, but they are made more acute by the severity of Uganda's economic predicament. For this reason, a range of energy demand projections are presented at the end of this Chapter, related to three different scenarios on the future devel- opment of the economy. Although the broad dimensions of Uganda's longer- term energy needs are discussed, the primary focus is on the outlook for the 1980s. In preparing the energy demand projections, allowance has been made for improvements in energy conservation and for some limited energy substitution. The scope for these are discussed in the first two sections of this Chapter. The institutional arrangements for implementing measures to improve energy efficiency, and related technical assistance proposals, are discussed separately in Chapter V. A. Scope for Energy Conservation 2.2 The efficiency of energy use in Uganda is extremely low. Some degree of improvement in energy efficiency will follow automatically, as rehabilitation proceeds and capacity utilization increases. However, addi- tional savings should also be achieved by implementing relatively simple and cheap energy conservation measures, especially in the industrial and transport sectors. These conservation measures would have an immediate impact on the energy situation, by constraining the growth of petroleum import requirements and the demand on generation capacity at Owen Falls Power Station. Woodfuel requirements could also be reduced by improvements in the energy efficiency of many agro-industrial plants and possibly over the longer term through better stove designs and charcoal production. Industrial Plants 2.3 Pattern of energy consumption. The level of industrial activity is very low, capacity utilization is typically less than 10% and many fac- tories are not in operation. Consequently current energy consumptions are extremely high relative to output and do not represent a normal demand pattern. The predominant energy sources for industry are heavy furnace oil and electricity. By-products such as bagasse and cotton seed husks are used for boiler firing in the sugar and vegetable oil refining industries  - 26 - together with small quantities of fuelwood. The major furnace oil consumer is the cement industry for kiln firing, when the plants are in normal oper- ation, followed by the textile factories which use heavy furnace oil in boilers for steam raising. Furnace oil, is presently used in the steel industry for reheating purposes. Electricity in industry is used almost exclusively for motive power and pumping, although it is understood that a small electric boiler is used at Mbale. 2.4 Efficiency of energy use. The efficiency of energy utilization in industries is extremely low due to the following factors: (a) most of the industrial installations have been operating at low production rates due to shortages of raw materials and spare parts resulting from foreign exchange constraints. Energy usage is excessive when plant is operating at low throughputs or with intermittent production, especially in industries wnich are nor- mally operated continuously. Under these conditions standing losses become significant and large amounts of energy are lost from frequent start ups. As a result the specific fuel consump- tion (i.e., energy per ton of production) becomes very high; (b) the majority of the plant is in a run-down state and hence opti- mum efficiencies are difficult to achieve even with good energy management. Certain energy saving plant items are available but these have not been put into operation. For example, an effi- cient boiler has been installed but not commissioned at the Uganda Sugar Corporation plant at Lugazi since 1970, and at Pampa Textiles in Jinja a new processing plant including a boiler has been awaiting installation since 1974. A new boiler has also been awaiting installation for some years at the Uganda Tobacco Corporation plant in Jinja. A textile effluent heat recovery plant is ready for recommissioning at Pampa Textiles; (c) the existence in some cases of inherently inefficient plant, par- ticularly in the steel and cement industry; and (d) the majority of plant operators arnd supervisors are unaware of modern methods of energy management. In particular, the princi- ples of good combustion practice are not appreciated, and elec- tric power factor correction is not applied (or is not fully understood). Records of energy consumption and related produc- tion are not kept regularly in most instances, and the absence of monitoring instrumentation aggravates the situation. 2.5 Energy conservation potential. Present retail prices for elec- tricity and fuel oil incorporate a substantial subsidy in economic terms (see Chapter IV), and the Mission's recommendation that prices for these products should be increased accordingly would provide some incentive to industrialists to improve energy efficiency. The scope for improvement in energy efficiency in industrial plants is illustrated by comparing energy consumption in the major industrial plants in Uganda at different levels of production. A comparison of energy consumption at 1982 production levels with estimated energy consumption at increased production levels for these  - 27 - plants is given in Annex III based on information obtained by the Mission. The comparison is based on specific energy consumption, comprising fuel oil, industrial diesel oil and electricity, with implementation of elemen- tary energy conservation measures but without substitution of fuels (except for the replacement of fuel oil with bagasse in the sugar industry). In some cases, the situation in 1982 was totally unrepresentative of normal production conditions with existing plant and production patterns, and therefore the Mission reconstructed the 1970 situation as an approximation to normal conditions for 1982. Achievement of the postulated increased production levels in some plants would depend upon substantial investment in capital plant and equipment to increase production capacity. The greatest scope for improvement in energy efficiency, as represented by reduction in specific energy consumption, is in the sugar processing plants (by utilizing bagasse), cement plants, breweries, tanning and vegetable oil refining. Reductions in specific energy consumption of more than 50% are possible in 10 out of the 21 plants considered by the Mission, and the average reduction for all the plants is about 50%. An immediate improve- ment in energy use can be obtained through a number of relatively cheap and quick measures which are discussed below. 2.6 Improvements in fuel-firing techniques. Improvements in fuel- firing techniques in boilers and furnaces can be obtained through operator training and the provision of instrumentation. A suitable instrument kit for boilerhouses would include portable indicators for measuring levels of oxygen, carbon dioxide and exhaust temperature together with a steam flow recorder. The carbon dioxide and temperature recorders would be used to ensure that optimum air-fuel ratios are maintained. The steam recorder would permit records to be kept on steam production and fuel used, and hence act as a monitor. Typical savings of 5% to 15% are achievable through these measures. In the case of furnaces, the use of oxygen indi- cators with higher range temperature indicators should result in substan- tial energy saving. Further benefits should accrue in the steel industry furnace where better control of the combustion process will reduce scaling effects. The use of infra-red detectors for kiln skin temperature in the cement industry could prove to be beneficial in reducing downtime through refractory failure, which would lead to an improvement in fuel efficiency. In addition to these measures, management and operators will need to be trained to interpret the results to ensure that optimum air-fuel ratios, firing rates and temperature control are being achieved. For example, the efficiency of boiler plant at Nyanza Textiles could be improved by 5% by simple adjustment of the fuel and combustion air controls. 2.7 Improvements in power factors. Most of the plants run at low power factors (i.e., less than 0.7) due to the predominance of induction electric motors. In some cases correction equipment exists but its per- formance has deteriorated due to age or alternatively it is not matched to the current load. The Madhvani Sugar Factory currently operates at a power factor of 0.5, and an increase to 0.95 would reduce the maximum demand by 700 kVA. A low power factor is undesirable in that it limits the output of alternators and transformers, increases losses in the main electricity dis- tribution network and reduces terminal voltages. This situation is toler- able at the moment only because there is ample generation capacity relative to system demand. Also there is no incentive in the tariff structure for  - 28 -- improvements. However, when the total system power demand approaches UEB's generation capacity, the maintenance of good power factors would be of significant economic benefit (paragraph 4.5). 2.8 Energy conservation plant investment. Improvements in energy demands can be achieved by relatively modest investments in energy saving equipment. These include recuperators on furnaces and kilns, effluent heat recovery plant and recovery of condensate. For example, the East African Steel Corporation plant in Jinja has a reheat furnace without adequate heat recovery on the exhaust. Fuel consumption can be reduced by 25% using recuperators. Packaged energy saving equipment that is awaiting installa- tion should be installed immediately, and includes new boilers at Pampa Textiles, the Uganda Sugar Corporation arid the National Tobacco Corporation (NTC) plants. 2.9 Improvements in general energy management. Substantial long-term energy savings can be obtained by improvements in general energy management (i.e., improvements in housekeeping techniques), and typical savings of 5% to 10% are achieveable. The importance of combustion control and power factor correction has already been stressed, bat there are other aspects of energy management which will become important as industrial activity in- creases, particularly as there appears to be a lack of knowledge of modern methods. It is considered that short practical training courses for man- agement and operators could be of benefit and such courses would include: (a) the importance of monitoring arnd record keeping; (b) the establishment of energy targets related to production and other indicators; (c) regular testing of the performance of energy consuming plant; (d) maintenance of good load factors and ensuring correct-sized plant to reduce standing losses; and (e) regular review of plant performance with regard to new improved designs, fuel availabiLity and costs, fuel substitution poten- tial and plant improvement. Agro-Industrial Plants 2.10 Tea drying. Boiler efficiency varies from factory to factory, ranging from 1.5 to 3.7 kilograms of woocl per kilogram of "made tea". Ways of improving boiler efficiency, and the economaic justification for instal- ling more efficient boilers, should be investigated. Annual tea production has fallen from a peak of 23,10() tons :in 19,71/72 to about 2,100 tons at present. If the tea industry is to recover to anything like its former production, it would be most unwise not t:o dry the tea with wood, since the increased demand for furnace oil could be in the region of 15,000 tons/year (at a foreign exchange cost of about US$6.6 million annually). 2.11 Brick and tile production. Brick and tile production is an important rural-based industry. Apart from one large factory on the  - 29 - outskirts of Kampala, which has a continuous kiln and uses coffee husks as fuel, all the other factories use wood as their energy source. Most of the kilns are small cottage industry operations but there are a few large kilns, for example, the Butende brick works just outside Masaka. On the road between Kampala and Masaka there are at least 42 brick-making units principally concentrated within a 20 kilometer radius of Masaka. From a cursory investigation, it appears that the Butende brick works uses about 0.60 kilograms of wood per kilogram of burnt clay and the small kilns use 0.65 kilogram of wood per kilogram of burnt clay. The quality of the out- side bricks in the small kilns is poor. Kiln design and burning efficiency could be improved without too much effort if the operator is given training and advice. 2.12 Butende brick works obtain their wood from the surrounding area but pay on average USh 1,600 per ton (USh 1,100 per m3 solid) delivered at the factory of which USh 1,200 is transport charges. With the present kiln design the factory requires about 6,000 tons per year of wood, which means they are paying USh 7.2 million per year in transport charges. Much of this transport cost (and related energy requirements) could be saved if the factory developed fuelwood plantations next to the factory. The wood could be transported to the factory by a sulky (an inverted U-frame on wheels) drawn by hand, so liquid fuel would not be required for transport. The financial and economic returns to fuelwood plantations are examined in Chapter III (paragraphs 3.44 to 3.46). Expansion of fuelwood plantations would be economically justified if the economic cost of fuelwood production is less than USh 1,100 per m3 solid, for which the opportunity cost of land used (expressed in terms of the economic returns to displaced crops) would have to be less than USh 20,000/hectare. The returns to bananas and coffee are substantially higher than this and therefore these crops should not be displaced. However, the use of more marginal land and existing wooded areas near to the brickworks would be economically justified. At the present kiln efficiency, the factory would require about 240 hectares of plantations to be self-sufficient, but less area would be needed if the kiln efficiency was improved. A detailed study should be undertaken into the supply of fuelwood from marginal land as well as means of improving kiln-burning efficiency. 2.13 Tobacco curing. Like the tea industry, tobacco growing declined substantially during the 1970s. At present, annual production is about 700 tons of which 500 tons is fire-cured and 200 tons is flue-cured tobacco. In 1971/72, 5,000 tons of tobacco was produced, 1,800 tons being fire-cured and 3,200 tons flue-cured. In 1982, about 13,000 tons of wood were used to cure tobacco whereas in 1972 about 140,000 tons were required. There are already 5,000 hectares of fuelwood plantations in the forest reserves for tobacco curing, and this should be sufficient to meet the 1972 output if it is in the right place. Savings in fuelwood consumption could be made by improving barn design and by increasing barn size. With the present barn design and curing techniques, the tobacco from a one acre plot and cured in a "one-acre barn" requires 25 to 30 kilograms of wood per kilogram of flue- cured tobacco, and about half that amount is required for fire-cured tobacco. In Kenya, some farmers are using about half this quantity of wood in a "one-acre barn" and in Malawi consumption is even less. Also, the larger the barn the less wood is used per kilogram of tobacco cured. If  - 30 - tobacco production from one hectare could be cured in a "one-hectare barn", then there would be a saving of about 25% in wood consumption compared to curing in a "one-acre barn". 2.14 Fish smoking. Fish smoking is a traditional method of preserving fish and improving iits flavor. The livelihood of many people, especially along the lakes of Uganda, depend on fishing yet some fishermen cannot obtain enough wood to smoke their fish with the consequence that they are only able to sell their catch locally. The Government could assist the fishermen by establi'shing plantations near to the catching grounds and by looking into the design of the fish smoking barns. Traditionally fish is smoked on wooden racks with a fire underneath. If brick hearths were introduced with the fish suspended on rods inside a partially closed brick curing chamber, then there could be a considerable saving of energy. The cost and efficiency of such fish smokers should be investigated to see if improvements are worthwhile and affordable by the fishermen. Transport 2.15 Railways. The 1982 consumption of diesel oil by the railways was about 4 million liters, equivalent to about 7% of total national consump- tion of diesel oil. The efficiency of energy use in the railways is low due to inadequate maintenance arising from shortages of spare parts, lack of suitable diesel-testing equipment, and poor condition and grading of track. Locomotives are operated outside normal servicing limits. There is a lack of suitable repair facilities and inadequate engine testing facili- ties since the workshop facilities at Tororo and Kampala are suitable for minor repairs only. Some diesel-testing equipment is in operation but is inadequate for the whole locomotive fleet. A testing rig has been deliv- ered to Kampala but has not yet been installed. The last major repairs to tracks were carried out in 1970 and speed restrictions are required (par- ticularly between Kampala and Kasese). Steep gradients, particularly be- tween Jinja and Kampala, necessitate the inefficient use of two locomotives instead of one. A proposal to regrade major routes would cost an estimated US$30 million, and the Government is seeking external assistance for this purpose. 2.16 The potential order of savings from provision of suitable repair and diesel-testing equipment would be 5% to 10%, with an economic value of US$120,000 to US$240,000 annually, provided that adequate skilled manpower and spare parts are made available. The testing equipment delivered to Kampala could be installed at minimal cost. The capital cost for imple- menting these measures would be less than US$100,000 and therefore is an investment that should be carried out immediately. 2.17 Roads. The situation regarding energy usage in the major road transport companies is similar to that in the railways, namely low energy efficiency due to lack of spare parts, inadequate workshop facilities and skilled mechanics, the poor state of the roads and high speed driving. Servicing is probably better in the pr:Lvate sector than in the public fleets due to the better availability of spare parts (via the black market) and ability to attract the best mechanics. In the public sector, the main operators are the Uganda Transport Company (UTC) and the Uganda Cooperative  - 31 - Transport Union (UCTU). The fleet of 100 buses operated by UTC is only about two years old and the availability of buses was reported to be about 70%. The company has had to restrict operations due reportedly to short- ages of diesel fuel. A similar level of vehicle availability was reported by UCTU which operates a relatively modern road haulage fleet of 150 vehi- cles primarily for transporting agricultural export products to Mombasa. UCTU's workshop facilities are very basic and include some testing equip- ment. UCTU does not have modern workshop facilities, and new machinery and equipment which was bought in 1975 has not yet been put into operation. 2.18 There is scope for energy savings of between 5% and 10% of diesel fuel consumption, equivalent to 2,500 to 5,000 tons per year at the 1982 consumption level (worth about US$1.5-2.5 million annually). To achieve these savings the following measures are required: (a) spare parts must be made available; (b) the existing diesel-testing equipment needs to be improved and extended, in order to increase the availability of vehicles and permit proper tuning of engines; and (c) training of drivers and mechanics to be conscious of energy effi- ciency. 2.19 There are no data available on energy consumption by private sector vehicles which account for most of the gasoline consumption. Government has provided an incentive to improve fuel efficiency by keeping gasoline prices above import-parity prices (Table 4.7). The poor state of the roads undoubtedly increases fuel consumption. However, the justifica- tion of a highway rehabilitation and maintenance program is beyond the scope of this energy assessment. Households and Commercial Establishments 2.20 Cooking accounts for about 80% of household fuel, and the house- hold sector is the dominant end user of energy accounting for 80% of total energy consumption (Table 1.6), practically all of which is woodfuel. Many restaurants, institutions and tea houses also use charcoal and wood to cook food and beverages. Most of them are using stoves with similar efficien- cies to the present household models. A number of countries in the region (e.g., Kenya) have recently started programs to improve stove efficiency. Uganda should follow these programs closely. To date, progress has been limited and even the potential for achieving higher levels of stove effi- ciency in everyday use remains unclear. Nevertheless, further development and promotion of better stove designs is justified by the potential longer- term impact on energy demand, especially for woodfuels. Further savings could be realized by reducing the proportion of wood raw material lost in the process of conversion to charcoal,1/ through promotion of better 1/ Energy lost in conversion of wood to charcoal is estimated to be equivalent to about 12% of total Ugandan energy consumption in 1980 (Table 1.6).  - 32 - wood-burning (rather than charcoal-burning) stoves and improved charcoal production techniques. 2.21 Woodfuel cooking stoves. The three-stone fire is the typical fuelwood stove, and the metal stove (sigiri) is the common charcoal device. The efficiencies of these stoves reportedly average 13% and 20% respective- ly. The three-stone fire has advantages in that it does not require an initial outlay of cash and it serves other purposes besides cooking, such as lighting and smoking the roof to keep it insect free. More efficient stoves, usually with a burnt-clay component, are already in use in other countries and could potentially be introduced into Uganda. Depending on conditions, efficiencies of up to 25% for a mud wood stove and 35% for a burnt-clay charcoal stove could possibly be achieved. However, for these stoves to be acceptable to the consumer, they must be portable, easy to use and low cost. Particular attention needs to be given to use of local mate- rials and the development of simple and non-customized designs. Investiga- tions should also be undertaken into alternative fuel options, including briquetting of charcoal powder/fines, sawdust and crop residues, and com- pressed and pre-dried wood. It is essential that Uganda learns from, rather than duplicates, the ongoing research :Ln these areas in other coun- tries. In this way, greater emphasis can be given to the practical prob- lems of producing and marketing improved stoves that are acceptable to Ugandan consumers. Related issues include materials testing and procure- ment, training of artisans in stove making, quality control, the organiza- tion of marketing networks and publicity. 2.22 Charcoal production. At present, most if not all charcoal is made in the earth or pit kiln. Yet Uganda ploneered portable metal kilns and also tested semi-permanent brick kilns. Unfortunately, these methods were abandoned during the troubles of the 1970s, but the production tech- niques and knowledge have not been lost. On average, it takes about 8.4 tons of air-dried wood to produce 1 ton of charcoal in an earth kiln, 5.8 tons in a metal kiln and 4.5 tons in a brick kiln, so considerable savin s of wood raw material are possible by switching to metal or brick kilns._/ However, these kilns are only semi portable and must have a guaranteed supply of wood near to the production unit. If the wood has to be trans- ported over distances of more than 10 to 20 kilometers or if the kilns have to be constantly moved to new supply sources, then these kiln types cannot compete with earth or pit kilns. 2.23 Much charcoal is made in open woodlands and rangeland areas, and the earth kiln is ideal for these areas, especially since it does not involve a capital cost. However, if plantaticns are grown specifically for charcoal production or if natural forests are being cleared or managed, 2/ If wood could be burnt directly in efficient stoves or boilers then even more energy could be saved. But charcoal is a more convenient fuel than wood, has twice the energy value on a weight basis and thus can be eco- nomically transported over longer distances. For wood to compete with charcoal, it must be grown near the consumers and if sold, cut into con- venient pieces and marketed properly.  - 33 - then brick (or steel) kilns should be used. Some earth kiln operators are far more efficient at charcoal production than others, and their techniques should be studied and courses organized in earth kiln production. The brick and metal kiln techniques should be revived and applied where appro- priate. Charcoal briquetting methods could be investigated and/or markets for powder and fines sought out (e.g., as furnace or kiln fuel or as a feedstock for calcium carbide production), since about 20% of charcoal production is in the form of charcoal powder and fines. Charcoal retorts are even more efficient than brick kilns and they produce by-products such as carbon monoxide, acetic acid, and wood tar. Unless these by-products can be sold however, the cost of charcoal production by retort is too high, especially in terms of foreign exchange. 