Haiti - Household energy strategy

Energy Sector Manag Assistance Programme Haiti Household Energy Strategy Report No. 143/91 ..  JOINT UNDP / WORLD BANK ENERGY SECTOR MANAGEMENT ASSISTANCE PROGRAMME (ESMAP) PURPOSE The Joint UNDP/Wodd Bank Energy Sector Management Assistance Progmme (ESMAP) was launched in 1983 to complement the Energy Aet Pogramm, established three years earlier. ESMAP's original purpose was to implement key recom ions of the Energy Assessment reports and ensure that proposed investments in the energy sector represented the most efficient use of scarce domestic and extermal resources. In 1990, an intemational Commission addresed ESMAP's role for the 1990s and, noting the vital role of adequate and affordable energy in economic growth, concluded that the Programme should intensify its efforts to assist developing countries to manage their enurgy sectors more effectively. The Commission also recommended that ESMAP concentrate on maing long-term efforts in a smaller number of e,ountries. The Commission's report was endorsed at ESMAP's November 1990 Annual Meeting and prompted an extensive reorpaniz on and reorientation of the Progrmme. Today, ESMAP is conducting Energy Assessments, performing preinvestment and prefeasibility work, and providing institutional and policy advice in selected developing countries. Through these efforts, ESMAP aims to assist governments, donors, and potential investors in identifying, funding, and implementing economically and environmentally sound energy strategies. GOVERNANCE AND OPERATIONS ESMAP is governed by a Consultative Group (ESMAP CO), composed of representatives of the UNDP and World Bank, the governments and institutions providing financial support, and representatives of the recipients of ESMAP's assistance. The ESMAP CO is chaired by the World Bank's Vice President, Operations and Sector Policy, and advised by a Technical Advisory Group (TAG) of independent energy experts that reviews the Progamme's strategic agenda, its work program, and other issues. The Manager of ESMAP, who reporti to the World Bank's Vice President, Operations and Sector Policy, administers the Progrmme. The Manager is assisted by a Secretariat, headed by an Executive Sertuy, which supports the ESMAP CO and the TAG and is responsible for relations with the donors and for securing funding for the Programme's activities. The Manager directs ESMAP's two Divisions: The Strategy and Progm Division advises on selection of countries for assistance, carries out Energy Assessments, ppares relevant programs of technical assistance, and supports the Secretatiat on funding issues. The Operations Division is responsible for formulation of subsectoral strategies, preinvestment work, institutional studies, technical assistance, and training within the framework of ESMAP's country assistance programs. FUNDING ESMAP is a cooperative effort supported by the World Bank, UNDP and other United Nations agencies, the European Community, Organization of American States (OAS), Latn American Energy Organization (OLADE), and countries including Austalia, Belgium, Canada, Denmark, Germany, Finland, France, Iceland, Ireland, Italy, Japan, the Netherlands, New Zealand, Norway, Portugal, Sweden, Switzeland, the United Kingdom, and the United States. FURTHER INFORMATION For fih information or copies of completed ESMAP reports, contact:- The Manager or The Executive Secretauy ESMAP ESMAP Consultative Group The World Bank The World Bank 1818 H Street N.W. 1818 H Street, N.W. Washington, D.C. 20433 Washington, D.C. 20433 U.S.A. U.S.A.  HAITI HOUSEHOLD ENERGY STRATEGY DECEMBER 1991  FIXCLANGE RATE US$ 1 ourdes S.0 a Hatan $1 USS 1 = FF 5.8 CONVERSION FAC?'ORS LPG 0.55 45.7 Kerosene 0.8 43.5 Diesel 0.84 433 Wood (air dry, 20% humidity) 0.7 17.0 Chara (10% humidt 29.0 Bagse (50% humidity) 7.5 Fia Enera 86 Primy EnerV 283 Otm 1 US Gallon - 3.785 ter I barrl 159 liters 1 TOE -41.87 0J o 10 mMioncal CUarwa poduction eiiency - 20% (in weight) 1 stere of wood -1 il (staked) * 0.6 m3 (round wood) 1 cureau = 129 ha  ACRONYMS AND ABBREVIUTIONS AFVP Association Frangaie des Volontaires du Progr&s AOP Agroforestry Outeach Project RAPP Bureau d'Approvisionnement en Produits P6troliers BDPA Bureau de D6veloppement de la Production Agricole B1T Bureau International dui Travail BME Bureau des Mines et de l'Energie BRH Banque de la R4publique d'Halti CPNAP Commbsariat & a Promotion Nationale et t l'Adminition Publique DR1E Diecton des Ressources Energdtiques (BME) EdH Electricit6 da ESMAP Ekergy Sector Management Assitance Progmme FAC Fonds d'Aide et de Coop6ration FAO Food and Agricl Orgnization FEPP Forety and Enironmental Protection Project OoH Govenmuent of Haiti IDA interational Development Association LPG Liquifled Petroleum Gas MARNDR Mlniste de lAgdcultur; des Ressources Naturelles et du D6veloppement Rural ODH Operation Double Harvest NOO Non Governmental Organization OLADE Latin Ameican Energy Organization TFAP Tropical Forestry Action Plan PFN Proet Forestier National UNDP United Nations Development Program !EI Renewable Energy Development nstitute SRF Service des Ressources Forestieres TOE Ton Oil Equivalent USAID United States Agency for International Development  . . . . . . . . . . . . . 2b . . . . .: .:. . .:.:. . . . . ... . .. . ... . . . . .. .. . .. ... .. .. .. .. . .. . . . .- . ........ .. ... . . . . . . . . . . . . . . . . . . . . . . . . * . . . . . . . . . - . . : :~~~~~~~~~~~~~~~0 . . . . . . . . . . . . . . . . . . . . . . . . . . . .. co~ * 0. * .... - a iFz g~~~I ~~~.~ .......- .--.- - . . ...... ..*a.*-................... ............R.  II. T E STRATEGY ................................................... 33 A . Interventions on Woodfuel Supply ............ ................................ 33 Current Economic Situation of Charcoal Exploitation ................. 33 Modlernizng the ChbarcoaSector .. * *.....*....................................... 37 Geographic: Reorientationof Charcoal Offtake ...................... 38 todernization of Supply Netvorks .......... ............................... 41 Improving theCharcoal Taxation System .............. . ........... 42 B . Demand-Side Interventions ............................................... 44 Economlc Conditions for Charcoal Substitution ........................... 