2.24 One of the National Research Council's projects in cooperation with the Forestry Research Centre is in the field of charcoal production and utilization. Work is being done on brick kilns, but the charcoal research could be expanded to look at earth and pit kiln methodology and ways of improving it, and the production of charcoal from agricultural residues such as rice and coffee husks. A survey should also be undertaken of potential sustainable supply and plantation area equivalents. Power Transmission and Distribution Losses 2.25 Present losses on transmission and distribution of power in Uganda offer considerable scope for energy conservation. Even after dis- counting the extraordinarily high level of losses during the troubled year of 1980 (45% on domestic sales), the level of losses has been running at about 20% of domestic sales. The losses on transmitting power for export to Kenya are presently only about 4%. A large proportion of the losses, probably about 15% of domestic sales, occurs in local distribution systems and is the area with the greatest scope for energy conservation. Urgent rehabilitation works are scheduled With ODA and IDA assistance. Additional rehabilitation will be required to reduce distribution losses to acceptable levels, and the extent of such works are included in the terms of reference for the study into uprating and rehabilitating Owen Falls Power Station (Annex IX). Improvements in power factors, especially for major industrial consumers, will also help to reduce distribution losses ,paragraph 2.7). B. Scope for Energy Substitution 2.26 The feasibility and economic viability of proposals for fuel substitution in industry should be studied on a case-by-case basis, taking into account projected production levels and the associated costs of energy investments. The Mission's preliminary findings suggest that there are several industrial and agro-industrial plants in Uganda in which there is a strong economic justification, as well as financial incentive, for fuel substitution. The form of substitution is usually fuel oil Dy woodfuel although a prerequisite would be a guaranteed source of supply of woodfuel which would require the establishment of plantations or designation of specific forest areas for commercial management. Other options for fuel  - 34 - substitution include switching to a heavier grade of fuel oil and the use of in-house agricultural wastes such as bagasse and coffee husks. The electrification of industrial boilers is unlikely to be economically justi- fied before the addition of a new hydroelectric power station, which would be in 1991 at the earliest (paragraph 2.32). The replacement of small diesel generators in isolated areas by hydroelectric power from extensions to the main transmission network is economically attractive. Other renew- able energy forms are theoretically potential substitutes, but in practice could not make a significant impact on the national energy situation and generally have yet to be proved technically feasible and economically attractive. This category of fuels includes calcium carbide for lighting as a substitute for kerosene in rural areas, producer gas (biogas) for engines in isolated areas as a substitute for diesel oil, and papyrus as a substitute for steam-raising fuel, especially woodfuel. Some of these alternatives are covered briefly at the end of Chapter III. Electrifica- tion of the railway system would not be economically justified at present or historical traffic levels, even though there is relatively low-cost hydroelectric potential in Uganda. Conversion to Woodfuels from Oil Firing 2.27 Industrial plants. New boilers would be required in industrial plants for conversion from oil to woodfuel firing since modifying the existing oil-fired boilers would be impracticable. In the case of the cement plants only new firing and handling plant would be required. One cement factory at Tororo used to use charcoal imported from Kenya but now it uses oil and is potentially one of the largest consumers of oil. From an economic point of view, it would be more profitable if both the cement factories changed to other fuels. It is possible that charcoal produced from coffee husks, or even dried papyrus grass, could be used if the cost of producing and transporting these fuels to the factory is not too great. However, the cheapest solution may be to establish fuelwood plantations near to the factory and either produce charcoal or burn densified wood produced by a process of drying and chipping. The economics of converting the Tororo cement plant to charcoal-burning boilers, and the conversion of the boilers at Nyanza Textiles back to wood burning, is examined in Chapter III (paragraph 3.45). Similar potential for conversion exists at the Uganda Baati metalworking plant. 2.28 Tea drying. One tea estate reportesd that it cost USh 1.2 for fuelwood and USh 50 for furnace oil to dry 1 kilogram of "made tea". The cost of growing and cutting the wood was probably not included in the above price of USh 1.2. On average it takes about 2.5 kilograms of wood to dry 1 kilogram of "made tea" (5 kilograms of green tea), so if the cost of growing and cutting the wood was included it would have doubled the price to about USh 2.4 per kilogram of dried tea, which was still some 20 times less than the cost of oil. These cost savings would justify the use of tea-growing land for woodfuel plantations (Table 3.5). The Government should therefore encourage those tea estates t:hat have oil boilers to con- vert to wood and indeed they should do this in any case as a profitable cost-saving measure. Most tea estates already have access to woodfuel plantations, and new plantations are to be established with assistance from the EEC.  - 35 - 2.29 Coffee roasting. There are two coffee roasting factories in Kampala. One of these companies is the Kaawa Kawomera Nabuka Company which uses 500 liters of furnace oil per month, costing USh 25,000. If charcoal was substituted for furnace oil with slight modifications to the roasting oven, about 600 kilograms of charcoal would be required. At present prices this would cost about USh 10,200/month, so the company could save more than half of its fuel bill and the country could save foreign exchange. How- ever, for companies to change to charcoal they must be guaranteed supplies of the fuel. The economic value of charcoal as a substitute for fuel oil is sufficient to justify the displacement of any major food or export crop for woodfuel production (see Tables 3.5 and 3.6 in Chapter III). The Government or the firms themselves should establish plantations for char- coal production, and such a company using 600 kilograms of charcoal a month would only require two hectares of plantation to be self-sufficient. Heavier Grades of Fuel Oil 2.30 The present grade of furnace oil used in industrial plants is 1,000 secs. With the exception of the steel industry, it would be feasible to convert the majority of the fuel oil processes to heavier residual grades of oil (1,500 and 3,500 secs) without technical difficulties or high investments as the majority of the boilers are designed for this grade. The handling plants (tanks, pumping pipework, heaters, etc.) would have to be modified to accept residual grades. New infrastructure, including new transportation and storage facilities, would be necessary to handle residu- al grades over 2,000 secs. The cost of residual grades is about US$10/ton less than the cost of heavy fuel oil. A substantial reduction in demand for heavy fuel oil would have an immediate impact on the product mix of Ugandan demand which could cause an imbalance in supplies of products from the Mombasa refinery. Therefore, the implications of a reduced Ugandan demand for heavy fuel oil should be examined before any major switch in fuel is considered, particularly with regard to the possible price effects on other petroleum products. Use of Bagasse 2.31 Efficient sugar factories and jaggeries should be self sufficient in boiler energy from the sugarcane waste (bagasse). Indeed, some modern factories in other countries have an excess of boiler fuel and produce surplus electricity for the use of their employees and to feed into the electrical grid system. However, in Uganda some factories are using wood to supplement their bagasse, some are using oil and even others are not using their bagasse at all but using wood. Unless the bagasse is going to be used for mulching, there should be no reason to use wood or oil in the production process of sugar and jaggery if efficient bagasse-burning boilers are installed. Plans for rehabilitation and modernizing both major sugar plants in Uganda have been prepared and financial support is being considered by bilateral and multilateral financing agencies. Under these plans, one plant would become entirely self-sufficient in energy and the other plant would be largely self-sufficient.  - 36 - Electrification 2.32 Industrial boilers. In general it is technically feasible to convert boiler plants from fuel oil to electricity although it requires very substantial electrical capacity to replace even modestly sized boiler plants (the Nyanza Textiles boilerhouse alone would require more than 21 MW). It is not practical to modify the existing boiler shells and there- fore it would be necessary to install new boi.lers and provide transformers (except in the case of Nyanza Textiles which was converted from electricity to fuel oil in 1969). In general, electrification of boilers would be economically unattractive at the relatively low levels of fuel consumption and plant capacity utilization projected for the 1980s when "economic" electricity prices are used (see Chapter IV, Section A), even when maximum demand charges are iLgnored or are very low (als in the current tariffs). However, at current prices electrification may offer good financial re- turns. In view of the need to install new electric generation capacity for the national system as soon as possible to meet projected national growth in demand without electrification of boilers (see Chapter III, Section A), any proposal to convert industrial boiler plants should be carefully evalu- ated at economic prices, particularly for the foreign exchange costs of conversion and for electric power. It is un:Likely that electrification of boilers would be economically justified before the addition of power gener- ation capacity from a new hydroelectric power station, which would be in 1991 at the earliest. 2.33 Household cooking and lighting. Analysis of economic costs of energy forms in Uganda (see Chapter IV, Section D) shows that electricity is a lower-cost option for lighting in urban households than kerosene and L.P. gas, and it is recommended that the Government and UEB should encour- age the use of electricity for this purpose by means of a low "lifeline" tariff rate for a given consumption of electricity by households (paragraph 4.11). The analysis also shows that electricity is an economic option instead of woodfuels, for meeting increased energy demand for household cooking in the main urban areas of southern Uganda. A substantial number of urban households are already connected to the electricity supply net- work, and therefore a significant degree of substitution of woodfuels by electricity could be achieved without being constrained by capacity to connect new consumers. However, for cooking purposes, heavier wiring than is presently installed may be required and ut:ilization of electricity may be constrained by the availability and cost of stoves and rings. 2.34 Isolated load centers. UEB have prepared plans to extend the main power transmission network to isolated load centers to displace local diesel generators with electricity supplied from the Owen Falls Station and to connect areas which presently have nc, access to electricity (para- graph 3.10). Because the alternatives are to continue operating the diesel generators or to install new ones, this program will substantially reduce the costs of electricity supply to these areas (paragraph 4.4). The ex- tended service will provide an additional energy option for households, coffee and tea processing plants, maize mills, saw mills and water pumping stations.  - 37 - C. Economic Projections and Energy Demand Economic Projections 2.35 Given the country's rich productive base, and the present low levels of capacity utilization in all sectors of the monetary economy, Uganda has the potential for relatively rapid growth in the years ahead. The economy's growth prospects are particularly good over the next two to three years, when the payoff from the Government's ongoing policy reforms and rehabilitation efforts should be realized. However, despite the unde- niable progress made over the past two years, a number of basic constraints to economic growth remain. First among these is the unstable security situation, which continues to disrupt economic activity and necessitates a diversion of scarce resources to military and police operations. On the economic front, there are a number of policy and institutional weaknesses which need to be addressed. These include: (a) further adjustments in the exchange rate and producer prices to provide adequate incentives for export development; (b) improvements in the allocation of foreign exchange to ensure that proper priority is given to the recurrent requirements of the productive sectors and supporting infrastructure; (c) strengthening of the Government's administrative capacity, espe- cially for tax administration, expenditure control and aid coor- dination (a basic prerequisite for this is improvements in the wages and working conditions of civil servants); and (d) decisions on an appropriate structure for the parastatal sector, on which public organizations should be closed down or priva- tized, and on measures to ensure the viability of the remaining parastatals. Finally, although both export earnings and aid disbursements have increased sharply over the past two years, critical shortages of foreign exchange continue to hamper Uganda's economic recovery and no significant improve- ment in the balance of payments is expected for the foreseeable future. This underlines the importance of substantially higher levels of external assistance to help finance the import requirements of rehabilitation and economic development. Without this assistance, Uganda's productive poten- tial will remain underdeveloped. 2.36 The base case projections for economic growth, together with higher and lower growth scenarios, are summarized in Table A2.1 of Annex II.3/ In the base case, GDP is projected to grow by 5.1% per annum during 3/ The economic projections have been prepared using the World Bank's Revised Minimum Standard Model. For a summary of the major assump- tion, see Annex II.  - 38 - the rehabilitation period (1983-85), led by recovery in the monetary agri- culture and industry sectors. Subsequently, GDP growth falls to a lower but sustainable trend of 3.7% per annum.4/ The viability of this projec- tion is dependent upon three key assumptions. First, it is assumed that the investment rate remains around its present level, estimated to be only 7% of GDP, through 1985. This, of course, imnplies an exceptionally low incremental capital-output ratio (ICOR), a reflection of the opportunities under present conditions to increase effective capacity with relatively small investments in rehabilitation, and to imnprove overall capacity utili- zation through increased recurrent imports andl other measures. Subsequent- ly however, it is assumed that the investment rate is increased, reaching 12% by 1995. This would restore the ICOR to a more normal level of 3.2. Second, export volume is projected to rise on average by 8.1% per annum over the next three years, supported by the recovery of export crop produc- tion and the development of some non-traditional items (e.g., maize exports to Tanzania). However, such rapid growth could not be sustained, given projected slowdown in economic growth and the poor market prospects for many of Uganda's exports (especially coffee, which accounts for 95% of export earnings). Therefore, the export volume growth rate is assumed to fall below 5% per annum after 1985. Finally, the projections are dependent upon substantially higher inflows of externa]L assistance. Although the current account deficit does not widen significantly in real terms, in- creased aid commitments will still be required to replace dwindling IMF resources (the net transfer, allowing for charges and repurchases, is expected to be negative by 1985) and to finance accumulated debt service payments. 2.37 The high case assumes more rapid population growth (an important determinant of household energy demand) and takes an optimistic view of the prospects for economic recovery, with GDP growth close to 8% per annum during 1983-85 and remaining above 4% per annum during the 1990s. Again, these projections are critically dependent upon the economy's capacity to increase the investment rate after 1985 and to sustain a longer-term im- provement in export performance. Note that the combination of more rapid export growth and more efficient use of foreign exchange (in part due to the opportunities for substituting imports with domestic supplies) helps keep the current account deficit to manageable levels. Indeed, external assistance requirements in the high case are no higher than in the base case through 1990. The low case, on the other hand, represents a pessimis- tic scenario with no acceleration of economic growth during the rehabilita- tion period and only marginal growth in per capita incomes after 1990 (even assuming slower population growth). Despite the somewhat optimistic as- sumption on export performance in this scenario, external assistance re- quirements rise to unrealistic levels. 2.38 These three alternative scenarios have been developed to assess the likely framework for energy demand. However, they are by no means 4/ The base case is less ambitious than the Government's own projections contained in the Recovery Program (GDP growth of 10% per annum in 1982/83 and 1983/84) and the earlier Ten-Year Action Program (GDP growth of 5.6% per annum for the remainder of the 1980s).  - 39 - exhaustive of the range of possibilities. On the optimistic side for example, present investment plans, if realized, could lead to a several- fold increase in the output of a number of major industries (e.g., sugar, fertilizer, cement and beer) during the 1980s. This in turn would generate an industrial growth rate far in excess of the 15% per annum assumed in the high case. The implications for energy planning should be considered in any proposals for development of energy-intensive industries. However, as it is unlikely that all or even most of these investments could in practice be implemented given the foreign exchange constraint, the prospect of very rapid industrial growth has been excluded from the projections used in this Report. Similarly, the projections in the low case, while pessimistic in relation to the country's productive potential, still represent a substan- tial improvement over historical performance. Without continued progress in the three key areas outlined in paragraph 2.35, the adverse trends of the 1970s could quickly re-emerge as instability and economic decline combine to create a downward spiral. Although this possibility cannot be discounted, in this event the basic institutional strengthening necessary for managing the energy sector would also not materialize, thus negating any attempt to improve energy policies and planning. Energy Demand Projections 2.39 Historical trends are considered to be poor indicators of future performance in the energy sector due to the abnormal conditions which resulted in economic decline during the 1970s and to the differences in economic circumstances between the 1960s and 1980s, particularly the de- cline in Uganda's external terms of trade, the poorer environment for over- seas investment, and the deterioration in the country's capital stock. The Mission has therefore approached the preparation of energy demand projec- tions in two ways. For the short term (through 1985), the energy demand projections take into account the prospects for rehabilitation and the ob- jectives of the Government's Recovery Program (see Chapter VI, Section A). For example, in the electricity demand forecasts (Table A2.4) explicit allowance has been made for the reopening of certain industries, such as the fertilizer plant at Tororo, the paper factory and the Kilembe copper mine. In addition, the short-term projections assume that action will be taken on the simple conservation measures in the industrial and transport sectors proposed in Section A above and that prices are adjusted to reflect economic costs (see Chapter IV). These price adjustments are expected to encourage energy conservation but not lead to any major changes in the relative costs of the three primary energy forms. In particular, electri- city is expected to remain the cheapest energy source for most household uses in urban areas, and therefore it is projected that there will be sub- stitution towards electricity for these purposes. However, no explicit al- lowance has been made for major energy conversions in industrial plants.5/ 5/ As discussed in Section B above, some industrial conversions from petroleum products to woodfuels and, in the 1990s, electricity may be economic. However, this will depend very much on projected production levels and the associated costs of energy investments. Further study on the feasibility of these proposals is therefore required.  - 40 - These short-ter.m energy demand Drojections have been related to appropriate sectoral and population growth rates Lo de:rive the demand elasticities shown in Table A2.2s For the longer term, assumptions have been made on appropriate elasticities) ta'zing into account UIgandan conditions and expe- rience in other countries, and energy demand derived accordingly. Note that both the short and long tem demand elasticities for commercial energy have been adjusted between scenar-os. This is intended to reflect the greater scope for improvements in energy efficiency in the industrial and transport sector at higZher rates oft rowth. and substitution away from woodfuels towards kerosene, L.P. gas and electricity in household use at higher income levels. 2.40 The projected growth rates of energy demand in the base case for the periods 1982-85 and 1985-90 are sumDmrized in Table 2.1. An illustra- tive projection to the year 9000 is also included. The imp'lications of the alternative assumptions on.1 economic performance, population growth rates and demand elasticities are summan,rized in Table 2.2. None of these projec- tions should be interpreted aS precise forecasts; they are simply intended to illustrate a range of possible developments and to highlight the major issues facing Uganda's energy sector. 2.41 Total energy demand. In the base case, total energy demand is projected to grow by 3.2% per annum through 1.985 and then continue growing by 2.8% per annum in subsequent years. This implies very little increase in per capita terms, despite rising per capita incomes, due to the continu- ing dominance of non-commercial energy use and, to a lesser extent, the impact of conservation measures on coD.mercial energy use. Even in the alternative scenarios, thne zrowth. rate of total energy demand remains with- in a relatively small range of 2 .5% tD 4. 0% per annum. However, these growth rates are still substantially higher than actually achieved during the 1970s (Table 1,2) and the cumulative impact on energy requirements is significant. By the year 2000, for example, energy demand in the base case is projected to be 68% higher than at present, 2.42 Commercial energy demand. In the base case, the growth rate of commercial energv demand is prolected to reacht 6.3% per annum during the period 1982-85, and then decline to about 4.7% per annum in subsequent years. These growth rates reverse the declirni.g trend of the 1970s but are comparable to the average gro-wth rate of 7.0% per annum recorded during the late 1960s (Table 1.2). The overall elasticil:y of commercial energy demand with respect to monetary GDP iLs projected at about 1.1 which is lower than estimated for the 1965-70 period (1.3) and alsco lower than presently esti- mated for most other developing countries. This reflects the scope under Ugandan conditions for improving specific energy consumption with increased production from existing capacity and through conservation measures. Per capita demand for commerc i4 ene:rgy is projected to rise from 57 kilograms of oil equivalent in 1982 to 83 L'4lograms of oil equivalent by the year 2000. However, this Is still 27%' less than in 1970. Even in the high-case projections, per capit:a demand ;or commercial energy does not recover to the 1970 level by 2000. The range of projected growth rates for commercial energy demand shown in Table 2.1 'Lead to cumu]ative increases over the next 18 years of from 115% to 203%. lThese increases are obviously substantially  - 41 - Table 2.1: PROJECTED TRENDS IN ENERGY CONSUMPTION a/ (for base-case assumptions) Growth rate (% per annum) Composition (% of total) b/ 1982-85 1985-90 1990-2000 1982 1985 1990 2000 A. Commercial energy 6.3 4.7 4.8 100.0 100.0 100.0 100.0 By source - Petroleum 5.8 4.4 4.4 20.8 20.5 20.2 19.5 - Electricity 11.6 6.2 6.2 3.2 3.7 3.9 4.3 - Fuelwood 6.1 4.7 4.9 51.9 51.6 51.6 51.9 - Charcoal 6.3 4.7 4.9 24.2 24.3 24.3 24.4 By end use - Household 5.6 5.4 5.6 40.3 39.6 40.9 44.1 - Commerce 6.9 4.2 4.3 14.8 23.6 23.1 22.0 - Industry 7.3 4.2 4.0 23.1 22.5 21.9 20.3 - Transport 5.1 4.4 4.4 21.8 14.3 14.1 13.6 B. Non-commercial energy c/ 2.6 2.4 2.3 100.0 100.0 100.0 100.0 By end use - Household 2.4 2.4 2.2 88.6 88.2 87.9 87.0 - Commerce 4.0 3.0 3.0 10.1 10.6 10.8 11.6 - Industry 4.2 2.9 3.0 1.2 1.3 1.3 1.4 C. Total energy 3.2 2.9 2.8 100.0 100.0 100.0 100.0 By source - Petroleum 5.8 4.4 4.4 3.4 3.7 4.0 4.7 - Electricity 11.6 6.2 6.2 0.5 0.7 0.8 1.0 - Fuelwood 3.0 2.7 2.6 92.0 91.3 90.4 88.5 - Charcoal 6.3 4.7 6.1 4.0 4.4 4.8 5.8 By end use - Household 2.7 2.7 2.6 80.6 79.4 78.6 76.8 - Commerce 4.9 3.4 3.5 12.3 12.9 13.3 14.1 - Industry 6.7 3.9 3.8 4.6 5.1 5.4 5.9 - Transport 5.1 4.4 4.4 2.5 2.6 2.8 3.2 D. Per capita energy consumption 341.6 346.1 347.3 348.8 (kg of oil equivalent) Commercial 56.6 62.6 68.5 83.4 Non-commercial d/ 285.0 283.5 278.8 265.4 E. Average elasticities Commercial energy with respect to monetary GDP 1.1 1.1 1.2 Total energy with respect to total GDP 0.6 0.8 0.8 a/ This table is derived from domestic consumption data in TOE, adjusted for transformation and losses. All exports are excluded. b/ Percentages show composition of commercial energy for Section A, non-commercial energy for Section B and total energy for Section C. c/ All fuelwood. d/ The declining trend reflects the fact that an increasing proportion of the population is moving into the monetary economy and using commercial energy. Source: Mission estimates.  - 42 - Table 2.2: ALTERNATIVE PROJECTIONS OF' ENERGY CONSUMPTION a/ Cumulative increase Growth rate (% per annum) over 1982 level (%) 1982-85 1985-90 i1790-2000 1985 1990 2000 A. Base case Commercial energy 6.3 4.7 4.8 20.1 50.9 141.9 - Petroleum 5.8 4.4 4.4 18.5 46.5 126.8 - Electricity 11.6 6.2 6.2 41.7 87.5 225.0 - Woodfuels 6.2 4.7 4.9 19.7 50.6 124.1 Non-commercial energy 2.6 2.4 2.3 8.1 22.0 53.1 Total energy 3.2 2.9 2.8 10.0 26.8 67.8 B. High case Commercial energy 9.8 6.8 5.1 32.3 83.7 202.8 - Petroleum 8.1 6.4 5.1 25.5 73.2 182.2 - Electricity 24.0 14.5 8.0 91.7 275.0 708.3 - Woodfuels 9.5 6.4 4.9 31.5 79.0 187.3 Non-commercial energy 2.7 2.7 2.6 8.4 24.0 59.5 Total energy 4.0 3.6 3.2 12.4 33.9 83.3 C. Low case Commercial energy 4.8 4.8 4.0 15.2 45.4 115.3 - Petroleum 4.4 4.4 3.7 13.4 40.8 101.9 - Electricity 6.1 6.1 3.0 20.8 62.5 116.7 - Woodfuels 4.9 4.8 4.2 15.3 45.7 118.8 Non-commercial energy 2.5 2.3 2.0 7.7 20.7 47.7 Total energy 2.9 2.8 2.5 8.9 24.8 58.9 a/ Based on alternative economic and populaltion projections presented in Annex II. Source: Mission estimates.  - 43 - higher than for total energy demand and have major implications for the discussion of supply options in Chapter III. Furthermore, there is rela- tively little difference between the projected growth rates of commercial energy demand in the base and low cases, and the projected elasticity of demand for commercial energy is higher in the low case than is the base case. This trend indicates that it would be extremely difficult to reduce the growth rate of commercial energy demand to less than 4% to 5% per annum for the foreseeable future, without serious implications for economic performance. 2.43 Petroleum products. Generally, demand for petroleum products is projected to increase at the lowest rate of all commercial energy forms in recognition of the heavy burden on the economy of the cost of petroleum product imports. In the base case, growth in demand for petroleum products is projected to decline from 5.8% per annum during 1982-85 to 4.4% per annum thereafter. These rates compare with an historical growth rate for 1965-70 of 16.2% per annum. During that period monetary GDP grew at 5.3% per annum (Table 1.1) compared to projected growth rates of 5.8% per annum for 1982-85 and 4.1% per annum thereafter. Clearly the projections display a much lower propensity to consume petroleum products than during the last historical period of economic growth. This decrease is due to the scope for improving energy efficiency in the industrial and transport sectors and to projected substitution of petroleum products by electricity and wood- fuels in the household, commercial and industrial sectors. Nevertheless, it appears that increases in demand for petroleum products are unavoidable even with these measures, as shown by the insensitivity of this demand to economic growth between the base case and low case (Table 2.2). As a result, the petroleum import bill (at 1982 prices) is projected to continue rising from US$100 million in 1982 to at least US$142 million by 1990 in the low case and up to US$175 million in the high case. 2.44 Electricity. Generally, Ugandan demand for electricity is pro- jected to increase at substantially higher rates than for any other energy form, which reflects the availability and potential for low-cost hydroelec- tric power. This trend assumes that there will be some substitution of other energy forms by electricity in the manner described in Section B above. This trend also assumes that expansion of UEB's generating capacity will be feasible, as outlined in Chapter III, and that the required invest- ment resources will be available. In the base case, growth in Ugandan demand for electricity is projected to decline from a high rate of 11.6% per annum in 1982-85 to 6.2% per annum thereafter, compared to an average growth rate of 4.3% per annum in 1965-70 when the policy was to switch to use of petroleum fuels. The sensitivity of projected electricity demand to the assumptions on economic growth is greater than for other energy forms, as reflected in a high and low range of demand growth rates for 1982-85 of 24.0% per annum and 6.1% per annum, respectively. The impact of these growth rates on the cumulative increase in electricity demand is shown in Table 2.2. This range of demand projections introduces a major element of uncertainty in planning power system expansion up to 1990, especially in planning means to deal with the possibility of a deficit in system genera- tion capacity relative to demand in the mid 1980s. From the analysis  - 44 - presented in Chapter III, it appears that the maximum growth rate in demand that can be met is that projected in the base case, by uprating Owen Falls Station. Therefore, it is essential that: any proposals to add major loads to the power demand in the next few years should be reviewed carefully to ensure that sufficient generating capacity exists to meet this demand. This restriction would cease once a new hydroelectric station is commis- sioned, which would be in 1991 at the earliest. At that time an expansion of electricity exports could also be considered. 2.45 Commercial woodfuels. In the base case, demand for commercial woodfuels is projected to increase at 6.2% per annum during 1982-85 and at about 4.8% per annum thereafter. These rates are substantially higher than the growth rates estimated for past years, including the 1965-70 period. This relationship reflects the consequences of a policy to develop commer- cial woodfuel production in the manner described in Chapter III as a domes- tic resource for substitution of petroleum products. Even in the low case, the projected rate of growth is between 4% to 5% per annum, and in the high case it is 9.5% per annum initially and drops to about 5% per annum there- after. 2.46 Projected energy balances. Corresponding energy supply scenarios to the base-case demand projections are given in the form of energy bal- ances in Tables 2.3 and 2.4 for 1985 and 1990 respectively. The projec- tions for energy supply are based on the current supply pattern and on lim- ited substitution between energy forms in view of the relative closeness of the reference years to the present time and to the anticipated shortage of investment resources that would be required to develop new energy sources and to bring about substitution. The only significant exception to this assumption is addition of generating capacity at Owen Falls Power Station in the mid 1980s and the substitution of marginal increases in demand for some petroleum products by electricity and woodfuel. The method of deriv- ing the projected supply requirements of woodfuel for fuelwood and charcoal (assuming improved kiln efficiency for charcoal production) is summarized in Table A2.5. An illustrative energy balance for the year 2000 is shown in Table 2.5 which is based on the same assumptions as the balances for the earlier years, particularly the absence of major energy substitution in- vestments except for a major new hydroelectric power station (and a related increase in export sales, over and above the levels shown in Table A2.4). The balance also assumes that no new source of primary energy becomes available in Uganda by the year 2000. The implications of these balances in terms of energy requirements are examined relative to supply options in Chapter III.  - 45 - D. Major Recommendations on Energy Efficiency 2.47 The Mission recommends that: (a) priority should be given to implementing a number of relatively simple and cheap conservation measures in the industrial (para- graphs 2.3 to 2.9) and transport (paragraphs 2.15 to 2.19) sectors. The Mission's proposals cover improvements in boiler and furnace efficiency, power-factor correction and maintenance of diesel engines. These measures would have an immediate impact on the energy situation, by constraining the growth of petroleum import requirements and the demand on generation capacity at Owen Falls Station; (b) measures should be taken to improve the efficiency of woodfuel use in agro-industrial plants (paragraphs 2.10 to 2.14). The scope for achieving higher levels of stove efficiency for cooking is less clear and progress in this area is likely to be slow. Nevertheless, further development and promotion of better stove designs is justified by the potential longer-term impact on wood requirements. Further savings of wood raw material could be realized through improved charcoal production techniques (para- graphs 2.20 to 2.24); (c) losses on power transmission and distribution should be reduced through system rehabilitation and power-factor correction (para- graph 2.25); (d) the feasibility and economic viability of proposals for fuel substitution in industry should be studied on a case-by-case basis. The Mission's preliminary findings suggest that some industrial conversions from petroleum products to woodfuels and, in the 1990s, electricity may be economic. However, this will depend very much on projected production levels and the associ- ated costs of energy investments (paragraphs 2.26 to 2.34); and (e) the Government should promote improvements in energy efficiency through appropriate pricing policies (see Chapter IV). A small section should be established in the proposed Energy Department to coordinate the many organizations involved in energy efficien- cy activities (see Chapter V). Short-term technical assistance, to provide expertise in conservation measures and to define an energy-efficiency program, is also proposed in Table 5.3.  Thble 2.3: PIJED RIMY BALU KR 1985 a/ (in '000 TCE) b/ SUPPLY CONSUNPTION Total Supply Tnirdssion & Net Supply Total Hw4w Ptiwry J After DistribtLon Available for twxda SDtwre Prodution InDrts Supply framfontion Losam GxoiuqtIon Household Camen ITuatry Dalmport Comuption Exports A. 0om.rdal e.r1 Aviation ftl - 20 201 20 - 2D - - - 20 20 - GRsoline - 52 52 52 - 52 - - - 52 52 - Karcene - 35 35| 35 - 35 35 - 5- - At". diei - 58 58 58 - 58 - 58 58 - InchUtrial d-esel - 1 1- - 1 - 1 - Fuel oil - 19 19 18 - 18 18 18 L. P. gm - . I 1 I 1 - 1 1 - - - 1 - Rlatrfrity - 1yio 18 180 60 7 53) 13 9 12 - 34 20 - tbem- - - 1 - 1) 0 Fualwocd 1,357 - 1,357 468 - 468 172 123 173 - 468 - Chw1I - - -; 22D - 22D 138 82 - - 22) - Total 1,537 186 1,723 934 7 927 359 214 204 130 907 20 Covien lolm - elecricitgy 12D - fu!l 1 ~~~~~~~~~~~669 - f~li.,o. I Total cmerdal eneqy 1,537 186 1,723 1,723 904 359 214 204 130 907 20 B. Non-c rca1 eary. Fuelvod 4,111 - 4,111 4,111 4,111 3,625 434 52 - 4,111 - C. Motal enrw' 5,648 186 5,834 | 5,934 5,038 3,984 648 256 130 5,018 20 a/ For the bune-ase asamptors. b See fract d report for cmaersiioz to it Smre: Mision estiates.  Table 2.4: PRJDU 1Y MBA= 1FOR 1990 a/ (in '000 TCE) b/ SUPPLY CONSUMPTION otal. SuppLy TrAmuiasdon & Net Supply btal Eneey Prry After Mstribution Awhilable for tbaia Source Prrd,tion limorts Su9ly Traxufozmation Icsses Comruption lkuehold C .rce Inlustry Tamport CoBumption Exports A. ComwxcW 42mmerg, A. Oseralg .rs I Aviation ful - 24 24 24 - 24 _ - - 24 24 - Gso]ine - 65 65 65 - 65 - - - 65 65 - Yercsene - 43 - 43 43 - - - 43 - hAto. dLesel - 72 72 72 - 72 _ - 72 72 - Indlsltial diel - 2 2| 2 - 2 _ - 2 - 2 - Rai1 O1 - 23 23 23 - 23 _ - 23 - 23 _ L.P.gas - 1 i1 - 1 1 - - - 1 - Elecricity - bdro 255 - 255 85 9 76) 17 11 17 - 45 31 - thle_ _ - - - - - - Filwood 1,674 - 1,674 588 588 231 149 20B - 588 - Charcoal - - - 277 - 277 174 103 - - 277 - Total 1,929 230 2,159 1,180 9 1,171 466 263 250 161 1,140 31 Goremion 1m m - electricity 170 - ful ood809 Tota cc.eaxcal energ 1,929 230 2,159 2,159 1,171 466 263 250 161 1,140 31 B. Nonomrcial enrgy F2elwxxd 4,639 - 4,639 4,639 4,639 4,076 503 60 - 4,639 - C. Tbtal mrgy 6,568 230 6,798 6,798 5,810 4,542 766 310 161 5,779 31 a/ For th base-cae asaptio. b/ See froit f report for coiwesions to itE. Swerce: Mission estirtes.  Table 2.5: TTTRAflVE BERFC BALAN F)R 2000 a/ (in '000 TOE) b/ SUPPLY CONSUMPTION Total Supply Tranumission & Net Supply Total Raergy Prmy After Distribution Available for panIa Sotrwe Prxhwtion Inports Supply Trarsfonnation Losses Corunption Household Commerce Industry lrhahspDrt onDumption Exports A. (oarcial energy Aviation ful - 38 38| 38 - 38 _ _ - 38 38 - Gasoline - 99 99 99 - 99 - _ _ 99 99 - Nerosene - 65 651 65 - 65 65 - - - 65 - Auto. diesel - ill ill ill1 - ill - -- ill ill1 T-ustrial disel - 3 3 3 - 3- 3 - 3 - Fuel all - 38 38 38 -38 - - 38 - 38 - L. P. gs - 2 2| 2 - 2 2 - - - 2 - Electricity - hydro 528 - 528 176 17 159) 31 20 27 - 78 81 c/ - thal _ _ - - - ) _ Fuelwold 2,580 - 2,580 947 - 947 429 216 302 - 947 - Char__l - - 445 - 445 279 166 - - 445 - TotaL 3,108 356 3,464 1,924 17 1,907 806 40 370 248 1,826 81 . Cmovemion lsse - electricity 352 - fuel aw 1,188 Total camnaial energy 3,108 356 3,464 | 3,464 1,907 806 402 370 248 1,826 81 B. no mercial energy Fuelwood 5,824 - 5,824 5,824 5,824 5,067 676 81 - 5,824 - C. Total aeergy 8,932 356 9,288 9,288 7,731 5,873 1,078 451 248 7,650 81 a/ For tbh base-caa asaimptions. b/ See front cf report for caoemions to TOE. c Ibdes ne exports of 120 NW and 600 GWh/year (Chapter III) vith 3% tranmmission Iceses. Source: Mission estimates.  - 49 - III. ENERGY SUPPLY: OPTIONS FOR DEVELOPMENT 3.1 Using the base-case projections presented in Chapter II, commer- cial energy demand in Uganda is expected to grow on average by about 5% per annum during the remainder of the 1980s. This growth rate could be signi- ficantly higher if measures are not taken to improve the efficiency of energy use. In addition, Uganda could potentially become an important exporter of electricity to other countries in the region during the 1990s, and planning for this must start now. Although Uganda has good potential for developing alternative energy sources (e.g, new renewable options such as solar energy and biogas, and possibly indigenous petroleum and geo- thermal resources), these are not expected to have a significant impact on the overall energy situation, at least for the 1980s. As a result, the emphasis in developing supply options will remain, for the immediate future, on the same three primary energy sources as in the past: hydroelec- tricity, petroleum imports and woodfuels. A. Hydroelectric Resources Power Potential on the Victoria Nile 3.2 Uganda is well endowed with hydroelectric potential. Although sites on a number of rivers have been studied,7! the potential is concen- trated on the Victoria Nile. The Victoria falls into three sections in Uganda, namely Lake Victoria to Lake Kyoga, Lake Kyoga to Lake Albert and Lake Albert to the Sudanese border at Nimule (see the map at the end of this Report). Profiles for the first two sections are shown in Graphs A4.2 and A4.3 of Annex IV. The last section is remote from the load centers and has the least power potential of the sections; it has therefore not been investigated in detail. The difference in elevation betweern Lakes Victoria and Albert is about 516 meters, of which 103 meters occurs between Lakes Victoria and Kyoga and 413 meters between Lakes Kyoga and Albert. The available head is concentrated in seven locations, with a total installed capacity of about 2,000 MW and annual firm generation capability of about 10,000 GWh. The major sites studied to date are described in Annex IV and summarized in Table 3.1. 3.3 The firm flow available for power at the Victoria Nile has been a matter of study since the inception of the Owen Falls power scheme. This is assessed from observation of movements in the level of Lake Victoria as shown in Graph A4.4 of Annex IV. The historic (1896-1946) mean of the flow from the Lake to the Victoria Nile was calculated as 631 m3/s. The firm 1/ Studies are available for three sites on the Nyakizumba, Muzuzi and Kiumi rivers. With a dam, the site on the Muzuzi river could supply up to 12 MW and 55 GWh per annum in a dry year. The other two sites studied have combined potential of 2.5 MW and 5 GWh per annum.  - 50 - Table 3.1: MAJOR SITES WITH HYDROELECTRIC POTENTIAL Potential Annual Generation Capacity Capacity Site (MW) (GWh/year) a/ Between Lake Victoria and Lake Kyoga: - Owen Falls (uprated) 210 b/ 981 - Bujagali 180 915 - Busowoko 160 840 - Kalagala 115 585 Subtotal 665 3,321 Between Lake Kyoga and Lake Albert: - Kamdini 230 527 - Ayago 540 2,900 - Kabalega 520 3,300 Subtotal 1,290 6,727 Total 1,955 10,048 a/ Based on firm flow available for power of 630 m3/s at Owen Falls. b/ Assuming the Owen Falls Station is uprated and based on 630 m3/s firm flow, the 60 MW extension project on the east bank of the river would not generate firm power or firm energy. However, it is possible that the detail- ed restudy of the hydrology of the Victoria Nile might modify this conclusion. In this; case, the hydroelec- tric potential of the river would be higher than shown in this table. Source: UEB's consultants' reports.  - 51 - (minimum) flow available for power through the year was taken by UEB's consultants in 1948 to be 505 m3/s, which was a flow level derived from historical records that represented a 95% probability that future flows will equal or exceed that level. The Owen Falls Power Station was designed accordingly based on 505 m3/s with a load factor of 60%, ten units each with a design capacity of 15 MW, maximum load of 135 MW and one unit in reserve, thus giving 150 MW as installed capacity. The annual firm energy was estimated as 710 GWh. 3.4 Since the completion of the Owen Falls dam in 1953 the river flow has been operated within its natural regime. Discharges from Owen Falls are balanced monthly in terms of amounts of water released downstream of the dam so that flow conditions downstream are controlled to simulate the conditions which would have prevailed if the dam had not been built. The maximum flow occurred May 1964 when the discharge rate averaged 1,698 m3/s (357 m3/s through the turbines and 1,341 m3/s through the sluices). Since 1966, UEB and its engineering consultants have considered that a flow of 630 m3/s is the firm flow available for power production. However, the average discharge for the period 1896-1981 was much higher than that for the 1896-1966 period, and therefore a higher firm flow would be derived from using records up to 1981 than was established in 1966. At present, an average of one-third of the total flow is passed through the turbines, while the remaining two-thirds flows through the sluices. This flow through the sluices shows that a large proportion of the potential energy flows down the river unutilized. Therefore there is a potential for uprat- ing the power station and/or extending the hydro scheme at Owen Falls. The extent of the potential depends on a reassessment of the hydrology of the Lake Victoria area. A multinational project sponsored by the World Meteor- ological Organization, which would determine new discharge and storage arrangements, was started in the 1960s and is still being carried out. Next Increment of Generation Capacity 3.5 According to the base-case demand projection and the reliability criterion of having one generation unit in reserve at Owen Falls Station, additional generating capacity will be required by 1986 to meet Ugandan demand and existing export contracts (see Table A4.5 of Annex IV). Al- though Uganda possesses many options for increasing its hydroelectric gene- rating capacity, the proposal for rehabilitating and uprating Owen Falls Power Station is the only project which could add capacity in time to avoid shortfalls in supply. Under this proposal, the generating capacity of each unit would be increased from 15 MW to 21 MW, resulting in an increase of about 60 MW (40%) to 210 MW in the rated capacity of the Owen Falls Power Station. The increased capacity could be operated at virtually 100% plant factor for base load since sufficient water is available for continuous generation. The proposed increase in capacity is due to four factors (see Table A4.1 of Annex IV): (a) increase from 505 m3/s to 630 m3/s in the available firm flow for power; (b) increase in the head available at the power station; (c) improvements in the design of equipment since the sta- tion plant was planned, especially for the turbine blades; and (d) use of better quality materials, especially for the generator windings, conductors and insulation. The preliminary estimate of the cost of uprating the units  - 52 - and rehabilitating the station :Ls about US$24 million (US$400/kW) at 1982 prices. Thus the proposal offers a very low--cost option for adding gener- ating capacity to the UEB system. Since the cost of new hydroelectric schemes is estimated to average US$1,000ikW, the proposal is also the least-cost option. 3.6 To avoid power shortages, five of the units at Owen Falls Station would have to be uprated, rehabilitated and commissioned by mid 1986. Due to shortage of surplus installed generating capacity over projected demand during the period of rehabilitation and uprating, only one unit at a time could be taken out of service to avoid power shortages. Project prepara- tion is still in the pre-feasibility stage. To meet this target, it is essential to start immediately a feasibility study of the proposal, as well as of the rehabilitation needs of the transmission and distribution system. Proposed terms of reference for this study (see Annex IX) were drafted and discussed by the mission with MlPED, UEB and ODA (which has subsequently agreed to finance this study). The Miss-ion has examined the program required and consider that the schedule for implementation is extremely tight. The critical activity for the uprating project is the long lead time for the manufacture of turbine blades. Least-Cost Development Program 3.7 The Owen Falls Power Station uprated to 210 MW would meet pro- jected demand until 1989 by when the station's firm energy generation capa- bility would become fully utilized (see Table A4.5 of Annex IV). The Mission considers that the earliest date that a new hydroelectric station could be commissioned is 1991. This would allow two years for site selec- tion, preparation of feasibility studies, resource mobilization, final design, preparation of bid documents and contracting arrangements, and another five years for construction. If this tight schedule can be adhered to, the energy deficits forecast for 1989 and 1990 of only 6 GWh and 39 GWh could be met by using some extra water from Lake Victoria. The next addi- tion to UEB's generation capability after uprating of the Owen Falls Station should be determined in accordance with the least-cost development program up to the year 2000. The program should be based on a new detailed demand forecast which should take account of the prospects of economic recovery, fuel substitution possibilities, additional exports and expansion of the transmission system. T7he least-cost power development plan for Uganda and possibly for exports to neighboring countries should also in- clude a detailed study of the hydrology of ;ake Victoria and the Victoria Nile, a detailed assessment of the optimal utilization of the hydro poten- tial, updating and comparison on a common hasis of the costs of the hydro schemes studied during the past 30 years, and the definition of the pro- posed development of the transmission system and of a system control center. The siting of the next major hydroelectric development is an important issue for the Ugandan Government in view of the Government's desire to promote development in the nDrthern part of the country. From information presently available it appears that the economic justification for developing one of the sites in the north (Kabalega or Ayago) rather than a site in the south (Bujagali, Busowoko or Kalagala) would be enhanced if a substantial export contract for power is obtained. The development of  - 53 - potential for exports is thus an important issue for planning the least- cost development program. Proposed terms of reference for this study (see Annex IX) were drafted and discussed by the Mission with MPED, UEB and ODA (which has subsequently agreed to finance this study). It is envisaged that the time required to execute the study would be about twelve months, and while the final report is being prepared a detailed updated feasibility study should be started for the site determined as the next development in the ]east-cost program. Additional Electricity Exports 3.8 Uganda already exports power to Kenya (paragraph 1.9), and the prospects of additional regional sales would have a significant effect on power development planning. The Government has recently agreed to export 16 MW (126 GWh/a) to Tanzania along the west side of Lake Victoria as far as Bukoba starting in 1985.2/ While Uganda has abundant reliable hydro- electric potential from a proven and almost constant outflow from Lake Victoria, most of her neighbors are not rich in electric potential. Uganda's neighbors are planning power development from indigenous resources for the rest of this decade at costs of up to twice the capital cost of Ugandan potential. There is a general lack of presently identified econom- ic power potential to meet requirements in some of these countries in the 1990s and beyond. The most promising market for Ugandan hydroelectricity lies in Kenya, and in 1968 Uganda and Kenya jointly studied the coordinated development of the electricity systems of the two countries. Although the study showed that joint development would be economically advantageous, the countries were not able to implement any joint development and presently Kenya is developing its own geothermal and hydro power schemes at consider- ably greater cost than Ugandan schemes would cost. However if Kenya demand increases as forecast, there will be opportunity for developing Ugandan hydro potential to supply up to 120 MW to Kenya by the early 1990s, as illustrated in Table A4.6 of Annex IV. 3.9 The World Bank's strategy is to encourage the development of re- gional projects where these projects have the potential to benefit all the countries concerned. Funds (US$2 million) have been included in the loan to Kenya on the Olkaria Geothermal Power Expansion Project for a study into the revision of the medium and long-term development program, considering among other options the possibility of developing regional resources for the benefit of Kenya. However, no formal discussions have been held be- tween the Government of Uganda, UEB and the neighboring countries, and the Ugandan authorities do not have any updated information about the cost of generation from undeveloped sites. The level of demand for export, in addition to the Uganda demand, would influence substantially the least-cost power development program. Therefore some scenarios (e.g., exports of 50 MW, 100 MW, 200 MW) should be chosen by the Government and UEB for evalua- tion and the least-cost solution(s) for these scenarios should be deter- mined. This exercise would provide Uganda with the prospective costs of 2/ The commissioning date agreed by the two Governments. However, the Mission considers it unlikely that the necessary transmission line and transformer station could be constructed before 1986.  - 54 - exporting power. Such costs are likely to be substantially higher than the existing selling price of power to Kenya (US cents 0.63/kWh and US cents 1.13/kWh), probably at least US cents 5/kWh (Table 4.1). This is still much less than the average cost of power fran new power projects in other countries of the region. Transmission System Extension 3.10 UEB has prepared 14 transmission extension projects on completion of which almost all of the population centers with at least 3,000 people according to the 1969 census would be connected to the national transmis- sion system. For some of the towns connected to the main system, UEB would retain the diesel sets as stand-by capacity for important consumers such as hospitals. The feasibility of seven of the schemes has been studied by the Resident Representative's Office in Kampala of the EEC, covering the con- struction of about 651 kilometers of 33 k'J transmission lines and the installation of 9.6 MVA transformer capacity in three years. Four of the schemes would replace existing diesel stations at Kitgum, Kabale, Rukungiri and Moyo. The alternatives to these scliemes would be to continue operating the diesel stations at high cost, both for supplying spare parts and conti- nuing to generate power inefficiently. In 1981 the financial cost of gen- eration from the diesel plants was USh 5.95/kWh of which USh 4.05/kWh was fuel cost (including transportation).3/ Thriee schemes would connect areas which presently have no access to electricity. The alternative for these areas would be to install diesel units. In addition to individual domestic consumers, these schemes would supply power to coffee and tea processing plants, maize mills, saw mills and water pumping stations. The schemes in total would provide a load of about 5.6 MW to the interconnected system and would serve areas with a total population of about 93,000 people. The total cost of the seven schemes is estimated by the EEC Representative to be US$5.4 million in 1982 prices. Because the alternatives are either to continue operating the existing diesel stations or to install new diesel plants, the rate of return of the schemes evaluated as a single project is very high. The remaining seven schemes include the construction of about 1,300 kilometers of 33 kV lines, and would permit the remaining diesel sta- tions to be shut down. The preliminary cost estimates by the Mission of the latter seven schemes is about US$9.0 million. When completed some of these 33 kV lines would form loops in the transmission network which would increase reliability of supply to the towns. B. Petroleum Products 3.11 At the present time, Uganda imports all of its requirements of petroleum products. There are signs that Uganda could have some petroleum 3/ The only reliable method of estimating the economic cost is to use import-parity prices, and on this basis the economic cost of supplying power from diesel sets in isolated load centers in Uganda was probably about US cents 20/kWh sent out.  - 55 - resources in the Lake Albert area and further work to assess the extent and economic viability of these resources is justified (see paragraphs 3.26 to 3.29 below). However, development of domestic petroleum resources is a long-term and still uncertain prospect. At least for the 1980s, Uganda will remain dependent on imported petroleum products and the main objective should be to minimize the costs, particularly in foreign exchange, of these supplies. There are four options available: (a) diversification of supply sources; (b) reduction in transport costs to the Ugandan border; (c) devel- opment of alternative supply routes; and (d) maintenance of a product stockpile. 3.12 Already, during 1983, significant progress has been made in two key areas: product purchases have been switched from the Kenyan market to the Middle East market and the processing of crude through the Mombasa Refinery, traditionally cheaper sources of supply; and, an alternative transit route through Tanzania has been established. These actions, to- gether with a fall in the price of crude oil, have reportedly helped to increase the volume of products received for each dollar spent by as much as 20%, and average stock levels have risen from one to four weeks' con- sumption. Even so, the Government still needs to take a more active role in monitoring petroleum imports, based on an informed understanding of changing market and regional conditions, to ensure that an appropriate mix of supply options is maintained. Diversification of Supply Sources 3.13 There are three potential sources of supply to Uganda for petro- lmum products: (a) purchase of products directly from the international market, principally from sources in the Middle East; (b) purchase of crude oil from the international market, viz Middle East sources, for processing in the Mombasa Refinery; and (c) purchase of products from the Kenyan market. The least-cost source is the purchase of products directly from Middle East markets. Comparative costs of these sources for products ex-Mombasa in December 1982 are given in Table 3.2. Table 3.2: COST OF PETROLEUM PRODUCTS FROM VARIOUS SUPPLY SOURCES (ex-Mombasa in US$/ton of "product") a/ Purchase directly from Middle East market: 361 Process crude oil in Mombasa Refinery: 410 Purchase from the Kenyan market: 395 a/ The comparison is based on a "product mix" at Mombasa obtained from one ton of crude oil processed at Mombasa Refinery and used in Uganda. Source: Annex V.  - 56 - The comparison of costs of alternative sources of supply shows that it would be in Uganda's interests to purchase as much as possible directly from Middle Eastern markets at present cost relationships. The relation- ships between the costs of alternative scurces of supply is not static, for example, Kenyan export prices are fixed in terms of Kenyan currency and therefore change when the Kenyan sxchange rate is altered. At the begin- ning of 1981, the cost of a ton of "product" from the Kenyan market was about US$467, at which time t1his source was by far the most expensive. This, in fact, is usually the case and a simi:Lar relationship has probably been re-established during 1983 as crude oil prices have fallen. Because of these fluctuations, it is essential that the Government closely monitors movements in the relative costs of: alternative supply sources. 3.14 The legal framework for the supply of petroleum products to Uganda through Kenya is a critical factor for the economic Issues. At the time of the East Africa Community (pre 1977), Uganda had the right to oper- ate an in-bond corridor through Kenya. Although there is some doubt about the present validity of this agreemnent, a. number of oil companies operating in Uganda have recently made direct product purchases from the Middle East and been allowed to transport t]hese through Kenya by pipeline and road transport. Under the original agrfeement for the Mombasa Refinery, partici- pants in the Refinery are obliged to supply the Kenyan market with products from the Refinerye At th-e time of the Mission, in October/November 1982, two oil companies operating in Uganda (SheLl and Esso) were purchasing crude oil on the international market for processing in the Refinery. These two companies supply about 50% of Uganda's petroleum requirements. However, all oil companies operating in Uganda have the right to use refin- ery facilities at Mombasa, three of them by virtue of their parent compa- nies' participation in the Refinery Processing Agreement4/ and the other three when the Refinery has unuseid capacity for white products. 3.15 The Refinery Agreement also operates to Uganda's disadvantage in some respects. Firstly, the Mombasa Refinery unit charges are presently about five times as great as unit charges in the major international oil- ref ining centers. Secondly there have been interruptions recently in refinery operations and hence su]Pplies t-o Uganda due to inadequate avail- ability of foreign exchange to Kenya for the purchase of the minimum quan- tities of crude oil required to maintain c,ntinuity of refinery opera- tions.5/ Thirdly, there are obstacles to the transportation of imported white products from Mormbasa to 1Zairobi, since by law all white products have to be transported by the petroleum produict pipeline (commissioned in 1978), yet the refinery operators control the use of the pipeline. In practice, this means that the refiiaery operatcrs consent has to be obtained for the importation Df white prodlucts; they can insist that efforts are made to purchase products on the Kenyan market, or charge a refinery loss fee, before making the pipeline available. 4/ Shell, Esso and Caltex. 5/ The minimum throughput quaptity for the Mombasa Refinery exceeds the volume of crude o`l that needs to be processed to meet Ugandan demand.  - 57 - 3.16 Since September 1980 the Kenyan Government has established mini- mum in-bond export prices (MEP) for petroleum products destined for Uganda, Rwanda, Burundi and Eastern Zaire. The minimum export prices are generally established with different margins to notional ex-refinery prices. In October/November 1982, four oil companies,6/ that together supply about 50% of the Ugandan market, bought products on the Kenyan market on the basis of NEP prices. However, following negotiations with the Government in late 1982, these purchases have now stopped. Instead, these companies have started making direct product purchases on the Middle East market or processing crude through the Mombasa Refinery, both of which are usually cheaper sources of supply. Further adjustments in supply sources should be made, as and when appropriate, to take advantage of changing cost relation- ships. 3.17 One complication in any negotiations on sources of supply between the Uganda Government and the oil companies is the relationship between affiliated subsidiaries of the oil companies in Uganda and Kenya. Some of the Ugandan oil companies purchase petroleum products from affiliated Kenyan companies. Pricing strategies amongst affiliated companies vary be- tween oil companies, for instance some companies seek to maximize the profits of their Kenyan subsidiaries. The margins on products imported into Uganda is a crucial factor in determining the mix of products that each oil company elects to import with its foreign exchange allocation (paragraph 4.18). Therefore, negotiations could be complicated by con- flicts of interests between maximising benefits to Uganda and oil company operating strategies. Reduction in Transport Costs to the Ugandan Border 3.18 At present, less than 10% of petroleum products are transported from Mombasa to the Ugandan border (Malaba) by rail (a distance of about 1,040 kilometers). Most petroleum white product imports are transported by a combination of the Mombasa to Nairobi pipeline and road tankers. This mode is not the cheapest available in principle. Presently there are four options for transporting white petroleum products to Uganda, namely by: (a) railroad from Mombasa to Malaba; (b) road tankers from Mombasa to Malaba; (c) pipeline from Mombasa to Nairobi and road tanker to Malaba; and (d) pipeline from Mombasa to Nairobi and railroad to Malaba. Since only white products can be transported in the pipeline, the options available for transporting black products are only all road tanker or all railroad, and generally these products are transported in road tankers. The costs of transportation by these options are given in Table 3.3. This analysis of costs shows that Uganda is paying about US$80/ton for transportation of white products from the East African coast to its border, and about US$74/ ton for transportation of black products. Transportation therefore imposes an additional cost equivalent to about 20% of the value of products at the Mombasa Refinery. 6/ Agip, Caltex, Mobil and Total.  - 58 - Table 3.3: TRANSPORT COSTS, FOR PETROLEIJM PRODUCTS THROUGH KENYA (in US$/ton at mid-1982 prices) Tariff Transport Losses a/ Total Cost Transport White Black White Black White Black Option Products Products Products Products Products Products Railroad 50 42 10 4 60 46 Road tankers 83 71 8 3 91 74 Pipeline + railroad 58 . 9 . 67 Pipeline + road tankers 73 . 7 . 80 a/ Due to the slow movement of rail transport, losses are expected to be relatively higher than for other modes. The values given here are rough estimates only. Source: Annex V. 3.19 It is generally accepted that the railroad potentially offers the most economic transport option for long-hatulage freight traffic in the region. The railroad is also the cheapest method of transportation at present freight rates in Kenya. For white products the requirement to use the pipeline adds about US$8/ton (13%) to transport costs in conjunction with rail transportation. Road freight rates are 85% higher than railroad rates and 43% higher than pipeline rates (see Annex V). If Ugandan demand on Kenyan railroad services was to increase substantially, it is possible that the present low railroad tariff rate would be raised. However, it is also possible that road truckers would be able to increase their rates to preserve the present differential whilst maintaining their present share of the freight market. Therefore, there appears to be a strong economic incentive for Uganda to maximise the use of railroads at present rates, especially since transportation costs are levied in foreign exchange. The potential saving from switching from road to rail is currently about US$2.3 million annually7/ which is about 20% of the present cost of transporta- tion by road and pipeline. 3.20 Potentially, there are sufficient railroad rolling stock avail- able in Kenya and Uganda to transport Uganda's present petroleum import requirements, provided they are allocated for this purpose. However, some rehabilitation of the railway systems would be required to handle substan- tially higher traffic volumes. It would also be necessary to reduce the 7/ In 1981, the savings would have consisted of about 130,000 tons of white products at US$13/ton and about 23,000 tons of black products at US$28/ton.  - 59 - influence of trucking vested interests on the allocation of freight to transport modes to realise a major switch of traffic from road to rail. A potential alternative to using Kenyan road transport is the proposal to extend the pipeline westwards from Nairobi to serve western Kenya, Uganda, Rwanda and Burundi. At present it appears that the volume of products transported to these markets is insufficient to justify such an invest- ment. Even so, a feasibility study on alternative options for extending the pipeline should be undertaken. EIB/EADB has offered financing for this purpose. Development of Alternative Supply Routes 3.21 Due to Uganda's landlocked situation, security of the supply route from the East African coast for petroleum products is of vital con- cern. Until recently, Uganda had only one supply route through Kenya, and thus it was vulnerable to interruptions in supplies through this route that were beyond Uganda's control. This aspect became sufficiently serious during 1982 to warrant strategic planning of an alternative route. Both the Refinery and the pipeline are sources of interruptions to supply. There have been interruptions during the past year to Mombasa Refinery operations arising from shortage of crude oil for the Kenyan market, and the pipeline between Mombasa and Nairobi was out of operation for some weeks in 1982 due to contamination. 3.22 The Ugandan Government has decided therefore to explore with the oil companies the option of developing a supply route through Tanzania. In this context, Shell recently agreed to carry out a trial run for importing petroleum products through Dar es Salaam and transporting the products by railroad to Mwanza and thence by lake ferry to Jinja. Although this trial is largely symbolic in relation to overall transport requirements for petroleum product imports, it will be a useful indication of the feasibil- ity of developing this route and of the costs involved, including loss factors and the time required. Such development would probably include provision of rolling stock and bonded storage facilities at Mwanza, and would be dependent on adequate progress in the planned rehabilitation of the Tanzanian railway system.8/ Reportedly, Tanzania has given permission for URC to operate its own block trains all the way from Dar es Salaam to Mwanza, for the movement of petroleum products. There already exists a rail ferry on Lake Victoria which is presently underutilized. 3.23 The proposed Tanzanian route is about 1,600 kilometers long, which is about 43% longer than the route through Kenya (to Kampala). The cost of imported products from the Middle East is about the same at Dar es Salaam as at Mombasa (see Annex V), and therefore is considerably less than costs ex-Mombasa Refinery and the Kenyan market. The Tanzanian route would also take advantage of relatively cheap rail transport, compared to the predominance of road transport presently used on the Kenyan route. Al- though firm cost estimates of transportation via Dar es Salaam have yet to be established, indicative cost estimates are shown in Table 3.4. 8/ The Tanzanian railway is used currently to transport petroleum products to Rwanda and Burundi.  - 60 - Table 3.4: COSTS FOR PETROLEUM PRODUCTS SUPPLIED THROUGH KENYA AND TANZANIA Product Transport Transported Landed Cost Source of Product Value Cost Rate Distance c/ in Uganda (US$/ton) (US$/ton-km) (km) (US$/ton) Middle East: Dar es Salaam 362 0.075 a/ 1,600 482 Mombasa 360 0.09 b/ 1,120 461 Mombasa Refinery 410 0.09 b/ 1,120 511 Kenyan market (Mombasa) 395 0.09 b/ 1,120 496 a/ By railroad and lake ferry, including transport losses. b/ By pipeline and road tankers, includig transport losses. c/ To Kampala. Source: Annex V. The cost to Uganda of importing products through Dar es Salaam appears to be of the same order as obtaining products from the Mombasa Refinery and from the Kenyan market. Therefore any premium that Uganda might pay by developing the Tanzanian route appears to be small relative to the poten- tial savings in costs through avoiding interruptions in supply. Any premium would be incurred on only a small proportion of total imports that would be transported through Tanzania to keep open this route. The Ugandan Government should encourage and follow closely the trial run organised by Shell, and should pursue this development with the oil companies by inves- tigating the physical facilities required and the economics of this route. Maintenance of a Product Stockpile 3.24 As part of a strategy to reduce vulnerability to interruptions in supplies of petroleum products, IJganda should maintain a "minimum safety stock" of products. This stock would be kept: in reserve to meet Ugandan requirements during a short interruption and to provide the time needed to mobilize alternative supply facilities, such as bringing into the service the full capacity of the Tanzanian route (paragraph 3.22), during a long interruption. The level of such a stockpile should be determined by the time required relative to Ugandan demand to mobilize alternative supply facilities, and one advantage of maintaining a route through Tanzania would be a reduction in the time required for mobilisation and thus a saving in the cost of stockpiling. 3.25 During the visit of the Mission (October/November 1982), the stock level decreased from seven days' to four days' consumption due to problems related to the Mombasa Refinery and the pipeline in Kenya, and consequently shortages occurred of some products. However, with the recent  - 61 - improvements in supply arrangements noted above, the level of stocks has now been built up to four weeks' consumption. From a preliminary assess- ment of the situation, this level of stocks would seem to be adequate to provide protection against a sudden disruption in supplies. The Government should also take an inventory of existing storage facilities, including those not in use at present, and investigate the possibility of selling or leasing idle storage capacity to the oil companies for stockpiling. How- ever, there would appear to be no general shortage of storage capacity at the moment, except possibly for some products (e.g., kerosene) and in some up-country areas. As shown in Table A1.6 (Annex I), the storage capacity is adequate for eight weeks of supply at present consumption levels. Petroleum Exploration 3.26 The most interesting area in Uganda from the viewpoint of petro- leum exploration is the Western Rift Valley, in particular Lake Albert (Lake Mobutu Sese Seko). The rest of the country appears to be of little prospective interest. Oil seepages along the sharp fault lines of the Rift Valley have been known for a long time. Although these oil seepages are not very strong, their documentation in the 1920s induced interest in exploration and several parties applied for concessions. About 20 test wells were drilled with an average depth of approximately 350 feet and one deep well to 4,043 feet. Oil shows were found in one well and the deep well identified oil sand and numerous black shales. Since 1956 no more wells have been drilled. However, interest in the prospects of the Lake Albert area continued. Comoro Exploration Ltd. applied for a concession in 1971 and subsequently Petro-consultants S.A. reviewed Uganda's geology. Petro-consultants estimated that the thickness of sediments could be as much as 13,000 feet in the Lake Albert area, which would indicate a situa- tion favorable to petroleum generation. However, in the one deep well drilled the thickness of sediments was only 4,000 feet, although this well was drilled close to the edge of the basin near the fault line. In conclu- sion, the Western Rift Valley appears interesting for further exploration, and the next objective of exploration should be determination of the thick- ness of the sediments in the Rift Valley. 3.27 Cooperation between Uganda and other countries in the region is an important prerequisite for further exploration work. The border between Uganda and Zaire runs through Lake Albert, and the entire lake area has to be surveyed to get meaningful geological information. Similarly, coopera- tion with Tanzania and Kenya is required to complete surveys of the south- ern portion of the Rift Valley and Lake Victoria. It is therefore encour- aging to note that these four countries have recently agreed to participate in a joint airmag survey of these areas, and to share the costs and results of this exercise. Hopefully, this same spirit of cooperation will be carried over into subsequent exploration and possibly development activi- ties. 3.28 Petroleum legislation does not yet exist in Uganda. In the past, the few exploration efforts were regulated on the basis of individual agreements between the Government and the exploration company, although a very general framework was provided by the Uganda Law of Mines. The recent  - 62 - discovery of oil in Sudan north of the Uganda border has created interest in exploration in Uganda. MPED and the Ministry of Finance have received a number of inquiries from oil companies, although the mission was not able to identify either those companies or the nature of their inquiries. How- ever, should a company make a formal request it would be difficult to begin serious discussions due to the lack of petroleum legislation. Preparation and passing of petroleum legislation therefore needs to be undertaken in the field of exploration promotion. 3.29 The Commonwealth Fund for Technical Cooperation (CFTC) has agreed to assist the Government in preparing petroleum legislation. The World Bank has also offered to finance an airmag survey, as part of a larger regional exercise, out of the IDA Technical ALssistance Credit. Based on the results of the airmag survey, a gravity survey is planned to identify prospective areas. The results of the various geophysical surveys would then be compiled into a promotion report that would be offered to inter- ested oil companies for sale. The objective of these exploration promotion activities is to attract oil companies' risk capital for exploring hydro- carbons in Uganda. The proposed timeframe for these activities is to complete the airmag survey in 1983 and the gravity survey in 1984, tender exploration blocks to oil companies in 1985 and explore concessions from 1985 to 1988. In the event of a discovery there would have to be technical and economic feasibility studies before any development activities could be started. An important issue to be addressed in this follow-up work would be whether to process the crude in Uganda for domestic use or export it to finance imports of products. It is important to bear in mind that, even in the event of a viable discovery, hydrocarbons could not be produced before the start of the next decade. C. Fuelwood and Charcoal 3.30 Development of woodfuel sources of energy requires little invest- ment of foreign exchange, whereas other sources of energy, such as kero- sene, solar panels and to a certain extent electricity, require substantial foreign exchange inputs. A number of options for increasing supply of woodfuels are examined in this section under the following categories: agro-forestry, urban plantations, industrial plantations and woodlots, forest plantations, management of natural forests and woodlands, and low- potential agricultural areas. 3.31 If wood is to meet an increased demand then sustained annual yields must be increased. The availability oE land for silviculture will be constrained by the demands for agricultural production to feed a growing population and for export crops. Some of the forest estate will be cleared for agricultural use. However, it is possible to increase tree production on farms without decreasing agricultural output by better management prac- tices and by the introduction of more suitable farm trees. By converting some of the natural forest to plantations, annuial production could increase four to five fold. By applying improved munagement techniques to the  - 63 - remaining areas of natural forests and woodlands, it should be feasible to double production. Therefore, it should be possible for Uganda to both in- crease agricultural and silvicultural production in order to satisfy the demands for food, cash crops and wood (including woodfuels). 3.32 The growing of more trees, the introduction of better tree man- agement techniques, and the expanded use of other renewable energy forms will take time to yield results, even using quick growing tree species on short rotations, for manpower has to be trained and plantations planned and planted. According to the preliminary estimates of stock, there is a con- siderable reserve of tree capital that could be used in the interim period until new sources of supply of fuelwood become available, but some of this stock may be too remote from the demand centers to be an economical source. The Forestry Department should therefore undertake an inventory of tree stocks on a district-by-district basis. Much wood for energy purposes comes from outside forest areas and from the tops and branches of trees. It is therefore essential to estimate total volume/weight (not just stem volume) and the total annual production. Plantation grown trees should be recorded by species and age and the Forestry Department should classify forest reserves into protection and production areas. This inventory should be complemented by a survey of consumption trends and projections for all wood products (particularly fuelwood, charcoal and poles) to help pinpoint the areas of immediate shortages. Measures can then be pursued to help remedy this situation. Agro-Forestry 3.33 To save fuel for transportation, wood should be grown as near as possible to demand centers. For the bulk of the rural population, wood has to be within walking distance from the home and ideally farmers' families obtain fuelwood from their own land or from nearby public lands. There are many benefits to be derived from planting multi-purpose trees, even though there is increasing pressure on land due to population increases and there- fore farmers have to increase the unit production of agricultural crops. It is possible to increase both wood and crop output on a piece of land or at least to increase total output if certain tree species are introduced into the cropping system and if they are grown in certain arrangements and up to certain densities. Many tree species fix nitrogen and therefore their pods and/or leaves can be used as animal feed. Also most leaves, whether from nitrogen fixing trees or not, can be used for green manure for the trees recycle mineral elements that have been lost from the top soil horizons. Trees also prevent excessive erosion and sun scorching as well as improving the microclimate. Trees of the legume family generally have a profusion of flowers which attract bees, giving rise to honey production, and several trees have edible leaves, pods and fruit. 3.34 In an agro-forestry demonstration center in Rwanda, for example, 300 Grevillea robusta trees per hectare on a six-year rotation yielded 7.6 tons of air-dried wood and 2.5 tons of green leaves. If the roots of the Grevillea trees are pruned at a radius of two meters from the tree then the loss in agricultural production is 2% or less. If a legume tree species were to be used then an increase in agricultural production is anticipated.  - 64 - These experiments are still continuing, but the above wood production, with present end-use devices, could satisfy the wood requirements for a family of seven or eight. If improved cooking devices are introduced, ten or more people could be supplied with their wood requirements. Of course it is difficult to achieve experimental results in the field, but half the above yield would be sufficient for a family of five owning one hectare of high potential agricultural land if stove improvements are introduced. There- fore, the agro-forestry solution seems ideal for most rural communities although there will be scope for woodlots and. plantations to meet specific needs. 3.35 Agro-forestry is not a new technique, but various practices are being codified and improved upon. A survey of farm tree systems should be undertaken within Uganda and agro-forestry methods should be taught in agricultural and forestry colleges and in the forestry and agricultural departments at the Makerere University. Extension workers should be train- ed in agro-forestry and farm tree management so that they can provide prac- tical advice to farmers. At present, the extension system in Uganda is basically non-operational and it will have to be substantially strengthened and reorganized to become effectiLve. However,, recommendations in this area go beyond the scope of this energy assessment. Urban Plantations 3.36 In urban areas shortages of woodfuel are either already manifest or are likely to occur in the near future, for the population of these areas is increasing at about 7% per annum, that is doubling itself every ten years. In Kampala evidence of the shortage is that people are switch- ing back to fuelwood because of physical shortages of charcoal, kerosene and L.P. gas plus electrical cookers (rings). The charcoal shortages may be due to transport restrictions arising fran security problems9/ as well as diminishing wood supplies. To overcome these shortages charcoal and fuelwood production must be increased. To avoid the conversion losses in charcoal production, people should be encouraged to burn wood, by growing fuelwood near the consumers, cutting it into convenient lengths and market- ing it properly in a package that includes modern stoves. One solution to this problem would be to establish fuelwood plantations around or near towns. Such plantations could act as recreational areas or provide animal feed and honey as well as fuelwood and poles. 3.37 As the main conurbation in the country, Kampala/Entebbe has a current population of about 0.5 million. By the turn of the century this population will have increased to 1.7 million if the annual growth rate averages 7%. Even allowing for increases in production and consumption efficiencies, the output from the equivalent of about 50,000 hectares will be required in the year 2000 to supply its needs of woodfuel and poles, an 9/ For example, chiarcoal prices in Kampala are about double those pre- vailing just 15 kilometers out of town. The Mission was told that this price differential was caused primarily by the fact that as much as one half of the charcoal transported to Kampala can be "lost" at roadblocks along the road.  - 65 - area larger than the total plantation area in the country today. For the country as a whole, the urban population will increase from its present level of 1.0 million to 3.4 million by the year 2000 at the same rate of growth. Therefore, the task of providing urban woodfuel is formidable and the Government should assist in the planning and execution of urban and other plantations. There may be problems of acquiring sufficient land near urban centers if there are not already forest reserves in the vicinity, for it would mean a change of land use. Such changes may need government legislation, although private individuals may invest in fuelwood planta- tions as the most profitable form of land use given the relatively high price of fuelwood and charcoal, as has happened near other major cities in the region (e.g., Addis Ababa). Government could give such individuals en- couragement by providing inputs (for a charge) and management assistance. Industrial Plantations and Woodlots 3.38 Fuelwood and charcoal is used extensively by industry, especially agro-industry, for drying or processing crops. In all cases wood is the cheapest energy source, but because its continual supply was not always planned for, some industries such as fish smoking, tea drying and cement manufacture are running short of it or have switched to other more expen- sive fuels. There are also several industries that would benefit economic- ally by switching to woodfuel but a continual source of wood energy must be guaranteed. In order to ensure this, industrial plantations and woodlots should be established near the industry. Government should assist the various industries by giving them technical help and leasing to them areas of forest reserves as they have done in the case of the tobacco industry. If necessary, the Government could also provide inputs and financial sup- port, for foreign exchange will be saved by burning wood and jobs will be created. Forest Plantations 3.39 Forest plantations will have to be expanded to provide the raw material for sawnwood and panel products and, most importantly, for fuelwood and poles. It is probably too early to think about establishing plantations for paper production but a study could be undertaken to see when, where and if such plantations should be started. The output per unit area could be increased up to fivefold by converting from natural forests to plantations. By planting the perimeter of natural forests to a depth of about half a kilometer, the natural forests are protected from encroachment, so it seems both prudent and necessary to at least convert the perimeters of forests to plantations. The actual area of plantation required will depend on the demand and supply forecasts and how successful other efforts to meet demand have been. Management of Natural Forests and Woodlands 3.40 At present very little management is being undertaken in the natural forests. A specified number of commercially valuable trees are being legally removed, and there is an unknown quantity that is removed illegally. However, less than 40% of trees in the natural forests are classified as commercial and many non-commercial trees are left either to  - 66 - grow and eventually die or are ring barked or poisoned and left to rot. However, in some areas a "uniform. management system" used to be practiced and unwanted and undesirable trees cut out for charcoal production or fire- wood. Line planting with Maesopsis emini was used to boost natural regene- ration.10/ Given some reorganization within the Forestry Department plus more manpower and equipment, this system could be revived, more forest areas could be included within tlhe management plans and charcoal/fuelwood production increased. At the same time an assessment of the area under bamboo plus the possibility of its management could be investigated, for in some countries bamboo is a very important forest species. Low-Potential Agricultural Areas 3.41 The north and north-east of the counl:ry are regions most affected by desertification. They are areas of low rainfall and high evaporation rates where pastoral agriculture is dominant. The people are nomadic but there are population concentrations living near permanent water supplies. It is near these settlements that shortages of fuelwood and poles occur. There is, therefore, a need to establish village woodlots and pastoral tree systems to meet the needs of the people. The costs of such woodlots are relatively high and the wood returns are low because the expected output is about one-quarter that of high potential areas and the costs are about double. Nevertheless, the existing tree population will be devastated and deserts formed if sustainable wood energy is not provided for these people. It is essential to establish a protective barrier of trees to prevent desert creep. With browse or feed trees it may even be possible to sustain a higher animal population as welL. The proposed supply and demand studies will pinpoint the areas of greatest need. However, parts of Karamoja are already becoming deserts. UNDP has proposed a US$2.5 million "forestry for rural community development" scheme for the Karamoja region to supplement the existing help being given to the Ugandan Government by the EEC and Oxfam. 3.42 The establishment of trees is the most difficult problem and micro-catchment techniques with solar electrical fences, using such species as Prosopis spp, Albizia spp and Azadirachta indica (neem) should be tried. Such systems have already been successful in the Baringo area of Kenya, which is close to Karamoja, and cooperation with this fuel and fodder pro- ject could be beneficial. The solar electrical fences have succeeded in keeping out domestic and wild animals where thorn fences have failed. In cases like this the use of relatively sophisticated technology may be justified. Back-Up Requirements 3.43 Sufficient seeds, cuttings and seedl:Lngs must be available in the right place at the right time to umdertake a large increase in the planting program and to expand farm tree planting. This will require planning, 10/ Various methods are described in a UNDP report by T. W. W. Wood en- titled 'An Appraisal of the Management of the Natural High Forest and Silviculture Research" (1978).  - 67 - organisation and trained manpower. Various back-up requirements are listed as follows: (a) indigenous and exotic tree seeds will have to be acquired and distributed to the areas where they are required. Temporary seed stores may have to be established in various parts of the country and a central seed store expanded. In order to save time, effort and money, seeds that can be directly sown should be used wher- ever possible, particularly on farms, but this will entail the training of farmers in the raising and tending of such species; (b) nurseries should be expanded or established in every area and farmers trained to grow their own seedlings. To this end pam- phlets should be produced to illustrate the raising of seedlings, the tending of them and their thinning and felling; (c) the Forestry Department should organize the collection of seeds from superior indigenous and exotic tree species and start seed orchards and a tree-breeding program; and (d) trials should be established for species, spacing, rotation, yield of wood and other products and the effect upon the soil. There are many multi-purpose tree species that could be tried. Returns to Fuelwood Plantation 3.44 The estimated production, costs and revenue from a commercial fuelwood plantation are summarized in Annex VI. These estimates are based on the Rwanda experimental model in which species are grown on a six-year rotation. Assuming no opportunity cost for land, a labor rate of USh 50/ manday and using the present stumpage fee of USh 200/m3 solid, the finan- cial yield to the producer of fuelwood works out to 26% per annum.11/ However, in practice, land often does have alternative economic uses and the retail prices for fuelwood (USh 1,200 to USh 4,800/m3 solid) are sub- stantially higher than the present stumpage fee. The economic values of fuelwood required to yield an economic return of either 10% or 20% for a range of values for the opportunity cost of land are summarized in Table 3.5. The economic value of fuelwood is in turn related to the fuel substi- tuted, the efficiency of use and also whether the fuelwood is used directly or transformed into charcoal. Substitution opportunities are mainly in the use of fuel oil for raising steam in industrial boilers and for crop drying in agro-industrial plants. Illustrative examples of the economic value of fuelwood in these uses are given in Table 3.6.12/ 11/ Alternatively, if direct labor is given zero opportunity cost, the yield is increased to 46% per annum. 12/ For simplification, these examples assume that trees are grown for fuel purposes only. However, in practice, trees can provide a number of joint products and also have a positive impact on the productivity of the land (if properly used within an agro-forestry system). Taking these factors into account, the economic value of growing trees would be higher than shown in Table 3.6.  - 68 - Table 3.5: ECONOMIC RETURNS TO FUELWCOD PLANTATION a/ Economic value required to Alternative land use Opportunity cost give economic return of to fuelwood plantation of land use b/ 10% 20% (USh/ha) ---(USh/m3 solid)--- Cotton, tea under 10,000 under 470 under 700 Maize, tobacco 30,000 1,350 1,800 40,000 1,700 2,200 Bananas 50,000 2,100 2,700 Coffee 60,000 2,550 3,250 80,000 3,250 4,500 100,000 4,000 5,500 Peri-urban housing over 120,000 over 4,800 over 6,000 a/ Based on a six-year fuelwood plantation mDde] as shown in Annex VI. b/ Given by estimated economic returns to land from alternative use. Source: (1) Annex VI. (2) Mission estimates. Table 3.6: ECONOMIC VALUE OF FUELWOOD (in substitution of fuel oil) ------At an exchange rate of--------- USh 100 - US$1 USh 200 - US$1 Economic value of fuel oil (USh/liter) a/ 44 44 44 88 88 88 Woodfuel used Fuel- Char- Char- Fuel- Char- Char- wood coal coal wood coal coal Economic value of woodfuel at plant gate (USh/m3 solid) b/ 8,002 26,329 26,329 16,004 52,658 52,658 Transformation ratio from fuelwood to charcoal c/ * 5:1 8:1 , 5:1 8:1 Economic value of roundwood at plant gate (USh/m3 solid) 8,002 5,266 3,291 16,004 10,532 6,582 a/ Based on import-parity price (Table 4.6). b/ Calculated as x x 1,060 x 0.7 where: y x z x - economic value of fuel oil (in USh/liter) 1,060 - no. of liters per ton of fuel oil 0.7 - no. of tons per m3 solid of woodfuel y - ratio of energy values: fuel oil/fuelwood - 2.55 fuel oil/charcoal - 1.24 z - ratio of end-use efficiencies: fuel oil/fuelwood - 1.6 fuel oil/charcoal - 1.0 c/ 5:1 for brick kiln, 8:1 for earth kiln. Source: Mission estimates.  - 69 - 3.45 The economic value of fuelwood ex-plantation is lower than the economic value of fuelwood at plant gate due to three costs, namely (a) the equivalent average cost of transportation of fuelwood (which would be low for a plantation adjacent to the industrial plant); (b) the capital costs of charcoal production kilns, which will be a relatively minor oncost to fuelwood, and (c) the capital cost of converting industrial boilers to han- dle and burn woodfuels, which could impose a substantial oncost. However, it would still be economically attractive to convert to woodfuels as illus- trated in the following two examples: (a) the estimated cost of conversion to automatic wood-fired boilers at the Nyanza Textiles Ltd plant is US$1.8 million (in 1982 pri- ces) for consumption of 60,000 tons per annum of fuelwood, which is equivalent to an annual oncost of USh 7,600/hectare on the required plantation area of 2,500 hectares.13/ In a cof fee- growing area, the total economic cost of land used for fuelwood and this plant conversion is about USh 67,600/hectare, and the required economic value of fuelwood to give an economic return of 10% would be about USh 2,800/m3 solid (Table 3.5). The estimated economic value of fuelwood in this use is about USh 8,000/m3 solid (Table 3.6) under the same assumptions, which is almost three times the required economic value; and (b) the estimated cost of conversion to charcoal-burning kilns at the Tororo cement works is about US$650,000 (in 1982 prices) for consumption of 25,000 tons per annum of charcoal, which is equiv- alent to an annual oncost of USh 1,850/hectare for charcoal pro- duction in brick kilns on the required plantation area of 3,600 hectare. In a tea-growing area, the total economic cost of land used for fuelwood and this plant conversion is less than USh 12,000/hectare, and the required economic value of fuelwood to give an economic return of 10% would be about USh 500/m3 solid (Table 3.5). The estimated economic value of fuelwood in this use is about USh 5,270/m3 solid (Table 3.6) under the same as- sumptions, which is ten times the required economic value. 3.46 This analysis of fuel substitution by fuelwood therefore con- cludes that fuelvood is a major potential source of indigenous energy that is economically attractive to produce for utilization in industrial and agro-industrial plants. The strategy for development should be to develop supplies for specific markets, especially particular plants, and to opti- mize fuel production for each plant in terms of form (fuelwood or char- coal), opportunity cost of plantation land, transportation distance for fuelwood (see the Butende brick works example in paragraph 2.12) and in- vestment costs in facilities for plant conversion and charcoal production. 13/ At a 10% discount rate and exchange rate of USh 100 = US$1, over a plantation life of 36 years with an annual yield of 24 tons/hectare.  - 70 - D. Other Indigenous Energy Sources 3.47 In western Uganda there are a number of areas which show promising geothermal potential. Howvever, as no detailed geothermal exploration has yet been completed, development of this potential is still an uncertain and long-term prospect. Several institutions are also looking into a number of renewable energy options in Uganda, including solar crop drying, solar heating and cooling, biogas technologies for rural energy needs, ethyl alcohol production from sugarcane and cassava, and the use of wind energy. None of these is expected to have a significant impact on Uganda's overall energy situation during the 1980s or even over the longer-term, but some could make a contribution in specific applications (e.g., solar drying and water heating), in certain industries (e.g., use of bagasse by the sugar industry) and isolated areas (e.g., biogas). Geothermal Energy 3.48 In Uganda, thermal waters are found in both Precambian and Ter- tiary rocks. Most thermal sprirngs issue fran the joint planes of granite gneisses of Precambian age along an-uplift axis some 30 kilometers east of the Albertine Rift Valley. This axis connects with a volcanic field (Katwe) to the south. Hot springs in Uganda tend to occur in areas asso- ciated with earth movements. This is especially true in the Rift Valley where some faults are still active. The western Uganda area, where most of the hot springs occur, has been divided inito three geothermal potential fields: (a) the Katwe volcanic field to the south; (b) the Buranga field at the foothills of the Rwenzori mountains; and (c) the Kibiro field in the northern part of the Rift Valley near Lake Albert. 