44 Comparative Fuel Prices ..................4**.4..* *.. ......... 44 Real and Theoretic Economic Costs of Cooking Fuels ................ 48 Impact on the Balance of Payments . ............................ . 49 7no Market for Fuels and Stoves .............................................. S0 Cufrent Users of Improved Stoves, Gas Stoves and Kerosene Stoves ...... 50 Consumer Accepta e of Different Stoves and Fuels ...... ............ 51 Potential Markets for Stoves and Fuels ........................... 52 Gas Promotion .......... . * .......................... 54 Energy Conservation ............................................... 57 Substitution and Conservation: Discarded Options .............. .......... 58 Leam-Pronising Short Term Options ............. ................... . 58 Secondy Options ... ........................................ 59 TV. ST TEGY COMPONBNI . ......................................... 62 ening the Gas Market ............ ................................ 65 Charcoal Conservation ....................... ...................... 67 ModoIern fDizngie liarcoalSectr .................. .................... 70 Supply bLu{aser Plan for Port-au.Prince .............................. 71 Assstmazce to Chasrcoal odue .......... * * * * * ........................ 72 onitoring tie Strategy .................. .......... 74 Eomoniic Aspects ........................ .................. 76 Strate Costs 4444444444 44444444444......... 44... 4 44.......4........... 76 Financa and EAononic nalysis . ............................... 77 Rik ..................................................... 79 Financing .................................................. 80 TABLES 1.1 Final Energy Balance, 1985 . . ...... . . . ........... . . . . . . . 3 12 Woodfuel Conasunption in Haii 1990 ....................... ....... . 4 1.3 Cooking Fuel Use in Port-Au-Prince according to Living Standard Quintile (in percentage) ................................ 6 1.4 Cooking Equipment in Port-au-Prince in percentage ofotal Househlds..............444.4.44..... 7 1.5 Bloman Standing Stoc in 1988 .. ........ ... 10 1.6 Charoad rea post counted in Port-au-Prine ........................... 11 1.7 Production areas of charcoal since 1979 in percentage .................. .... 13 13 Charcoal price swr ture ......................................... 14 19 Clsrwltaixrevenues .............. ...................... ..... 17 1.10 BIP price ....... .................................. ... 22  3.1 Compared Costs of Cooking Port-au-Prnce, 1990 ....................... 46 32 Cos.s of Importing Gas and Kerosene ..... ................................. 47 3.3 Comparison of Propane Price Suctures in Haiti and Dominican Repubic ............ ................44.... .. 48 3.4 Compared EconomcCostsinPort-au-Prince, 1990 ........................ 49 3.5 Bip Tichri'sshareofthegasmarket in thousands of households ............... 50 3.6 Reints ofthemarketstudylnPort.au-Prince ................ . 40 ...... 53 4.1 Summarized Budget of theStratey,s Compov'.nts ......................... 77 4.2 Fiancial Analysis ... ..... *4444*44.. . *. .. 4*. . .4*....4..*. . 78 4.3 Economic Analysis ..........4.. o4.. 4 ...4 79 FIGURES: 1.1 Charcoal Supply Systems in Haiti according to Stevenson (1985) 12 3.1 Owcoal price * 1953-1990 Trend ...... . ...... ... . ...... . . 34 3.2 ProducersShareofCharcoal Price -1979-199OTrend ...................... 35 33 Main ChaacteristicsofCLarcoalProduction .................... 40 3.4 Costs of Cookingwith Charcoal Keaosene and Gas . ..... ..... 45 4.1 Epeded Impact of the StrateVU . . ... ... .... .... ......... . . ... 64 ANNEXES: L Foresty Resourcs and Deforestation .. ........... ..82 IL Energy Prices in the Cariean and Central A aerica. .................. 90 W.. Cookstoves i use n Haiti .................................. 94 Iv. Volumes of Charcoal Entries in Port-au-Prince . ... ...4.. ...4...4.. 99 V. Price Strcures of Charcoal, Gas and Keroene ....... .............. 100 VL Fiancial and Economic Coss of Cooking ............................ t 104 VIL Factor Analysis of the Household Survey .................. .. ...... 111 VIII. Market Study for Household Fuel ... . . ............................ 116 IX Projected Consumptions (Trend and Strategy) . ...... 124 X General Data .....4.. ...4................ ..... 127 XI. Short-Tern Actions.......... ........o .................... 144 XIL EconomicCostofCharcoal ... ...... 149 XIIL Reports and Studies Canied out .......... 4.............. , .......... 156 TV. Budget of the trt Coponents ..................... .............. 158 MAPS IBRD No. 23061 - Haiti Progession of Charcoal Poduction Areas IBRD No. 23062 - Haii, Charowal Production Areas IBRD No. 22676 - Hait5 Forestry and Environmentad Protection Project  FOREWORD This study was completed as an activity of the Joint World Bank/UNDP/Bilateral Aid Energy Sector Managem.ent Assistance Program (ESMAP). Financing for the study waa provided main by the N.3rweglan Government, as well Ps by UNDP, the French Government and the Jai.-nese Fund. The study was implemented with the assistance of and in close coordination with lie BME/UNDP/OLADE Energy Sector Strengthening ?roject (HAI/89/031) and with the teams involved in the preparation of the IDA financed Forest and Environmental Protection Project ESMAP and its consultants worked with their main counterpatv, tde Bureau of Mines and Energy (BME) as well as wil the Forest Resources Sevice (ERS), the Bureau for Petroleum Product Supply (BAPP) and the main Haitian organiztions operating in the sector. The study coordinator in the BME was Loctamard Antilus, Director of Energy Resources. Nicole Dieudonne, Guilen Aurelien and Naquin Medina also participated in the study. violette Derosiers was responsible for svey data entry (carried out at the BME), with assistance from Marie Mercie Dieujuste and Eunide Germain. The main authors of the following report are Philippe Durand (Task Manager, ESMAP) and Michel Matly (Energy Economist, coordinator of the consultant team). The study team indluded Jocelyne Durany-Jakob and Carole Roy (Consultants, Sociologists), Jos6phine Arpaillange (Energy Statistician, ESMAP), Chrsa Bonaparte (Consultant, Marketing Specialist), CAlite Cldrism6 (Consultant, Sociologist) and Kevin Fitzgerald (Energy Statistidan, ESMAP), as well as about thirty presenters and pollsters. Thi study is based on a major data collection effort comprising both quantitative and qualitative information collected from December 1989 to May 1990. During this time, the folloig main suveys and in-field inquiries were carried out: household surrey, market pols and *uveys of small businesses in Port-au-Prince; surveys of charcoal retailers and transporters in Port- au-Prince and in the main cities; sureys of rural merchants and charcoal producers in the main areas which supply charcoal to Port-au-Prince; counting the number of charcoal entries at Port- au-Prince (see Annex XM).  EXECUTIESUMMARY 1. Haiti's environment is undergoing a gradual p-ocess of severe degradation, which has immediately perceptible and sometimes irreversible effects, secialy with regard to loss of soils and decreasng agricultural yields. Even though this enviromental problem has arisen partiall due to multiple non-energy factors (demographic prassures, agricultwal techniques, irregular land tenure practices), the fact remains that woodfuel consumption, especialy urban charcoal consumption, is a major contributor to deforestation. Tbis is not a newly noted phenomenon: experts have cited it for more than a decade. However during that time, no coherent, comprehensive program has been proposed or undertaken by the Haitian Government with assistance from international development agencies. 2. It is signficant that in Haiti - frequently cited as one of the countries in the world where environmental desuction has reached an alarming phase - there is no recent inventory of forest or biomass reources, nor any serious analysis quantifijng the extent of deforestation. Without fundamental data needed to accurately evaluate the current state of the biomass stock and its annual growth, alarmist accounts of the approaching demise of the last Haitian tree persisL Independently of the uncertainties surrounding wood-energy supply potential, deforestation in Haiti is an inconttabe reality; it is therefore important to reduce the pressure on forest resources generated by woodfuel consumption, especiall charcoal consumed in Port-au-Prince. Reforestation d ManagM Fofreist Rmmu 3. The only national-scale projects related to this problem have been carried out by the private sector. They comprise mainly the agro-forestry tree planting projects run by non- governmental organizatons operating throughout the country. Although these plantatins are expnsive and their overall suwess has been relative, several million new trees grow in Haiti each year as a rest of these efforts. An additonal advantageous outcome is that tree planting has boeome ingrained into traditional peuant practices However, tree, planting efforts represent only partial solutions to Hait*s deforestation problem. Based on the latest analyses of deforestation, at least 23 million surviving trees would need to be planted each year, i.e. nearly 10 times the current rate, in order to reverse the destucion of national forest resources. 4. Public mn ent of forest resources, already relatively ineffective, has continued to Wons In 1986, a new stage was reached: monitoring of forest exploitation disappeared almost totalDy. Since then, charcoal exploitation has spread throughout the country, starting from areas where lage-scle production was already the nonn, so that now almost all regions of Haiti market charcoal for the capitaL In some parts of the country, parularly along the aiorthwest peninsula,  v i. charcoal production plays a major role in the rural eco:aomy and has become an important source of income for numerous peasants. 5. The large scale plantations approach has been abandoned because of the lack of financial piofitabflity which in turn prompted landholders away from this option. Also, investments for reforestation and forestry research under the National Forestry Project have produced only minor results relative to the deforestation problem. The project achieved somewhat limited progress with increasing the effectiveness of forestry control; but the Forestry Service still monitors only 10% of the charcoal trade. Management of Household En,r Demand 6. Several projects have addressed the problem of charcoal conservation (BME, NGOs). Results were negligible due to the smal size of the projects, lack of resources and dispersed efforts. Although studies show that the charcoal cookstoves currently in use are highly inefficient, there have been no commercial-scale promotion of improved models. 7. The initiative for developing charcoal substitutes has been taken by the pri:ate sector, which introduced portable butane gas stoves (Bip Ticheri) and engaged in large-scale publicity campaigns to promote this fuel. The Government assisted private sector efforts by decontrolling gas prices and app.ying favorable foreign exchange conditions to fuel imports in 1987, wih extension of the same conditions to the import of Bip cookstoves in 1989. Costs of Household Fuels 8. Extensive use of charcoal in Haiti's cities is, to a very large extent, an indirect consequence of the high gas prces which result from current supply and distnbution arrangements. In fact, the Haitian gas market is a de facto monopoly, characterized by small and often outdated instaflations and gas prices among the highest in the Caribbean, if not the world. It is a captive market for the petroleum company which holds a monopoly over the entire supply of gas - from gas purhases on the international market to retail distribution. 9. Under the current situation - no other supply alternatives, decontrolled gas prices, no taxes or tariffs - the offiial gas policy is prejudicial against the State itseW, depriving it of much needed resources. This policy costs the economy about $25 million annually, has no direct benefits to the consumer and only marginally provides incentives for the use of g's as a charcoal substitute. Under a more norral price structure, the gas could be lightly taxed, and the supplier margins would be reduced so that the final price is much less costly for the consumer. 