3.49 Of these, the Katwe volcanic field is the most promising. The area is famous for its explosive craters and saline lakes. Although hot water manifestations are not numerous in this field, the Na/K atomic ratio gives a postulated subsurface lemperature of 2300C. Geologically, the rocks are volcanic tuffs, agglomerates and sandy clays. Structurally, the field is situated in a fault block terrain, a condition which is favorable to the formation of a good geothermal field. The Katwe field is also well located: it is only 35 kilometers from the terminus of a 132 kV transmis- sion line at Kasese; waste heat from the generating plant could be profit- ably utilized to concentrate mineral brines located in the crater lakes of the area; and a geothermal power plant could provide energy to the proposed salt industry at Lake Katwe (although the feasibility of this project is still in doubt). By comparison, the Buranga and Kibiro fields are located in sparsely-populated and remote areas, with little potential for local consumption of power. 3.50 A UNDP study, completed in 1971, concluded that Uganda's geo- thermal potential could be as much as 450 MW and that further surveys and exploration were justified. However, the Mission does not consider this an immediate priority, given the country's resource constraints and the availability of sufficient hydroelectric capacity to meet the country's power requirements for many years to come. In Uganda hydroelectricity is  - 71 - both reliable and economic, conditions which are unlikely to be matched by geothermal development. Solar Energy 3.51 Drying is essential for preserving most food crops, beverages, tobacco, fish, meat and sawn timber. Solar drying can and is used for dry- ing many crops. Improvement and the quickening of the process of solar drying could preserve some foods that are at present wasted and also save some (wood) energy. However, the smoking of fish not only dries the fish but adds flavor to it. Also several drying processes have to be continuous to achieve satisfactory results as for example the drying of tobacco and tea. Solar tobacco kilns have been designed in several countries but they need a back-up heating system at night and also an electric fan to circulate the air which necessitates being connected to an electric power supply. The Uganda forest research station has produced prototype solar timber kilns (again with a fan) and with a little more work these could be commercially produced. Therefore, the efforts to improve solar drying should be supported but concentrated on food crops and on crops where con- tinuous drying is not essential. 3.52 Solar water heating can save considerable energy in hotels, hos- pitals, factories and houses. Commercial solar heaters are already on the market in many countries and licences could be obtained to manufacture them in Uganda. Therefore, very little, if any, new research needs to be under- taken in this field. The Government should consider providing special tax incentives to encourage use of solar heaters in new and existing commercial buildings, hospitals, schools, factories and houses, especially in areas not served by UEB's transmission network. 3.53 Much work has been done on developing solar cookers and success- ful stoves have been produced. However, they have not been accepted by the housewife because they can only be used from mid morning to mid afternoon without installing an expensive storage system. Therefore, work on solar cookers should be given a low priority. Other Renewable Energy Options 3.54 Crop residues. The 1980 production of crop residues in Uganda is estimated to be in the order of 4.1 million tons, of which about 30% (1.2 million tons) is available for fuel.14/ This is equivalent to about 0.4 million TOE. Although a higher proportion of crop residues could be used for fuel in the short term, this would be to the detriment of the soil or at the expense of other uses such as animal feed and fertilizer. Actual consumption of crop residues for fuel is obviously very difficult to esti- mate but is probably less than 100,000 TOE or only 2% of total woodfuel consumption in the country. For cooking, crop residues are generally con- sidered inferior to wood as they burn rapidly and the fire has to be tended 14/ These estimates are derived from crop production data, using crop- specific "residue" and "fuel availability" factors.  - 72 - constantly. However there are some specific applications, such as the use of coffee husks to fire brick kilns and bagasse for sugar-factory boilers, which are well worthwhile pursuing. For example, after rehabilitation of the Lugazi works of the Sugar Corporation of UJganda, bagasse will provide all of the plant's boiler energy requirements and steam generation for the turbo generators should be such that all internal electrical demands will be satisfied as well. Indeed, the plant: is expected to produce a surplus of electricity, possibly for sale to UEB. Bagasse should also be utilized at the other two sugar factories to be rehabilitated. 3.55 Alcohol. Uganda already grows sugarcane for food purposes. How- ever, an alternative use of sugarcane could be to produce alcohol, pri- marily for blending into gasoline. The economics of this option should be studied carefully before any decision is taken to divert arable land from food and export crops for this purpose. The relative benefits of producing sugarcane to substitute for sugar imports or for petroleum imports should also be considered. Experience with alcohol production in Kenya has been disappointing so far, with very high costs and no net saving of foreign exchange. However, progress in Zimababwe has been more promising and costs are estimated to be below the landed price of premium gasoline.15/ 3.56 Biogas. Biogas cannot solve the energy requirements of the sub- sistence sector or the urban popu:Lation for they do not have enough dung or capital to build and run self sustaining units. However, biogas digest- ers could potentially be useful on large farms and where sewage is collect- ed, provided units were properly maintained and the process monitored. Therefore, only a moderate amount of time and effort should be put into the development of biogas units. 3.57 Wind. Wind pumps for pumping of water are already in commercial production in neighboring countries. Therefore, no new research is really required. A survey is required to locate areas in Uganda that have enough wind energy to support wind pumps aLnd are in need of water pumping. 3.58 Peat and papyrus. There are substantial peat and papyrus depo- sits in central Uganda, especialLy around Lake Kyoga. However, these products are bulky forms of energy and are located far from potential consumers. Although further work on the economics and practicability of production and marketing is probably justified, the Mission does not con- sider this a high priority. 15/ These alcohol programs are discussed in the energy assessment reports on Kenya (No. 3800-KE, May 1982) and Zimbabwe (No. 3765-ZIM, June 1982).  - 73 - E. Major Recommendations on Supply Options 3.59 The Mission recommends that: (a) Owen Falls Power Station should be rehabilitated and uprated on a priority basis, to help meet the expected growth in electricity demand during the second half of the 1980s. The proposed feasi- bility study should be started immediately so that the works can be completed by 1986 (paragraph 3.5); (b) preparation for a second power station, to supply Uganda's needs during the 1990s and possibly for regional exports, should also be started immediately. The earliest a second station could be brought on stream is probably by 1991. Project selection should be based on a least-cost power development program up to the year 2000 (paragraph 3.7); (c) the Government should take a more active role in monitoring petroleum imports. The main objective should be to minimize the cost of supplies. The options available are diversification of supply sources, utilization of alternative transport modes and routes, arnd maintenance of a product stockpile (paragraphs 3.11 to 3.25); (d) further work to assess the extent and economic viability of petroleum resources in the Lake Albert area is justified. A petroleum exploration package should be prepared to determine the interest of oil companies in further exploration and develop- ment (paragraphs 3.26 to 3.29); (e) the Forestry Department should undertake an inventory of tree stocks, complemented by a survey of consumption trends and projections for all wood products to help pinpoint the areas of immediate shortages (paragraph 3.32). Measures can then be taken to help remedy this situation. The Mission has prepared prelim- inary proposals relating to agro-forestry, urban and industrial plantations, and forest management. Special emphasis needs to be given to low-potential agricultural areas, such as Karamoja (paragraphs 3.33 to 3.43); and (f) the economics of new renewable energy options should be studied carefully before any major investments are undertaken. In the short ternm, one of the most promising options is the use of bagasse, which will make sugar-processing factories virtually self-sufficient in energy. The Government should also consider providing special tax incentives to encourage the use of solar water heaters, especially in zreas not served by UEB's transmis- sion network (paragraphs 3.47 to 3.58).  - 74 -  - 75 - IV. ENERGY PRICING POLICIES 4.1 Energy prices have a major impact on energy consumption patterns and levels. As outlined in Chapter I, there was a significant decline in all retail energy prices in Uganda during the 1970s. This led to waste and inefficient use, reduced the sector's contribution to public revenues, and encouraged smuggling of petroleum products to neighboring countries. Since 1981, the Government has substantially increased petroleum prices. How- ever, the potential for profitable smuggling has not been fully eliminated and there are still significant price distortions between products. Com- parable adjustments have not been made in domestic electricity tariffs and they are now substantially below the long-run marginal cost of supply. Woodfuel prices, which are not officially controlled, have risen (in real terms) over the past two years, reflecting the increased prices of alterna- tive fuels, higher transport costs, security bottlenecks and the steady de- pletion of the most accessible and economic forest resources. This Chapter provides a more detailed analysis of the present distortions in energy prices and recommends appropriate adjustments in the Government's pricing policies. These adjustments are intended to encourage improvements in efficiency, through conservation and substitution. More realistic energy prices (especially for electricity) will also help ensure that consumers contribute towards the costs of the major energy sector investments required to meet projected demand during the 1980s and beyond. A. Electricity Tariffs 4.2 Uganda's present tariff structure and levels are based on histor- ical factors. However, Uganda is now facing the need to embark on a program of investment in new generation facilities and therefore there are economic costs to meeting rising demand which render the present tariff system to be out of date, as shown below. UEB presently keeps its tariff level under review in its efforts to maintain the financial viability of the organization. However, a review of tariffs according to economic prin- ciples is also required and should be undertaken immediately after prepara- tion of the long-term power development program (paragraph 3.7). UEB Tariff Structure 4.3 UEB's tariff schedule for domestic sales is given in Annex VII and incorporates the latest revisions brought into effect on July 1, 1982. In addition, there is a government levy on electricity sales of 10% applic- able to all consumer categories except industrial. UEB's tariff structure has remained unchanged since 1961, shortly after eight of the ten genera- tion units at the Owen Falls Power Station had been installed. At that time the level of sales was running at about 50% of firm generating capability from the Station, and therefore the medium-run marginal cost of power was virtually zero. The structure was designed to promote energy consumption by incorporating regressive tariff rates for additional blocks of energy consumption and for additional increments of demand. At that time such a  - 76 - structure had economic justification. However, at present the situation facing UEB is different and the priority is to increase generation capacity so that there is now a substantial marginal cost in the long run for sup- plying increases in demand. The present tariff structure is therefore giving the wrong signals to consumlers, and the element of regressive incre- mental rates should be removed. 4.4 UEB applies a uniform tariff rate nationwide for each consumer category, irrespective of cost of supply. UEB recognizes that a result of this policy is that consumers in isolated areas served with power generated from small diesel sets are heavi:ly subsidized. For example, the Mission estimates that the economic cost of power delivered to consumers in these areas is the equivalent of about US cents 20/kWh, of which about US cents 16/kWh is accounted for by fuel costs based on the import-parity price for fuel. At a shadow exchange rate of USh 200 = US$1, the economic cost is about USh 40/kWh, and even at the official "wjindow one" exchange rate the economic cost is about USh 20/kWh. In contrast, the domestic tariff is USh 3.375/kWh for the first 6 kWh per month and USh 0.4725/kWh for all units consumed above 24 kWh per month (excluding the government levy). The low voltage tariff for industries and commercial establishments is between USh 0.3713/kWh and USh 0.4725/kWh. Clearly, economic schemes for connecting these areas to UEB's main transmission network (paragraph 3.10) should be given priority. UEB plans to phase out all diesel stations by 1985, al- though achievement of this target will depend upon progress in extending the transmission network. The associated costs of expanding system genera- tion capacity should be reflected in the general tariff levels (paragraph 4.11). 4.5 At present the power factors for demand by industrial plants are generally very low, below 50% and even as l-ow as 20% (paragraph 2.7). These low power factors cause transmission and distribution losses for UEB and increase the demand on UEB's system, and undoubtedly are a significant component of the high level of system losses (paragraph 1.10). Once system demand increases to UEB's present. supply capacity, projected to occur in 1986 (Table A4.5 of Annex IV), additional demands imposed on the system through low power factors would, unless proper measures are taken, have to be met by investment in new capacity. The 'least-cost option is through uprating the Owen Falls Station at an esltimated capital cost of US$400/kW. The equivalent month:Ly cost for adding an increment of 1 kW to system capa- city is US$4.0,1/ or USh 400 at the "window one" exchange rate of USh 100 = US$1 and USh 800 at a shadow exchange rate of USh 200 = US$1. The pre- sent demand charge varies between USh 0.35 and USh 0.47/kVA, and therefore is less than 0.1% of the economic cost. At the present low tariff the demand charge does not give industrialists sufficient financial incentive to install power-factor correction equipment. The Mission recommends that UEB should increase maximum demand charges as soon as possible to the eco- nomic cost of low power factors, as well as implementing its campaign to improve power factors through providing technical assistance. 1/ At a 10% discount rate over a 40 year economic life and with 20% trans- mission and distribution losses, at 1982 prices.  - 77 - UEB Tariff Level 4.6 The underlying principle adopted by UEB for setting tariff levels is that revenue should be sufficient to cover operating costs, depreciation or loan repayments (whichever is greater), loan interest and a reasonable surplus for contingencies and contribution to capital development. For the period 1961 to 1979, UEB was able to keep to this principle without in- creasing its tariffs despite the considerable cumulative effects of infla- tion during this period.2/ The main reasons for this outcome were the absence of major capital expenditure on power generation facilities, al- though there was substantial extension of the transmission network, and the maintenance of an increasingly overvalued exchange rate for the Ugandan currency which prevented increases in the real costs of imported equipment and fuel from being passed on to UEB. 4.7 Eventually the irresistible force of cost inflation and finally the massive devaluation of the Ugandan currency in June 1981 precipitated UEB into a position of extreme financial difficulty. The severity of the situation was aggravated by the need to import substantial quantities of materials, parts and equipment to execute repairs to the transmission and distribution system to make up for a backlog of maintenance work and the effects of destruction and looting during 1979 and subsequently. Due to the total unpredictability of future costs at that time, UEB embarked on a series of increases in tariffs in increments that it judged to be the maxi- mum tolerable. Accordingly, UEB increased tariffs by 50% each time in July 1980, January 1982 and July 1982, which resulted in a cumulative increase of almost 240% over the pre-July 1980 level. 4.8 Despite the increases in tariffs, it is still not certain that UEB's revenues are adequate to meet commitments even excluding contribu- tions to capital development. According to unaudited accounts for the year ending December 31, 1981, UEB earned a net surplus on revenue of USh 28 million (US$280,000 at the "window one" exchange rate) after charging loan interest. However, this statement does not reflect the full effect of the massive increase in foreign loan repayment obligations in Ugandan currency terms due to the devaluation. For example, the authorized loans incurred up to 1972/73 totalled USh 382.1 million, whereas the balance of these loans at the end of 1981 was about USh 1,124 million. This trend will have continued since 1981 due to the further devaluations of the Ugandan cur- rency. 2/ Although tariffs remained constant between 1961 and 1979, average revenue per kWh sold increased steadily between 1969 and 1979 (Table A1.3 in Annex I). This increase was due probably to the effect of a decreasing tariff structure on average revenue from sales, particularly industrial sales, as sales declined during this period. Therefore it could be said that the present tariff structure did work to UEB's ad- vantage during the 1970s.  - 78 - 4.9 For the future, tariff levels should reflect the cost of increas- ing power generation capability to meet rising demand. Under Ugandan cir- cumstances, in which the cost of new power capacity will be substantially greater than the historical cost of existing power capacity, economic tariff levels would generate substantial cash surpluses that would be available for financing the local currency contributions to capital costs. The obvious benefit from such an outcome is that the development of power generating capacity would not impose any burden on the government budget. 4.10 The Mission has made an indicative estimate of the long-run marginal cost of power in Uganda at 1982 prices, as represented by the average incremental economic cost of power from new generation facilities. The Mission examined two scenarios, in the first of which capacity is in- stalled to meet only increases in domestic demand and existing export con- tracts, while in the second scenario extra capacity is installed for addi- tional exports of 120 MW from the early 1990s. These scenarios are con- sistent with the projections in Tables A4.5 and A4.6 of Annex IV. The evaluation is detailed in Annex VIII and is summarized in Table 4.1. The lowest estimate of the long-run marginal cost of power without additional exports is about US cents 5/kWh sold, and is higher at discount rates above 10%. At a 10% discount rate, the cost is 15% lower with the inclusion of additional exports, but at higher discount rates the estimates are less sensitive to the export assumptions. The discount rate used should be the opportunity cost of capital for Uganda, which is extremely difficult to estimate under the present circumstances. A value of 10% is commonly used for project evaluation in neighboring countries. Table 4.1: INDICATIVE ESTIMArES OF LONG-RUN MARGINAL COST OF POWER IN UGANDA To meet increases in domestic To meet demand and additional increases in exports of 120 MW domestic demand from 1991 At discount rate of At discount rate of :L0% 15% 20% 10% 15% 20% Long-run marginal cost: - US cents/kWh produced 4+.1 5.9 7.7 3.5 5.5 7.8 - US cents/kWh sold a/ 4.9 7.1 9.2 4.2 6.6 9.4 USh/kWh sold at USh 100 = US$1: 4.9 7.1 9.2 4.2 6.6 9.4 USh/kWh sold at USh 200 = US$1: 9.8 14.2 18.4 8.4 13.2 18.8 a/ Assuming losses to be 20% of sales. Source: Annex VIII.  - 79 - 4.11 Present tariff levels are only a small proportion of the economic cost of power as represented by the long-run marginal cost. The Mission estimates that the average yield during 1983 from the present (post-July 1982) tariffs will be about USh 1/kWh sold. At a 10% discount rate, the economic cost of power is about USh 5/kWh sold at USh 100 = US$1, and USh 10/kWh at USh 200 = US$1. A fivefold or tenfold increase in real terms (i.e., excluding the effect of general inflation) on top of the recent tariff increases is commensurate with the order of magnitude of devaluation and inflation in recent years. The Mission recommends that the Government and UEB should plan to raise tariffs to a level that corresponds to the economic cost of power as soon as is practicable, and at least by 1991 when the next new hydroelectric station will be required.3/ While such major tariff adjustments will obviously be politically difficult to implement, they are consistent with similar price increases already introduced by the Government for petroleum products. In some other countries in the region, a "lifeline" low tariff rate is adopted to enable the poorest households to obtain electricity at affordable rates for essential uses, particularly lighting. This approach would be justified in Uganda to avoid imposing an undue burden of price increases on those least able to pay. Sales to Kenya 4.12 The present rates for the sale of power to Kenya were negotiated in 1980, although there is not any explicit provision for renegotiation of rates during the 50-year term of the contract. Under the present terms, Kenya pays KSh 0.075 (US cents 0.63)/kWh for supplies up to 30 MW at 90% load factor, and KSh 0.135 (US cents 1.13)/kWh for additional units. There is also a capacity charge of KSh 30.25 (US$2.52)/kW per annum with a mini- mum payment for 30 MW. The average yield on the first 30 MW of supplies is thus US cents 0.66/kWh. In the past, when Uganda had substantial surplus capacity, sales to Kenya at even lower rates than the present levels were justified as extra income from power supplied at zero marginal cost, and these sales constituted an important source of income for UEB and of for- eign exchange for Uganda.4/ The terms are also favorable to Kenya since the rates are less than 10% of marginal generation costs in Kenya. 4.13 However, once the demand on the Ugandan system surpasses existing generation capacity, projected to occur in 1986, power supplied to Kenya will have an opportunity cost for Uganda equal to the cost of installing new capacity to make up for power sent to Kenya, which the Mission esti- mates to be at least US cents 5/kWh in 1982 prices (paragraph 4.10). Even 3/ A fivefold increase in seven years would require compound increases of 26% per annum, and a tenfold increase would require 39% per annum. 4/ Before 1980 Kenyan sales constituted about 10% of total UEB revenue, but this portion rose to 37% in 1981 and 49% in 1982. The proportion is expected to remain high during 1983, especially as long as UEB is allowed to convert revenue from Kenyan into Ugandan currency at the "window two" exchange rate.  - 80 - with these costs fully reflected in the export: tariffs, Ugandan power would remain a low-cost source of supply for Kenya. B. Petroleum Prices Retail Prices and Taxes 4.14 The Government of Uganda sets retail prices for premium and regular gasoline, kerosene, and auto. diesel. Fuel oil, jet fuel and lubricant prices are set by the oil companies, following discussions with the Government. As discussed in Chapter I (see Table 1.3), the government- controlled petroleum prices have been increased substantially since 1981, largely to reflect the devaluation of the Ugandan shilling. The retail prices are "rationalized" in terms of a price formula (see Table 4.2), which makes provision for the cost of crude oil, processing, transport and oil company margins, and taxation. In practice, however, the formula is distorted by three factors: (a) the base cost of crude oil used in the formula has not been adjusted since 1981 and therefore does not represent present market conditions; (b) the exchange rate used in the formula is only adjusted periodic- ally,5/ and therefore increases in oil company costs (in terms of Ugandan shillings) resulting fromn interim devaluations have to be "refunded" in the form of a rebate;6/ and (c) the retail prices set by the Government do not always correspond exactly to the prices derived from the formula, necessitating balancing subsidies/surcharges on the ad valorem tax rates. The latest price formula, relating to the new petroleum prices introduced in December 1982, is not yet available. However, the formula prevailing at the time of the Mission (October/November 1982) is shown in Table 4.2 for products obtained from the Mombasa Refinery. An illustrative computation for fuel oil based on the same formula is also included in the table. 