10. Analyses show that under normal conditions of supply, gas becomes the best financial option for the consumer and the best economic option for the country. Market studies underscore that gas is also the fuel which is most likely to sucessuly substitute for charcoal in urban Haitian  lv- households: the market potenW'al is for conversion of 25% io 50% of households In the capital depending on which pricing assumptions are retained, as compared to a potential market share of le than 10% for its direct competitors, kerosene and electricity. The upper estimate, 50%, is substantially bhiger than the current rate of market penetration: only 15% of Port-au-Prince households use as as the only or primary fuel However, this is far below the estimated 70% uthiation rate by households in the capital of the Dominican Republic, though the fuel is heavfly subidized (and the price of charcoal is two times higher than in Haiti). 11. The household energ sector is in complete economic disequilibrium and thus rapid intervention is needed. Peasant charcoal producers subsidize urban consumers since wood is acquired at a price far below its replaceme"nt cost - prices which reflect a "mining" mentality in the exploitation of forestry resources - thereby contributing to degradation of the environment and the rural economy. International organizations subsidize reforestation activities and even part of the operadng costs for the forestry services, while elsewhere ineffective collection of charcoal-related ues represents a recoverable loss of nearly $1 million annualy. The public authorities subsidize the gas sector to a considerable extent through its favorable import policies; unfortunately, no benefits accrue to either the consumer or the State, even though the State itself is sorely in need of more sources of operating revenues. 12. The primary objective of any intervention in these areas is to gradually restore the financial and economic equlibriums in the sector through: lowering significantly the price of gas; allwing charcoal prices to increase substantialy; introducing mechanisms for the State to generate rvenues from the marketing of both fuels, sufficient for financing operations and partal Investments by the public entities responsible for managing the sector. 13. Intervention also will be centered only on several main themes which represent important stakes for the countty and which are likely to have significant impacts. These themes wil be addressed in a manner so that the scale of intervention will have national impacts. 14. Intervention requires the State to affirm its political willingness to act at the highest levels, and to make a more accurate accounting of its strengths and weaknesses, by seeking intervenon of eising non-governmental and private entities, and by negotiating when necessary with the local populations. 15. Finally, intervention in the household energy sector must be designed as a contribution to solve the problem of environment degradation. A household energy strategy would therefore not hinder but rather complement and reinforce the countrys global strategy for natural resource management  -V. 16. The proposed strategy contains measura designed.to effectively redress quickly the current dramatic situation in the sector by intervening on four levels - the four components of the rdcommended strategy: a. opening the gas market, b. generating charcoal savings, C. modernizing the charcoal sector, d. sector monitoring and evaluation. 17. Te frst m of th_. opening the gas market, wi create conditions for real competition in the gas sector, thus producing downward pressure on the gas prices to the benefit of consumers It also will facilitate access to gas as a primary fuel and to gas appliances by the less prieged classes (middle classes and upper levels of the poorer households). This will be accomplished by either installing new receivin& storage and filling in or by elarging exsting inftucture, with possibly temporary recourse to gas imports supplied from the Dominican Republic. The various pibilities of partnerships between private or institutional operators, both local and foreign, should be studied, as well as the legaL regulatory and fiscal aspects of the prerequisites and consequences of partnership choices. 18. The second component of the strategy is the lgW-scale distribution of improved wcoal stves in Port-au-Prince and in the four major secondary cities in Haiti In fact, even under the most favorable assumptions of subituto by other fuels, the proportion of households which will continue to use charcoal will remain ignificant: about one in every two, on average in the major cites. he distibuton wil target household users of round stoves and "potaje" cookstoves, as well as snall roadside food vendors ("manje kwit'). A total distrbution of 80,000 stoves i8 anticipated at the end of three years 19. The third component targets the reorgpnization and.ratonalzaton of charcoal exploitation. Charcoal exploitation will be redistrbuted geohicall through the use of regional production quotas, which will ensure that charcoal production is carried out in areas where the risks to the enionment are acceptable. The component wi help the Haitian Government to more effectiey control the flows of charcoal, especially for marketing in the capital, and to collect taxes associated with charcoal transport. The component is based on definiton and implementation of a Supply Master Plan for Port-au-Prince, strengthening and improvement of controb on charcoal transport and of the tax colection system, contribution to defining a priority rural development progrm to compensate local popuations for losses generated by application of charcoal quotas, and  -vi- implementaton of an assistance program for charcoal producers in zones where charcoal exploitation wil continue. 20. 31M fourth coment indudes creation of a Steering Unit for the strategy. The Stering Unit wl asre permanent supersion and coordination of the household energy sector and wil assure cooperation between the various sector participants, as well as coordination and evaluation of activies carried out as part of the strategy. In order to promote coherence in the sector, the Steering Unit wil also oversee adjustments to pricing and fiscal policy and legislation cmoening the different fuels. At the fiscal level, the strategy includes improved collection rates for charcoal preos and a now tax on the price of gs once the gas price structure has been revised so that gas can be distributed throughout Haiti at lower prices (para. 16). 21. The main outcomes of the Strategy wfll be to stabilize charcoal consumption in Haiti and to gradually transform as much charcoal production as possible into a sustainable, financially profitable and envionmentaly benip economic activity. Specific outcomes for the strategy once the four components have been implemented are the following: a. gradually incrase in the proportion of urban consumers using gas as the primary fuel, with a target percentage of 35% of Port-au-Prince households in 1996 and 50% in 2001 (10% and 20% respectively in other cities); b. developmeat of gs consumption from 7,000 tons at present to 23,000 tons in 1996 and 41,000 tons in 2001; C.X ue of improved stoves by one of every two charcoal consumers in Port-au-Price by the year 2001; d gradual drop in charcoal consumption (and thus decreased pressure on national forest resources) by 43% of projected 2001 consumption, through substitution by gas and distribution of improved stoves. i. mprovement of conditions for charcoal production and rationalizaton of the acce to forest resources in the main charcoal eploitation diticts (Supply Master Plan for Port-au-Prince), including improved tax polcy and better enforcement of regulations concerning charcoal commerce (tax collection rates for charcoal tranwport to g from 10% in 1990 to 60% in 1996); f vocationa taining and assistance for developing professionalism among 500 small- scale charod makers  vii - 22. Tho impacts of this strategy on the national economy are the folo.wing a. relative increase in petroleum product imports of 12% by 2001, all other things being equaL This is equivalent, in terms of the current level of ternal trade, to 4% of the countries total exportsL b. income genermted by charoal exploitation in the main production zones to remain constant (or, under normal conditions, to increase due to efforts to increase the real producer price). Cot and FEnancing 23. Implementation costs for the strategy total USS 10.5 mfion over the 1992.1994 period: USS 0.8 million for the stove distribution progam, (which is included in IDA!s Forestty and Envionmental Protection Project) US$1.4 million for modernizing the charcoal sector, USS 7.5 million for petroleum investments (ir 't private), and US$ 0.8 million for monitoring and evaluation. Total recurrent costs are estimated at US$900,000 over the period 1995.2001. The simplffied economic and financial analysis gives a financial IRR of 10% and an economic IRR of 40% for the period 1992.2001. 24. Considering the urgency of implementing the Strategs components, obtaining the necesay funding should be among the Government's priorities. Several bilateral donors traditionally active in the energ and foresty sectors could be approached, such as France, Canada, uSAYD and Germany. Other resources might also be tapped, such as the Haiti Economic and Socal Fund in the recently-reated Global Environment Facility.  L OVERVIEW Genra Socio-Econemic Osve Dtioationu of the Eggffi= 1.1 By far the poorest country on the Latin American continent, Haiti has been in the midst of a sustained economic and social cisis for over a decade. Tis crsis is manifested by a continuous decline in per capita GDP averaging 1 to 2% per annun since 1985 and estimated at nearly 4% in 1989. The drop in GDP has had repercussions in all economic sectors. The decreased demand for imports, associated with the economic slowdown and the resulting reduction in the population's purchasing power, has been totally offset by declining exporis, decreasing of foreig aid and the loss of tourism revenues. Thus the trade deficit, which was about 40% in 1989, remains substantiaL In the formal commercial sector, the level of salaried employment has continued to drop: accrding to the National Bank of Halti (Banque de la Rpublique d'ati - BRH), the number of employees in the formal sector declined by 10% between 1986 and 1989. The effects of the crWis have not spared the informal sector: during the last three years, about 20% of the 3300 mini and small business located in the capital have been forced to shut down. Declininf of Purchasin Power in Urban Areas 12 Inflation was relatively moderate until recently but now is on the rise; officially estmated at 7%, actual inflation figures were in the range of 15 to 20% in 1989. The prices of most basic products have increased substantialyr. food products such as oil, meat, eggs and sugar rose by 30 to 80% between February 1989 and February 1990. Construction materials Otso increased by 25% during the same period (30% for timber alone). By contrast, the minimum salay in HaIt4 (H$ 100/month) has not changed since 1984. A progressive weakenin of HaIti's crrency is linked with this inflation. At the beginning of 1990, there was a 40% differential between the nominal rate of exhange (5 Gourdes per $1) and the actual rate offered on the paralel market. There results a continued erosion of purchasing power in urban households, particuarly among the middle dasses, whether it be for purchases of imported supplies and products (automobile purchases, for example, have dropped from nearly 6000 in 1985 to 2500 in 1989) or more simply for purhases of local products Acity Slowdow in Rural Areas 13 Agriculural actiy has been on the dedline for several years, as evidenced by a decrease in its value added ir 1989; in conatant terms, value added already had fallen by 10% from 1986 to 1989. Agricutural activity decline is parwtiuly noticeable at the level of crop exports for products sch as cocoa (production was halved between 1985 and 1988) or sugar cane (production went from 2 milion tons in 1985 to 350,000 tons in 1989). Inflation coupled with a decline in international market prices act as disincentives to ural producers: the tendency now is  2- to move away from eport-oriented cash crops and raise food crops instead. The eradication of pigs also has deprived the peasants of one of their rare sources of monetaty income. Marketed production and monetary income in rural areas thus have been reduced considerably, producing infi4tionaty rcions on the prices of products flowing onto the national market. 