5/ The pricing formula is supposed to be adjusted for exchange rate move- ments every six weeks. However, in practice, the adjustments are less frequent and generally only after major exchange rate/price changes. 6/ The rebate covers the difference between import costs at the pegged rate used in the formula and at the prevailing exchange rate. It is approved by the Treasury and deducted from taxes paid to the Government by the oil companies. These payments are made at ten-day intervals on the basis of sales from main depots.  - 81 - Table 4.2: RETAIL PRICE FORMULA FOR PETROLEUM PRODUCTS (pre-December 1982 in USh/liter) a/ Gasoline Fuel Premium Regular Kerosene Diesel Oil Product value 24.30 24.30 24.30 24.30 24.30 Processing fee at Mombasa Refinery 2.21 2.21 2.21 2.21 2.21 Pipeline transport Mombasa/Nairobi 2.32 2.32 2.32 2.32 - Nairobi handling 0.30 0.30 0.30 0.30 - Transport Nairobi to Malaba 3.29 3.27 3.32 3.35 5.68 c/ Transport losses 0.49 0.49 0.49 0.49 0.49 Cif value at Malaba 32.91 32.89 32.94 32.97 32.68 Customs duty 24.69 24.67 - 10.97 - Transport Malaba to Kampala 1.81 1.81 1.26 1.37 1.37 Company margin (22%) b/ 7.64 7.63 7.54 7.55 7.49 Sub-total 67.05 67.00 41.84 52.88 41.54 Sales tax 47.95 44.22 6.65 11.63 - Ex-depot price 115.00 111.22 48.49 64.51 41.54 Dealer margin + local transport 5.00 5.00 5.00 5.00 5.00 120.00 116.22 53.49 69.51 46.54 Subsidy 0.00 6.22 13.49 9.51 6.54 Established price at retail outlets 120.00 110.00 40.00 60.00 40.00 a! Based on a pegged exchange rate of USh 76.67 = US$1. Unit rates for product value, processing fee and transport costs are derived in Annex V. b/ Operating 15%; development 7%. c/ Transport Mombasa to Malaba. Source: MPPT.  - 82 - 4.15 Taxes are levied on gasoline, kerosene and auto. diesel. All other liquid petroleum products are exempt from taxation. Customs duty is computed on the cif value at Malaba at an ad valorem rate of 75% for gaso- line and 33.3% for kerosene; sales tax is computed on the bulk depot value at an ad valorem rate of 71.5% for premium gasoline, 66% for regular gaso- line, 15.9% for kerosene and 22% for auto. diesel. The product values for taxation purposes are derived from the price formula shown in Table 4.2. The effective tax rates are, however, affected by both the rebates and subsidies/surcharges noted in paragraph 4.14. As shown in Table 4.3, net taxes as a proportion of retail prices prior to December 1982 were only 49% for premium gasoline and 38% for regular gasoline, while kerosene and fuel oil were heavily subsidized. Although a revised formula relating to the new petroleum prices introduced in December 1982 is not yet available, it is clear that the net tax rates have been increased on all products and that the differences in tax raties between products have been reduced. In particular, the subsidy on kerosene has been eliminated. However, there is still an effective subsidy on fuel oil. These trends are illustrated in Table 4.3 on the simplifying assumption that there was no change in costs or the prevailing exchange rate. 4.16 The effect of pegging the exchange rate used in the price formula is to provide a subsidy to consumers whenever the Ugandan shilling is de- valued. In other words, government revenue absorbs the effect of exchange rate movements and related price adjustments are entirely retroactive. To avoid this revenue impact, the Mission recommends that petroleum prices be adjusted wherever there is a major change :Ln the exchange rate. Under present conditions, the Government should also consider fixing the exchange rate used in the formula above the prevailing rate to anticipate future depreciation of the Ugandan shilling and thereby avoid more frequent price increases. Foreign Exchange Payments to Oil Companies 4.17 The availability of foreign exchange for payments for imported petroleum products is ensured through an agreement between the Government and the oil companies. According to this agreement the Government commits itself to making available foreiLgn exchange at a certain rate for as long as is sufficient to meet the cost of im ortation of bulk liquid products and other petroleum-related commodities,_/ oil company services in Uganda and repatriation of dividends. Under this agreement the Government origi- nally in 1981 provided US$3 million per week. Subsequently, the amount has been adjusted periodically to reflect changes in demand for petroleum products and supply costs. In June 1983, the foreign exchange allowance was down to US$2 million per week. 7/ Including lubricating oils, greases and other special oil products, bitumen, chemicals, capital goods, containiers and spare parts.  - 83 - Table 4.3: NET TAXATION ON PETROLEUM PRODUCTS a/ (in USh/liter) Before price increase in After price increase in December 1982 b/ December 1982 c/ Retail Net Net tax as Retail Net Net Tax as price tax % of price price tax % of price Premium gasoline 120 59.21 49 150 89.21 59 Regular gasoline 110 42.25 38 140 72.25 52 Kerosene 40 (20.10) (50) 80 19.90 25 Auto. diesel 60 (0.16) - 90 29.84 33 Fuel oil d/ 40 (19.67) (49) 50 (9.67) (19) a/ Net taxation is customs duty + sales tax - subsidy - rebate. Negative net taxation (subsidies) are shown in parentheses. b/ Net taxes have been calculated for the period October 11 to November 20, 1982, when the prevailing exchange rate averaged USh 100.97 = US$1 com- pared to the pegged rate in the price formula of USh 76.67 = US$1. The rebate has been calculated accordingly as 32% of costs as shown in Table 4.2 (including oil company margins but excluding customs duties) or ap- proximately USh 13.1 to USh 13.4/liter (depending on the product). c/ On the simplifying assumption that costs and the prevailing exchange rate remained unchanged. d/ Fuel oil is not subject to the same price formula as the other products, but a notional computation is included for comparative purposes. Source: Mission estimates. 4.18 The oil companies have instituted an independent audit of their operations to serve as a basis for controlling the foreign exchange allow- ance amongst themselves. The allocation of exchange is based on agreed historical market shares (paragraph 1.28). Each company calculates the minimum amount of bulk liquid products to be imported from 80% of its foreign exchange allocation at Kenyan minimum export prices.8/ Each com- pany is obliged to import its respective minimum quantity of bulk liquid 8/ The use of one standard set of prices favors those companies that im- port products from relatively cheaper sources.  - 84 - petroleum products in order to keep its market share. The remainder of the foreign exchange allocation is used for payments for other goods and serv- ices. If a company fails to import the required minimum quantity, the allocation of foreign exchange to that compaly is reduced in the following three months by the proportion of the shortfall. The re-allocated allowance is distributed amongst the other companies that imported more than the required minimum quantity of products to make up for the shortfall. 4.19 Before the agreement on foreign exchange allowance for petroleum product imports was brought into operation, the oil companies extended credit to the Government to cover delays in payments for such imports. Accumulated arrears on payments for petroleum imports reached US$50 million by the end of May 1981, equivalent to about six months supply. This ar- rears was subsequently reduced to US$12 million, currently equivalent to six weeks' supply, and it has been frozen at this level by agreement with the oil companies. A revolving credit is also extended by the oil compa- nies through the pattern of foreign exchange payments. At present the Bank of Uganda transfers US$1.5 million to New York each week, and at the end of each thirteenth week the Bank transfers an additional US$6.5 million, for a total payment of $26 million in each three month period. This revolving credit is equivalent to a constant level of about two weeks' supply, and thus the total credit extended by the oil companies is equivalent to eight weeks' supply. At present, foreign exchange transfers are approved only after presentation of invoices by the oil companies for purchases of products. The oil companies in Uganda supply the local currency to commer- cial banks for the purchase of foreign exchange. 4.20 The product value usel in the price formula (paragraph 4.14) is derived from the net value of products obtained from the Mombasa Refinery, after resale of surplus product (fuel oil), from a representative blend of crude oils supplied by the oil companies :Eor refining (Annex V). The current product value was established in January 1981 at the then weighted average market price for crude oil ex-Middle East of US$36.42/barrel, and the value of crude oil used in the formula has not been changed since even though crude oil prices have declined. As illustrated in Table 4.4, the current weighted average officiaL selling pr:Lce of crude oil purchased for Uganda declined by about US$2.2/barrel (6%) from January 1981 to February 1983; subsequently, the price hias fallen even lower to below US$30/barrel in mid 1983. In so far as these cost savings have been realized by oil companies using the Mombasa Refinery, they should have been passed on to Uganda through lower invoiced values (in terms of foreign exchange) for crude purchases. A related adjustment should also be made in the price formula, especially for the calculation of oil company margins.  - 85 - Table 4.4: OFFICIAL SELLING PRICES OF CRUDE (ex-Middle East in US$/barrel) Crude Oil Type January 1981 February 1983 Arabian Light 32.74 34.00 Qatar Marine 37.75 34.56 Murban. 39.95 34.49 Dubai 38.31 34.00 a/ Weighted average 36.42 34.27 a/ Approximate price. Sources: (1) Oil companies operating in Uganda (for January 1981 prices). (2) World Bank, EPD (for February 1983 prices). 4.21 Under the pricing formula the oil companies receive a margin computed as 22% of the value of premium and regular gasoline, kerosene and auto. diesel delivered to Kampala (excluding customs duty). The margin covers company operating expenses and profit in Uganda (15%) and an allow- ance for rehabiLitation and development of assets in Uganda (7%). This margin is not directly related to the actual operating and capital expendi- tures by the oil companies. The value of this margin for the 1982 level of product imports is shown in Table 4.5. Table 4.5 : OIL COMPANY MARGINS IN 1982 Conversion Marginal Value of 1982 Imports Factor Rate margin Product (tons) (liters/ton) (USh/liter) (USh million) Gasoline 43,900 1,330 7.64 446.08 Kerosene 28,900 1,220 7.54 265.85 Auto. diesel 49,500 1,180 7.55 441.00 Total 122,300 1,152.93 Source: Mission estimates.  - 86 - The oil companies obviously generate revenue from other products as well and are able to allocate profits between Kenyan and Ugandan affiliates, especially on products purchased on the Kenyan market. Therefore, the Ugandan Government is not able to assess the actual level of profitability of the oil companies operations in the cou;ntry, nor is it possible to ascertain precisely the use of the foreign exchange allowance. Another shortcoming is the inclusion of all operating expenses in the oil company margin. In practice, these expenses are unlikely to remain a fixed per- centage of supply costs, independent of crude prices and import volumes. As an alternative arrangement, it might be better to separate out these expenses and account for them directly, adjusting the margin from a gross to a net basis. Local expenses should be expressed and reimbursed in Ugandan shillings, without automatic protection against exchange rate fluc- tuations. In addition, the Government should take measures to ensure that the margin allowed for capital expenditure, currently 7%, is used only for this purpose or is reduced accordingly. More generally, the Mission recom- mends that the Government reviews the formula for reimbursement of all operating and capital expenditures with the oil companies. Import-Parity Prices 4.22 The present retail prices for petroleum products do not neces- sarily reflect economic cost, whiLch in Uganda's case is represented by the import-parity prices of products. The computation of import-parity prices for products according to source of product, namely Middle East market, Mombasa Refinery or Kenyan market, is summarised in Table 4.6. Import- parity prices are shown for alternative assumptions about the foreign ex- change rate, firstly at the official ("window one") rate of USh 100 = US$1, and secondly at a shadow exchange rate of USh 200 = US$1. There is little difference at present between parity prices for white products obtained from the Middle East market, the Mombasa Refinery, or the Kenya market, but the parity price of fuel oil obtained from the Refinery is about 30% higher than from the Middle East market or the Kenyarn market. 4.23 The most accurate indication of economic cost is the import- parity price for products obtained from the Middle East market valued at the shadow exchange rate. The extent of cross-subsidization between petro- leum products and price distortions relative to economic costs at current (post-December 1982) retail prices is shown in Table 4.7. Even after the latest increases, the retail prices of fuel oil, kerosene and auto. diesel are substantially lower than import-parity prices. The retail prices of gasoline are higher than the import-parity prices. There is some cross- subsidization of fuel oil from the other products due to the uniformity of product values from the Mombasa refinery. There is no significant cross- subsidization between gasoline, diesel and kerosene. 4.24 In conclusion, the present distortions in the retail prices for petroleum products arise primarily from three factors: (a) the overvalued exchange rate; (b) the cross-subsidization of products at the Mombasa Refinery; and (c) differences in effective tax rates. The Mission there- fore recommends that the Government continues to adjust petroleum prices in line with exchange rate changes. The Government should also realign the  - 87 - Table 4.6: DORT-PARITY PRICES FOR 1ETR1DIM PR0DUTS a/ (in US$/liter, unless specified otherwise) Premium Reular Auto. Ftp1 Product Source Gasoline Gasoline Keroene Miesel Oil A. Middle East market Product value Lf Maibasa (US$/ton) b/ 383 373 380 339 204 Convernion factor (liters/ton) 1,330 1,330 1,220 1,180 1,060 Product value cif Manbsa 0.288 0.280 0.311 0.287 0.192 Trarsport Manbasa to Malaba c/ 0.083 0.083 0.083 0.083 0.074 Border value: 0.371 0.363 0.394 0.370 0.266 Domestic costs c/ 0.188 0.188 0.181 0.181 0.175 Import-parity price at retail outlet: 0.559 0.551 0.575 0.551 0.441 Import-parity price at USh 100 = US$1 (USh/liter) 56 55 58 55 44 Inport-arity price at USh 200 = US$1 (USh/liLter) 112 110 115 110 88 B. Montasa Refinery Product value cif Manbisa d/ 0.290 0.290 0.290 0.290 0.290 Processirg fee 0.029 0.029 0.029 0.029 0.029 Transport Manbasa to Malaba c/ 0.083 0.083 0.083 0.083 0.074 Border value 0.402 0.402 0.402 0.402 0.393 Danestic costs c/ 0.188 0.188 0.181 0.181 0.175 Import-parity price at retail outlet: 0.590 0.590 0.583 0.583 0.568 Import-parity price at USh 100 = US$1 (USh/liter) 59 59 58 58 57 liport-parity price at TESh 200 = US$1 (USh/liter) 118 118 116 116 114 C. Kenyan market Product value cif Manbasa e/ 0.340 0.330 0.320 0.330 0.170 Transport Mombasa to Malaba c/ 0.083 0.083 0.083 0.083 0.074 Border value 0.423 0.413 0.403 0.413 0.244 Dbmetic costs c/ 0.188 0.188 0.181 0.181 0.175 lInport-parity price at retail outlet: 0.611 0.601 0.584 0.594 0.419 linport-parity price at USh 100 = US$1 (USh/liter) 61 60 58 59 42 Import-pErity price at USh 200 = US$1 (Uih/liter) 122 120 117 119 84 a/ For Decemier 1982. b/ From World Bank, EPD. c/ Fran Table 4.2, converLted at the pegged ewcharge rate of USh76.67 = US$1. d/ Based on crude oil blend cost of US$34/ton and current values of surplus fuel oil, according to fomiula given in AnneK V. e/ Kenya Minimum Export Prices converted at KSh12.5 = US$1. Source: Mission estimates.  Table 4.7: IMPORT-PARITY AND RETAIL PRICES FOR PETROLEUM PRODUCTS (for December 1982) Prices Cross-Subsidization c/ Petroleum Import Retail/Import Effective Ex-Mombasa Ex-Kenyan Product Parity Retail Parity Taxation b/ Refinery Market (USh/1)a/ (USh/1) (%) (USh/1) (USh/l) (USh/l) Premium gasoline 112 150 134 89 6 10 I Regular gasoline 110 140 127 72 8 10 Kerosene 115 80 70 20 1 2 Auto. diesel iiO 90 82 30 6 9 Fuel oil 88 50 57 (10) 26 (4) a/ From Table 4.6 for imports from Middle East market at USh 200 = US$1. b/ Preliminary estimates from Table 4.3. c/ Equal to difference in import-parity prices based on Mombasa Refinery or Kenya Mombasa market and the Middle East market (Table 4.6) at USh 200 = US$1. Source: Mission estimates.  - 89 - taxation structure to offset the cross-subsidization of products. This implies relatively higher tax rates on fuel oil in particular, but also on kerosene and auto. diesel; tax rates on gasoline could possibly be reduced, especially if the Ugandan shilling depreciates further. Taxes on petroleum products make a major contribution to government revenue. Therefore, as the official exchange rate tends towards the shadow exchange rate, it may be necessary on revenue grounds to raise retail prices above import parity (as has already occurred for gasoline). This would also help discourage smuggling and the uneconomic use of petroleum products within Uganda. C. Woodfuel Prices 4.25 Retail prices of woodfuels are not officially controlled and therefore tend to reflect economic costs (see Table 4.9). The major con- cern of the Government is the apparently high level of retail prices for woodfuels in Kampala. As shown in Table 4.8, households are paying up to USh 6.82/kg for fuelwood and USh 21.2/kg for charcoal. 9/ Table 4.8 also shows that only a very small proportion of the price (between 4% and 24%) is accounted for by production costs, assuming that fuelwood is obtained from forests at the official stumpage fee of USh 200/m3. There are a number of factors which account for this differential: (a) as noted in Chapter III, Section C, woodfuels have a high econom- ic value in substitution for alternative energy sources (espe- cially fuel oil), and indeed this is necessary to justify the use of fertile rural and peri-urban land for growing wood. For this reason, it is probably very difficult in practice to acquire wood at the official stumpage fee near to Kampala; (b) transport costs are exceptionally high in Uganda, due to the recent increases in gasoline and auto. diesel prices, the scar- city of foreign exchange to import trucks and spare parts, and the impact of poor road conditions on vehicle depreciation; and (c) significant quantities of fuelwood and charcoal get "hijacked" in transit to Kampala and these losses are recouped by the supplier through higher prices to the consumer. 4.26 In practice, many consumers are probably avoiding these high re- tail prices by collecting their own fuelwood (either legally or illegally) and transporting it in small loads from the forested areas around the city. This, however, is only a short-term solution because it could soon lead to 9/ Retail prices for non-households are substantially lower as they are better able to make bulk purchases directly from the Forestry Department and do their own cutting and transport. Demand is also probably more elastic in the non-household sector due to better access to alternative energy options.  - 90 - Table 4.8: COSTS FOR FUELWOOD AND CHARCOAL MARKETED IN KAMPALA Standing price at forest gate for fuelwood a! (USh/m3) 200 Standing price at forest gate for fuelwood (USh/air-dried ton) 280 Non- Household household Retail price of fuelwood in Kampala (USh/ton) 6,820 1,720 Fuelwood transport and distribution costs b/ (USh/ton) 6,540 1,440 Fuelwood transport and distribution costs (% retail price) (96) (84) Cost of license for charcoal production USh 30,000/year/person Transformation rates for charcoal production using earth kiln 12m3 fuelwood/ton charcoal Production capacity of earth kiln 30 tons/year/person Labor and overhead costs USh 700/ton Production cost of charcoal d/ USh 4,100/ton Non- Household household Retail price of charcoal in Kampala (USh/ton) 21,170 16,940 Charcoal transport and distribution costs c/ (USh/ton) 17,070 12,840 Charcoal transport and distribution costs (% retail price) (81) (76) a/ According to latest revision of forest fees of July 8, 1982. b/ Retail price less standing price. c/ Retail price less production cost. d/ Calculated as 200 x 12 + 30,000 + 700. 30 Source: Mission estimates.  - 91 - a serious depletion of the most accessible wood resources. Obviously, woodfuel prices will come down as transport and security around Kampala is improved and this is a priority of the Government. The stumpage fee should also be reviewed. At present, the fee does offer an attractive financial return (paragraph 3.57) on fuelwood production from land with low economic potential. However, large amounts of such land are unlikely to be avail- able near to urban areas such as Kampala. There is therefore a strong case for raising the stumpage fee to reflect more fully the economic value of wood as a fuel. An initial adjustment to at least USh 1,000/m3 for forests serving urban areas would seem justified. This is unlikely to have any significant impact on retail prices, as the stumpage fee is a relatively small component of total costs and the increase would be partially absorbed by a reduction in the "windfall" profits of distributors. Provided the Government utilizes the revenue generated to strengthen its tree planting and management program, the overall impact on supplies will be positive over the longer term. In the final analysis, the only effective way to reduce retail prices for woodfuels in Kampala is to improve supply. D. End-Use Energy Costs 4.27 The present market price relationships in urban areas between alternative forms of energy display a pronounced gradation which reflects the distortions introduced by government price controls and indicates the existence of considerable non-price barriers to switching between energy forms. A comparative analysis of energy costs for three consumer catego- ries is given in Table 4.9. Analysis for switching fuels in industrial plants is given earlier in this Report (Chapter II, Section B and Chapter LII, Section C). By far the cheapest form of energy at present market prices for lighting, cooking and small-scale power usage is electricity. For example, on an equivalent usable energy basis after allowing for dif- ferences in conversion efficiencies of applicances, the cost of energy for cooking from electricity is about one-nineteenth of the cost from fuelwood, which in turn is about one half the cost from kerosene and L.P. gas. Like- wise, the cost of electric lighting is less than one tenth of the cost of lighting from kerosene. These enormous differences are due to present res- trictions on the supply of fuelwood and to a lag in increasing electricity tariffs in line with domestic inflation. The differences would be reduced if fuelwood supplies to Kampala were to be increased. Further downward pressure on fuelwood and charcoal prices would be exerted if electricity supplies to households became more easily available. To assist the lowest income sections of the community, the Government and UEB should plan to expand the electricity distribution system in towns and to facilitate con- nections to households. 4.28 There is little difference between the cost of energy from fuel- wood and charcoal at present market prices for household cooking, which reflects total substitutability between these forms. However, charcoal is a more expensive form of energy than fuelwood for non-household consumers  Table 4.9: EtD-(EE ENE! awSTS iN IN ARFAS Ead-Use 0ODnrr1ion Retail Effective Market Ccat Econani Cost Effective Econczoic Ccat EnelEy Source . Unit Ene3gy Value Efficiercy Price a/ of Usable Enety at Consumer b/ of Usable Energy (TOE/000 units) (% (USh/milt) (UShOOO/T0E) (tEh/unit) (WbOO/OE=) Urban hmsehold lighting Kerasene liter 0.83 45 80.0 214.2 115 307.9 Electricity - UEB n network kIh 0.25 70 3.7 21.1 10 57.1 - local diesel station lih 0.25 70 3.7 21.1 40 229.6 Urban hseld cooiLng Kerosene liter 0.83 45 80.0 214.2 115 307.9 Electricity - UEB mnin network 14M 0.25 70 1.1 6.3 10 57.1 Fuelwod kg .0.38 13 6.8 137.7 3.1/4.5/8.1 c/ 62.8/91.1/164.0 Charcoal kg 0.77 20 21.7 140.9 9.5 d/ 61.7 LP Gas kg 1.07 55 140.0 237.9 168 295.5 Urban on-household powr Fuelood kg 0.38 25 1.7 17.9 V5/4.11/8,4 c/ 15.8/43.2/88.4 Cbarcoal kg 0.77 40 16.9 54.9 8.6 d/ 27.9 Electricty - UE mnin netork Wh 0.25 70 0.5 2.9 10 57.1 ul. liter 0.95 40 50.0 131.6 90 236.8 a/ Retail prices as of December 1982 fraa Table 1.3 except electricity which is fran TEB Tariff (Am-A VII). b/ Erconamic costs based on followdng estimates with 10% discomt rate ani USh 200 = US$1 dcarge rate: - petrolem products: import-parity prices for supplies fram the Middle East (Table 4.6). - IPG: Kenyan NUP KSh 7.19/liter and traspDrt ard domestic cats. - electricity: long-rm marglnal cat estimete (Table 4.1). - fuelwood: plantation cests. - charcoal: fueLiod cast an laind with opportunity cast of TSh 10,000/ha, transfomation ratio of 12 m3 fuelwood/ton dharcoal. c/ The eonanc values of fuehaod are based on opprtuiity casts of land equai to tSh 10,000/50,000/120,000/ha respectively (Tale 3.5 with 1 m3 solid/0.7 toes). Trarsport caots wder sDre nonaal supply conditiois are estimated at USh 21/ton-km for Ixuseholds and IEh 14/ton-kn for non juehlds, aoer an averar- distance of 100/50/20 km respectively. Mtrmal distribution marglrs are assuneI to be 1(O. d/ Tie eoanic value of diarcoal is Ibsed on an cppDrtuai ty cast of land equal to lEh 10,000/ha (Table 3.5 with 1 m solid/O.7 tons) assumingr cone rsion in an earth Idln (8.4 toms fuelwd/ton charcml). TrarefoDnation costs are tdken fran Table 4.8. Transport costs are as in (c) above, over an average distance of 100 km. lbrmal distribution murgins are assumed to be 1C%. Source: Mission estinates.  - 93 - which indicates that charcoal can command a price premium due to non-price advantages over fuelwood for these users. In all cases, the market price of energy from petroleum products is much higher than from the indigenous energy sources, and in this respect market prices correctly reflect econom- ic priorities. 4.29 Insufficient data are presently available to carry out a similar analysis for rural energy consumption. However, it can be concluded from observation that price relationships differ from urban areas in that fuel- wood is marketed at lower prices due to lower transport costs and more plentiful supplies relative to demand. To a lesser extent the same obser- vation applies to charcoal. Furthermore, most fuelwood consumed in rural areas is not marketed but is gathered by the consumers themselves. The Mission estimates that about 77% of fuelwood consumed in Uganda in 1980 was not marketed (see Table 1.6). 4.30 The estimated economic costs of available energy sources are also given in Table 4.9, based on import-parity prices for petroleum products (Table 4.6), woodfuel plantation costs (Table 3.5) and the long-run marginal cost of electricity (Table 4.1). All of these estimates are preliminary and they should be interpreted with caution. Nevertheless, a number of tentative conclusions can be drawn. Electricity produced from Ugandan hydropower potential is the least-cost form of energy for lighting in urban areas, and the Government and UEB should encourage the use of electricity for lighting instead of kerosene and L.P. gas. In other uses, the comparisons between electricity and woodfuels are very much dependent upon the economic value of land used for wood production and transport and distribution costs. Table 4.9 assumes that present transport and distribu- tion costs for woodfuels (see Table 4.8) are inflated by the abnormal supply conditions and that economic costs would be substantially lower. However, even so, electricity remains the cheapest form of energy for urban household cooking. This probably approximates the case for the major population centers in southern Uganda, where the surrounding land is suit- able for production of medium-value crops such as bananas and coffee and where supplies from more-distant low-value land involve substantial trans- port costs. For urban areas in the northern part of the country, and most rural areas, woodfuels would become a more attractive option. Similarly, woodfuels are also more attractive in non-household uses, where transport costs are generally lower and the end-use conversion efficiency higher than in household uses. In these instances, the economic costs of electricity and woodfuels would have to be evaluated carefully on a case-by-case basis. However, in all uses, both electricity and woodfuels are substantially cheaper than petroleum products.  - 94 - E. Major Recommendations on Pricing Policies 4.31 The Mission recommends that: (a) a general review of electricity tariffs according to economic principles should be undertaken immediately after preparation of the long-term power development program (paragraph 4.2); (b) electricity tariffs should be raised to a level that corresponds to the economic cost of power as soon as is practicable, and at least by 1991 when the next hydroelectric station will be re- quired. Based on preliminary estimates of the long-run marginal cost of power prepared lby the Mission, this would involve tariff increases of five to ten times in real terms, depending on the exchange rate used. The burden on the poorest households could be alleviated through introduction of a "lifeline" low tariff rate for essential uses (paragraph 4.11); (c) the electricity tariff structure should be adjusted to eliminate the element of regressive rates for additional blocks of consump- tion (paragraph 4.3) and to increase maximum demand charges (paragraph 4.5); (d) the Government should continue to adjust petroleum prices when- ever there is a major chLange in the exchange rate. Under present conditions, it might be preferable to fix the exchange rate used in the petroleum price formula above the prevailing "window one" rate to anticipate future depreciations of the Ugandan shilling and thereby avoid more frequent price increases (paragraph 4.16); (e) the Government should also realign the taxation structure to offset the cross-subsidization of products. This implies rela- tively higher tax rates on fuel oil in particular, but also on kerosene and auto. diesel; tax rates on gasoline could possibly be reduced, especially if the Ugandan shilling depreciates further (paragraph 4.24); (f) the Government should review the formula for reimbursement of operating and capital expenditures with the oil companies (para- graph 4.21); and (g) the stumpage fee should be raised to reflect more fully the economic value of wood as a fuel. An initial adjustment to at least USh 1,000/m3 for forests serving urban areas would seem justified (paragraph 4.27).  - 95 - V. INSTITUTIONS AND MANPOWER 5.1 The present organization of the energy sector in Uganda was re- viewed in Chapter I, Section B. It was noted that the sector had suffered from the general institutional collapse which characterized Uganda during the 1970s. Many signs of this remain: (a) manpower: wages in the civil service (e.g., MPPT, Forestry Department, MPED) have been seriously eroded by inflation, and are now totally inadequate to provide a minimum standard of living, motivate staff and contain corruption. Employment condi- tions in the parastatal and private sectors (e.g., UEB, oil com- panies) are substantially better, and these institutions have had less difficulty in attracting qualified personnel. Nevertheless, there remains an overall shortage of skilled manpower in the energy sector, caused by the disruptions to higher-level educa- tion and training in recent years as well as by the exodus of technicians and professionals overseas; and (b) facilities: the sector's facilities were generally poorly main- tained during the 1970s, and much of what survived was damaged or destroyed during the 1978-79 war. The oil companies have been least affected because of their direct access to foreign exchange to replace and repair equipment. UEB has also been able to undertake minimal rehabilitation with external assistance from ODA and IDA. But, the plight of the Ministries is critical. Many have no vehicles let alone the technical and laboratory equipment needed to perform their functions. Offices are generally poorly maintained, with only rudimentary furniture, few typewriters and copying machines, and no systematic filing systems. Even where these facilities are available, they are often not utilized due to shortages of fuel, spare parts and office supplies (e.g., paper and pens). 5.2 These are economy-wide problems, which cannot be resolved in the context of the energy sector alone. The Government recognizes that any lasting solution to the vicious circle of manpower and budgetary con- straints, weak administration and poor economic performance will require mutually reinforcing improvements in all areas.1/ This is inevitably a long-term process and initial progress will be slow. It would go beyond the scope of this Report to make recommendations on such general issues as wage levels and the allocation of manpower and budgetary expenditures be- tween sectors. However, two observations are relevant. First, the energy 1/ The Government has begun to tackle these problems. A Salaries Commission was established in August 1980 and presented its report to the Government in December 1982. Two inter-ministerial task forces are presently preparing white papers for Cabinet consideration on the Commission's recommendations and other aspects of administrative re- form.  - 96 - sector has a critical role to play in the revLval of the economy and almost certainly deserves greater attention and priority than it has received in the past. Second, even with a drastic reordering of priorities, manpower and budgetary constraints will remain a fact of life in the energy sector for the foreseeable future and proposals for institutional reform must take this into account. A. Sector Coordination and Planning 5.3 In addition to these economy-wide problems, the energy sector has suffered from a particularly weak and poorly coordinated institutional structure. In particular, no one institution (or clear heirarchy of insti- tutions) has been given primary responsibility for energy planning and policy formulation. As a result, day-to-day management has been left to a plethora of separate entities, while higher-level decisions have been taken without due regard for their impact on sector performance and development. Most other countries in the region, recognizing a similar weakness in their institutional structure, have now established a new ministry or expanded an existing ministry to take the lead in energy matters: e.g., the Ministry of Energy (1979) in Kenya, the Ministry of Water and Energy (1980) in Tanzania and the Ministry of Industry and Energy (1981) in Zimbabwe. In similar vein, MPPT in Uganda has proposed the creation of a new Energy Department within the Ministry, which would have separate sections respon- sible for electricity and new and renewable sources of energy, as well as a petroleum inspectorate. The structure and staffing of the Department is summarized in Table 5.1. This; proposal has not yet been discussed by Cabinet, apparently because of opposition from the Ministry of Finance. 5.4 Given the manpower and resource constraints in Uganda, the Mission considers it would be impractical and premature to establish a separate Ministry of Energy at this time. Instead, a more modest and phased program of institutional development would seem appropriate. In this context, MPPT's proposal to establish an Energy Department has some merit. However, in the view of the Mission, the proposal requires elabora- tion and revision in a number of areas: (a) in its present form, the proposal is basically just for a struc- ture, without any indication of what the Energy Department will do or how it will work with the many other institutions involved in the energy sector; (b) the professional manpower requirements -- 21 of the 25 positions would have to be filled with new staff appointments -- are un- realistic at the present time. It would be counterproductive if the Energy Department were given formal responsibilities which it could not then carry out because of manpower or other con- straints. It would also be counterproductive to draw staff from subsector institutions (e.g., UEB), as this could jeopardize what has already been achieved in the management of day-to-day opera- tions;  - 97 - (c) the proposed staffing would seem to be top heavy, with too many managers (e.g., Principal and Senior Assistant Secretaries) and not enough economists and energy specialists; (d) there is inadequate provision for coordinating the three sections in the Department. In particular, no one is assigned responsi- bility for monitoring overall energy trends and providing the analysis necessary to develop an integrated approacha to the sector; and (e) the three sections provide inadequate coverage of two important areas: the woodfuels subsector and energy efficiency. 5.5 Bearing in mind these concerns, the Mission has prepared an al- ternative proposal for strengthening the organization of the energy sector as a whole (see Table 5.2). Broadly speaking, the proposal provides for a three-tiered organizational structure: day-to-day management would be largely left to the subsector institutions as at present, sector planning and policy formulation would be coordinated by an Energy Department in MPPT, and higher-level policy making would be the responsibility of an Inter-Ministerial Policy Committee.2! It is not envisaged that this full structure would be put in place all at once. Instead, the following phased approach is recommended: (a) the first priority would be to form the Inter-Ministerial Policy Committee. / This Committee would be formally responsible for submitting energy policies and programs to Cabinet for approval, as well as commenting on the energy implications of other Cabinet decisions. It would also take the lead in discussing energy concerns with other countries in the region (e.g., on electricity exports and petroleum imports). The Committee would be chaired by the Minister of Power, Posts and Telecommunications, and have the Ministers of Planning and Economic Development, Finance, and Agriculture and Forestry, and the Governor of the Bank of Uganda, 2/ These "layers" of responsibility have been simplified in Table 5.2 for illustrative purposes. In practice, relationships between institutions will be less clear cut. For example, many ideas on development plan- ning will emerge from subsector institutions (e.g., UEB) and the Energy Secretariat will seek feedback from MPED and other Ministries before formally submitting proposals to the Inter-Ministerial Policy Committee for approval. 3/ A similar approach has already been adopted for economic policy (coor- dinated by the President's Economic Advisory Committee) and agricultur- al policies (coordinated by the Agricultural Policy Committee). There is a danger that the creation of too many inter-ministerial committees will unduly concentrate decision-making powers and abrogate normal line responsibilities. However, the Mission still feels this approach is justified in the case of energy, at least on an interim basis, by the virtual absence of a sectoral approach to date and the many Ministries and institutions that have a direct interest in energy matters.  Table 5.1: MPPT'S PROPOSAL FOR AN ENERGY DEPARTMENT a/ Minister Permanent Secretary I . Under Secretary iUnder Secretary F under Secretarv Energy | Communicatin j Administration| Electricity Section: Petroleum Inspectorate: Principal Assistant Secretary 1 Principal Assistant Secretary I New and Renewable Energy Section: Senior Assistant Secretary 1 Senior Assistant Secretary 1 Economist 1 Economist 1 Principal Assistant Secretary 1 Statistician 1 Statisticians 2 Senior Assistant Secretary 1 Executive Engineer 1 Executive Engineer 1 Scientific Officers 2 Engineering Assistants 2 Engineering Assistants 4 Assistant Secretary 1 Assistant Secretary 1 Assistant Secretary 1 5 8 11 a/ Structure below Under Secretary is shown for professional staff in Energy Department only. Source: MPPT.  Table 5.2: MISSION'S PROPOSAL FOR FUTURE ORGANIZATION OF THE ENERGY SECTOR a/ Inter-Ministerial ) Policy Committee ) . ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~) )Policy )Making Ministry of Agri. Bank of MPPT Ministry of MPED and Forestry Uganda Finance ) Energy ) Department b/ ) Makerere National Research ) University Council )Sector )Planning Energy )and Policy Secretariat )Formulation I~ ~~~~ ~~~~~~~~~~________ ._______ I________ t ________ I_______ ) '.0 Woodfuels Electricity Petroleum Energy Efficiency Alternative Section Section Section Section Energy Section Forestry l l UEB Oil Mins. of Industry, GSMD )Subsector Department d/ Companies Transport, HUD c/ )Management a/ Major institutions and responsibilities only. b/ The Energy Department would initially only include the Energy Secretariat and the Energy Efficiency Section; the various other subsector sections would be added later as appropriate manpower and resources became available. c/ The Ministries of Industry, Transport and Housing and Urban Development could be coopted to the Inter-Ministerial Policy Committee for discussion of efficiency issues and programs. d/ The Forestry Department would have primary responsibility for wood production from forest plantations, natural forests and woodlands. However, other institutions would be involved in agro-forestry (e.g., the Agriculture Department), urban and industrial plantations (e.g., private enterprises) and stove efficiency (e.g., the National Research Council). Source: Mission proposal.  - 100 - as full members. Other Ministries could be represented or co- opted as appropriate; (b) the second step would be to establish an Energy Secretariat in MPPT, to support the Inter-Ministerial Policy Committee and for- mulite proposals for Cabinet consideration. The Secretariat would also liaise with MPED and other institutions on energy planning. Its functions would inaclude monitoring of energy trends and developing an integrated approach to energy issues. Initially, it is proposed that the Secretariat be established with five professional staff (three economists and two statisti- cians) working to a Director or Under Secretary. This would require the recruitment of only two additional staff (both econ- omists). As and when additional manpower becomes available, the Secretariat could be expanded to include an energy planner and possibly an engineer. An expatriate energy advisor could also be hired to help set up the Secretarial: and provide initial techni- cal support; (c) the third step would be to establish an Energy Efficiency Section. This would be a small unit of possibly four staff, comprising two engineers and two economists. Their primary function would be to coordinate the activities of the many public and private organizations which are or should be involved in energy-efficiency activities (paragraph 5.11). In particular, the Section should assist the relevant ministries to establish their own energy efficiency programs, especially in the indus- trial and transport sectors. The Energy Efficiency Section should also propose, through the Energy Secretariat, appropriate policies for promoting more efficiert use of energy in the eco- nomy; and (d) the final step would be to improve the coordination of other sub- sector activities. Eventually, this could be done through four further sections in the Energy Department, dealing with wood- fuels, electricity, petroleum, and alternative energy sources. However, manpower and budgetary resources for this are simply not available at the moment. Instead, it is proposed that the Government identify suitable staff and begin training them, with- in the existing institutional structure, to form the nucleus of an expanded Energy Department at some future date. In the mean- time, these staff can mLake a contribution through their normal line responsibilities as well as in liaison with the Energy Secretariat. Much of this coordination can be done in the sub- sector institutions themselves (e.g., UEB and the Forestry Department), which need to strengthen their own planning capacity in any case. Additional support could be provided by the Bank of Uganda (monitoring of the petroleum subsector) and the Ministry of Finance (energy pricing).  - 101 - The Mission recommends that the Government seeks technical assistance to review proposals for an Energy Department in MPPT, prepare a program for implementing institutional changes and develop job descriptions for man- power requirements. B. Subsector Management and Development 5.6 The institutional structure for energy management at the sub- sector level is largely in place and has performed remarkably well given country conditions. The major requirement right now is resources for both operations and development, and the priorities in this area are outlined in Chapter VI. However, there are also some institutional and manpower weak- nesses which should be addressed in the near future. The Mission's recomm- endations in this regard are summarized below. Electricity 5.7 UEB should strengthen its development planning capacity. At the moment, UEB is well staffed with operational personnel and engineers but has no economists to help analyze the viability of future investment pro- posals. This has not been a major constraint in the recent past, when the emphasis has been on keeping the system working and getting by with limited resources. But this will soon change as the country's excess generating capacity is fully utilized and further expansion is required. Priority tasks are to: (a) update demand projections, taking into account export options and the scope for improvements in energy conservation and efficiency; (b) prepare and update a least-cost program for power development; (c) identify areas where extension of the distribution system is required and where rural electrification schemes can be justi- fied; and (d) propose revisions in the tariff structure and level, to encourage power conservation and to provide a contribution towards develop- ment costs. It is appropriate that UEB takes the lead in preparing positions on these issues and manpower should be allocated for this purpose. Correspondingly, the proposed electricity section in the Energy Department can be very small (probably no more than two people when fully staffed), serving primarily as a link between UEB and the Energy Secretariat. 5.8 UEB's accounts are generally up to date and well managed. Col- lection of UEB's current bills is prompt, and there are no unusual arrears  - 102 - either from the Government4/ or from other consumers. UEB is also under- taking a campaign against illegal connections and meter thefts. The major problem now is billing itself, which has often been delayed by computer malfunctions and breakdowns. As similar problems are faced by other parastatal bodies (including the Uganda Posts and Telecommunications Corporation) the Government should consider establishing a central computer center, possibly managed and serviced by the supplier. Petroleum 5.9 Management of petroleum importing and marketing should remain with the oil companies. However, there is an urgent need to strengthen the Government's capacity to monitor developments in the subsector and to nego- tiate effectively with the oil companies.5! It has already been noted that the import arrangements are complex aind specialized training will be required to develop expertise in oil industry operations. The Bank of Uganda has recently established a petroleum desk and this should be strengthened to assume primary responsibiLity for overseeing petroleum importing and marketing, at leiast until a separate petroleum section (or inspectorate) can be established in the Energy Department of MPPT. Prior- ity tasks are to: (a) determine the composition of foreign exchange payments made to oil companies for petroleum products, oil-related products, services and dividend remittances; (b) record systematically information provided by the oil companies on import costs, broken down by crude purchases, processing charges, and transport and oil company margins; (c) keep track of world market trends for crude and petroleum products and of general developments in the oil industry; (d) evaluate alternative supply and transport options, including direct purchases from the Middle East and use of the Tanzania route; (e) analyze petroleum consumption trends, by product and region, and monitor stock levels; 4/ There are old receivables amounting to USh 100 million dating from the time of the military regime; about 50% of these are estimated to be uncollectible and eventually will have to be written off. 5/ The Petroleum Act of 1957 provides broacd powers for the Government to restrict and regulate the import, transport and storage of petroleum. It is generally accepted tlhat MPPT is the responsible Ministry, but this is not specified in the Act. This should be clarified in the revision of petroleum legislation, which is required in any event to provide a sounder basis for the exploration of domestic resources (paragraph 5.12).  - 103 - (f) ensure that adequate foreign exchange is being released to meet priority domestic requirements for petroleum while maintaining a satisfactory level of stocks; and (g) review impact of pricing policies and suggest changes in the pricing formula, including oil company margins and effective tax rates. A start on these tasks can be made by the staff already assigned to the petroleum desk, provided there is adequate training and technical assist- ance. However, they will need material support, especially for reference books on oil industry operations and periodicals on world market develop- ments. Woodfuels 5.10 The Forestry Department should continue to have primary responsi- bility for wood production from forest plantations, natural forests and woodlands. Until a separate woodfuels section can be established in the Energy Department, the Forestry Department should also take a broader interest in woodfuel production issues. Although the Forestry Department is already well staffed, it will need additional vehicles, equipment and supplies to perform these functions effectively (see Table 6.2). Improve- ments in agro-forestry techniques should probably be promoted through one extension system under the Agricultural Department,