1.4 With an estimated population of 5.8 million in 1990, Hait has a population density of 215 inhabitants per bnm, one of the highest in the hemisphere. This creates a growing pressure on the environment due to demand for new agricultural land. Even so, the annual growth rate for the rural population is estimated at only 0.7% for the period 1982-2000, while the growth rate for the urban popuation would reach nearly 5% during the same period; the capital, Port-au-Prince, represents nearly 20% of the total population. This urban population, most of whom consume charcoa, thus constitutes a rapidly growing threat to the environment 3b E3n=ag Sector E=Wa BDemand 1.5 Per capita conumption of commercial energy in Hait is among the lowest in the world and certainly the lowest on the Latin American continent. Estimated at 1.8 barrels of oil equivalent (b.o.a), annual per capita consumption is only half that of the Dominican Republic, one third that of Jamaica or Barbados, and one quarter that of Cuba. These comparisons are even more strikng fqr elecicity consmption, which measures about 51 kWh per capita annually in HaKit versus 430 kWh in the Dominican Republic, 510 kWh in Jamaica, and 1410 kWh in Cuba. Only a fraction of Haiti's population has access to =ommercial energy, kerosene, used for lighting, is the only fue in widespread usage. Gas 1/ and electricity are used by a very small segment of the population, almost acc*el in urban areas (the national electrification rate is less than 15%). 1.6 The most recent energy conmption data (for the year 1985) are summarized in Table 1.1. Residential energyr cWns on represents more than half the total energy demand, followed by the commercial sector (21% of national demand), industry and transport. The table emphasizes the preponderance of the woodfuels sector in national energy supply fuelwood, which by itself represents 70% of the national supply, charcoal and sugar cane bagasse. Haiti currently is more than 80% slf-sufficient in energy due to residential use of wood in rural areas and of charcoal in urban area StiDl, it also should be noted that fuelvood, bagasse and, to a lesser extent, charcoal are used largely by the industrl sector; thus these fuels cumulatively represent 63% of energ consumtion by industry. 1/ References to gas in this rMpr impl Liquid Petroleum Gas (LPG).  .3. Igkl1.ls Final tnwry _orution, 198 (1000 TOE) Residential Indtry Transport Corceial Pubtlc Toxal Cost 36.S 36.5 Petroltepro icts 17.4 49.0 121.5 187.9 LPG 2.7 1.2 3.9 Electricity 10.1 11.1 1.2 3.1 2S.5 Charcoal 88.9 1.5 90.4 Fualwood 635.1 49.8 286.S 971.4 Sao"" 86.0 86.0 TOTAL 754.2 233.9 121.5 287.7 4.3 1401.6 1.7 Several srveys and evaluations by the BME and within the famework of this study show the breakdown of charcoal and fuciwood consumption by category of consumer. This breakdown is presented im Table 12, which shows both the levels of woodfuel consunppdon and their equivalent in terms of wood cuttin& One wi note: (a) the importnce of charcoal which is ony consumed by less than 30% of the populadon, but represents 40% of the demand for wood, (b) the importance of non-residential charcoal onumption, estinated to represent 20% of the total dharcoal consumption, (c) the importance of Port-au-Prince for cbarcoal consumption; with two-thirds of the naon con  -4. Tabte .2': Voodfuel Consuption in Haiti, 1990 Charcoal Woodfuel Vood equivalent (1000t) (1000t) (1000t) (a) (1000.3) (b) Port-au-Prince Households 160 800 1,143 21 Informal sectors, others 40 200 286 5 Bakeries/dry cleaninw 20 20 29 1 sub total 200 20 1020 1458 27 Other urban HouseholdsZc) 65 30 355 S07 9 Informal sectors, others IS 75 107 2 Bakeries/dry cleaning 20 20 29 1 sub total 80 50 450 643 12 Rural Households- 2100 2100 3000 56 aSuldivas 130 130 186 4 Oil factories 30 30 43 1 sb total 2260 2260 3229 61 Total HaNit Households 225 2130 3255 4650 87 others 55 200 475 680 13 Total 280 2330 3730 5330 100 jggess (a) Charcoal production efficiency estimated at 20X (in weight). (b) Wood dnsity estimated at 0.7 tons/i3. (c) 905 of households use charcost. firews &Surveys of ESIAP/D1E/OLADE/PID, 1990 (Annex IX). Revised BNE data (1987). n2 XLgi of PetroleuImn 1.8 Despite the dominance of woodfuel in the national energy balance, petroleum imports represent a considerable weight on the Haitian economy. Petroleum imports amounted to nearly 50% of totale portsin 1982, when oil prices peaked. After falling sharply to just 20% in 1987, the ratio between oil imports and total exports has begun to increase once again, due partly to the continued erosion of export of goods and services and partly to the sustained demand for energy (Annex X). 1.9 Thus in 1989, the energy bill absorbed more than 30% of the countys export revenuea Kerosene and LPG represent 10% in volume and about 20% in value of all petroleum product imports to the countiy. Although ps consumption is very limited (about 10,000 tons in 1989), the value of gs imports alone is equWalent to 4% of the total value of HaIti' eports  Cats of Commial Enery in Hati 1.10 One of the reasons for low consumption of commercial ener is without doubt that Halti is one of the Central American and Canibbean countries where petroleum products and electricity are most expensive. A comparison of prices in Haiti and the Dominican Republic demonstrates this: even taking into account the overvaluation of Haitian currency ( a 40% margin estimated between the official market rate and the parallel rate relative to the dollar), the price of kerosene and diesel are about two times higher in Haiti, that of electricity three times, and LPG prices nearly five times higher than in the Dominican Republic (see Annex II), where LPG is heavily subsidized. 1.11 Conditions for marketing and distrbution are such that major differentials also exist between the economic cost of the various fuels in Haiti and in other Central American and Carbbean countries. TI is particularly true for LPG, a fuel which is not under government price controt Household and Small Indut Consumtin in Urban Area =naeidetal Conumotion 1.12 Tle categorization of Port-au-Prince households by fuel (primary and secondary) is summarized in Table 13. Charcoal is the most utilized fuel, consumed by 90% of the capital's households. The most affluent clases are abandoning the use of charcoal (only about 2/5 of the 20% most affluent households use charcoal as their primary fuel). Those who consume charcoal only (62% of the population) use about 0.44 kg per capita daily. Taking into account secondary uses of charcoal, daily per capita consumption in the capital for all households combined is estiated between 0.36 kg (based on weighin auried out in 22 households) and 0.42 kg (based on a survey of 400 households). This results in a total residential consumption for Port-au-Prince approaching 160,000 tons in 1990 (based on the results of the household survey).  .6. ilb 1.3: Cooking Fuet Use in Port-Prince acording to Living Standrd Quintile (in percentap) Nedium Nedita LoW Low Nediua Nigh NHlh Total CHARCOAL Only or main fuel 89.7 96.6 97.8 83.8 41.1 80.8 Sacmndmry fuel 6.4 2.4 2.2 11.3 32.8 10.9 Total 96.1 99.0 100.0 95.1 73.9 91.7 GAS only or wain fuel 1.3 1.3 2.5 13.9 55.8 15.1 Secondary fuel 1.3 3.? 3.7 31.0 22.0 12.0 Total 2.6 5.0 6.2 44.9 77.8 27.1 FUELU10 only or mafn fuel 7.7 1.2 0.0 0.0 0.0 9.3 Secondary fuel 6.4 4.9 1.3 0.0 0.0 2.5 Total 14.1 6.1 1.3 0.0 0.0 11.8 ELECTRtICITY Only or main fuel 1.3 0.0 0.0 0.0 2.5 3.5 secondary fuel 2.6 0.0 3.8 11.3 6.3 4.5 Total 3.9 0.0 3.8 11.3 8.8 8.0 Only or main fuel 0.0 1.2 0.0 2.5 1.3 1.3 Secoday fuel 0.0 0.0 3.8 3.8 1.3 1.8 Total 0.0 1.2 3.8 6.3 2.6 3.1 sLE : Survey ESIP/W/LADE/P, 1990 llos S" Amx X for tho definfitn of the living standard vrale and aerae incoe and food epdture ftor oah quintit*. 1.13 Among the other fuel gas is of spe l note, as nearly 27% of households currently are equp with Vgs stoves or gs cookers and more than half of these (15% of total) use gas as the primaty fuel (see Table 1.4). In fack use of gs is well _sablished mainly in high-income households (half use it as the primary fuel) and middle-income households (those with incomes of more than $250/month). Use of gas among households with more modest incomes s1 is relatively limited: however, the arrival of "popular gs with the disemination of Bip single-flame butane has generate an iital penetrin of this market. More than 50% of portable gas cookstoves are sold to households with incomes of ler than S2W0/month (the last three quintiles of the population). 1.14 Other fuel are used on* by small seents of the popuation: wood used in poor households (2.1% in open fires and 4.1% in portble cookaoves); kerosene, which one finds in use among the fringes of middle- to high-income populi (0.8% primay usage and 53% in stoves/cookers); and electicity (0.8% primary usage and 53% in stoves/cookers) used uniquey in affluent households.  -7.. 1.15 Household energ consumption in the seoondary cities of Haiti (the main ones being Cap Haitien, Les Cayes, Gonaives and St Marc) apparently Is compable so that in the capttal, with charcoal predominating The proportion of householb using pas and kerosene for cooking is much lower, however, and residential consumption of fuehwood is stifl widespread, spially in cain port neighborhoods situated along city outsits laife 1.4: Cooktng Equipwnt In Port-Princ* In Prcentage of total Nouhholts Nan stove Secondery stove Other stove Wood Three ston 1.8 2.5 Charcoal Round stove 55.3 10.3 PotaJS6 stove (36tal) 13.6 3.S "Potaj6' stove (blocks) 7.0 3.8 Squre stow 4.0 2.3 3qwoved stove 1.5 0 0 total S1. 19.9 2.3 Gan alp Tich4rl 3.5 S.8 1.5 Plate 1.S 0.3 0 Cooker 9.8 7.0 2.5 Toatl 14A 1.1 4.0 Keroesne 1.0 1.5 1.3 Electricity 1.0 4.5 1.3 TOTAL 100.0 41.5 9.0 Scumse: urveyESNAP/W/OLADE/PNW 1990. * - Less than 1X. 1.16 There are various metallic traditional cookstoves available (see Table 1.4); the dominant types are the following: (a) the round stove (or its square version which swls for prices vatying from $1.5 to $2, is used by twohirds of the households, (b) the wpotaje stove, consisting of two or three burers supported by a metallic or cement block fiame, is used by near 30% of the househol, geay the most affluent. The metlic model of this Nt of the lne' stove coss fom $15 to $20, and the %loW model sells for about $3 to $S per burner.  -8- 1.17 The lifespan of these stoves is approimately one year. The round sngle-burner stoves (as well as their square version) are produced by 'r6chauliers (craftsmen who specalie in making portable cookstoves), and the potaj6 stoves by iron or metal masons. An estimated 200 r&chauliers can be found in the capital CosuMlin Sma lAdut 1.18 With regard to wood and charooal consumption by the urban productive sector, a distinction must be made between two main categories: (a) productive activities in the informal sector, which often consume charcoal; (b) bakeries and dry-cleaners, which use mostly firewood. 1.19 Among the activities of the informal sector, small roadside restaurants and hot food vendors represent the bulk of charcoal consumption: a study carried out in 1989 by the Center for the Promotion of Female Workers estimates that there are nearly 80,000 sale points of food products in Port-au.Prince, of which 20,000 consist of small Omanj6 kwit restaurants (one for eight households). Even if these figures are overestimated, they show the informal sector's importance in the capital. The household survey estimates that charcoal consumption linked to home-based productive activities of this sort accounts for 20% of total residential consumption: thus in Port- au-Prince, about 40,000 tons of charcoal are consumed by these activities annualy, bringing the total consumption of charcoal in the capital to 200,000 tons/year. 1.20 The BME's 1987 census identifies 836 bakeries (146 in Port-au-Prince) and 113 dry- leamers (94 in Port-au-Prince). Some of these companies have switched from kerosene to gas (approximately 30 bakeries and 13 d&yv-eaners): hence 95% of the bakeries and 90% of the dry- leaners in Hati consume firewood elusively. Total consumption, respectively, is estimated as 24,000 and 10,000 tons of wood. Cnsumption by ousehojds and Small ndustry in Rural Areas Rlda Consumption 121 Various activities have been undertaken to assess rural fuel consumption, generally within the framework of reforestation projects. They show that firewood is almost exclusively the only fuel used in rral areas for cooking food and related acivities. Various estimates of the quantities consumed give figures convergig around 500 kg/year of firewood per capita. In most cases, this wood is gathered free of charge by the peasants on their land; in certain regions, however, especiay in large agrinuural zones where sugar cane or 'vetiver are cultivated, rural marketing of firewood is beginming to appear, sometimes even for residential uses. Ihe use of  . 9 - charcoal Is relatively limited and concentrated in a few areas with large charcoal production activities. ==bL bmauJDd. 1.22 In 1985 there were 473 'guildivese and 'moulinsguildivese (small companies producing supr and drinkng alcohol), mainly in the southwest peninsula of the country. Some of the moulin-guildives use bagasse from sugar cane and more modern guildives use diesel but most of these small industries use large wood. This is also the case with 31 essential oil factories; a few more than half use firewood for their oil extraction process. It is estimated that in total, the annual wood consumption of these small enterprises is 130,000 tons and 30,000 tons, respectively. ^cQet ~ ~ ~ ~ ~ Wodfe Resources 1.23 The extent of Haitis forest resources is not well-known; the last estimation was based on aerial photographs taken in 1978, and comprised only natural forest formations. A World Bank environmental study (BDPA) undertook an evaluation of forest resources based on the 1982 land use map which was established using the aerial views from 1978 (see Table 1.5). As this map was not created specificaly for foreatty needs, the estimates which are drawn from it - the most complete and most precise of all existing estimates - remain at best approximate, even according to the authors of the map. Still, it is possible to note the importance of fallow land and tree crops as standing imventory for biomass, and probably the principal source of wood in Haiti  - 10. IAM*.JJl Ille_oS Stncdtlg Stocks In 19 8 Areo Volim (1000 he) (2) (10000 aS (2) Bread-tleaf foret dse 14 1X 2l1O 46 degraded SO 2X 2500 5X rogneration s0 32 3200 62 Pine fret dne a oX 1200 2 opn 20 1X 600 1X wry open 40 12 520 1 Aro-psturel fost prosopic forest S 0A 250 1X ghrnis (ds) 100 4 2500 5X shrus (opn) 15s 7x 2405 5s Nrove 1S X 525 1X en treo crops 170 a 5100 10X (fruit tre, coffee) open tree roPs 40 14 "40 13X Othr (h eous 1682 612 10095 202 crops, swrt rottion fellos, sesom, rock, etc.) TOTAL 2769 100 37395 76 jgg*s UDPA, 19 Ch SUDD1VNetworlu 124 Charcoallsdistibutedthroughanextreme deentraledurbanetailandwholesale network In the greater Port-au-Prince reion (Port-au-Pince, Delmas, Carrefour, and P6tionville), there are an estimated 3000 sales points, as presented in Table 1.6. There are nearly 200 large varehouses with capacity for more than SO bap and, at the other end of the scale, nearly 1500 mostly female, small reiers (the marketing of cbarcoal, both at the wholesale and retail levels, is carried out mosty bywomen). Much of the harco is sold in the markets, induding the wharfs, wch comprise nearly 3/4 of the market for sales in ba.  * 11. Iglas.6. Charcoal retail poste couted In Port-w.Prtnce LWS0 than N Nllr of retall posts Ibg I to 10 11 to so thn SO Total X Port*au-Prince 749 277 260 11? 1,423 62 Carrefour 282 134 131 44 591 26 * 0t1e 119 37 24 2 182 8 P6tioittll 57 34 23 2 116 5 Nwkets S81 283 309 151 1324 5? am" 23 1S 10 3 51 2 AloN the street 0 184 139 t1 334 14 Alon the road 603 0 0 0 603 26 Total conted 1,20? 42 4SG 165 2.3i2 100 X 52 21 20 7 100 Total esttzted 1.800 520 5o0 180 3,000 Z60 17 17 6 100 IU1CRV~~~~~~~~~~~~~ .NE 1% 12S Approimatey 37% ofte charcoal consumed in the capital is transported by boat kmo the portsoftewo pen 1sa nd from onave island; the remaiing 63% is delivered by truck or other vaious modes of road traport The practioe of Ocharbon-provione which refers to c od transported in passenger vehicles to escape regulation by public officials - a practice tlaerated by the Foresty Svice up to a limit of 10 bap -swidespread. About 15% of the charcoal supply enters the cit by this meas Charcoal generaly is delivered by rural merchants (uswaly female, referred to as madame Saras) who hire transpot to come and seU charcoal in the city (see Fi4ure 1.1). 1.26 Carca production, entirely done with tradionl methods, has been the object of only some limited inestitions Ihe charcoal producers use small capacity sak (5 to 10 bag) of various forms (pasol circlar, or recaular bateau' kilas). the obsved yield by weight are quite high (20%) in areas wth largrscale prduto but it seems that yields are lower in areas where carbonizaton acdvity rcently has been esbied. As there is very little quantitative data, a campain to take measuem-ents fom a pre ative sample of kilns i very much needed.  .12 - Eigpre1.1: Charcoal supply systems in Haiti according to Stevenson (1985) Produce/charoaler Reron Ma\rket Trasprtler Collector Wholesaler In Port-au-Prnc Lal Merchant Retailer  .13- 1.27 IL comparison of two maps, one created in 1980 (Adas CNRS 1985), the other in 1990 based on Information obtained from various suiveys conducted during this study, shows how wharcoal production has become widespread throughout Haiti The extension of charcoal production dates primarily from 1986, when public control over forest exploitation in Hait disappeared for all practical purposes. 128 The origin zones of charcoal delvered to Port-au-Prince in 1980 and in 1990 are shown in Table 1.7; the results of a revised count of the volumes of ' rcoal entries in Port-au- Prince are also given in Annex IV. A progressive diversication of supply to the capital can be seen; to the traditional production zones in the northwest and the south have been added new zones, notably the Grande Anse (southwest peninsula) and the Central Plateau. There are no longer hardly any zones in the country, with the cweption perhaps of the far northeast, where dharcoal is not produced for Port-au-Prince (see maps for 1980 and 1990 in Annex XIV). Table 1k,: Production areas of Charcoal since 1979 in percentaee Department 1979 198S 1990 North vest 50 33 21 south 30 36* 2 Central 10 1s** 13 Gonae 10 3 5 Grande-Anse 0 ` 13 south Eant 0 9 3 Uast 0 4 26 Artibonite 0 13 North 0 0 4 t tncludinG Grande-Anse I Including Artibonite Source: Karl Voltaire, 1979 University of Maine, 1986 Supply ME rgin and Rvne 1.29 The cost to the residential buyer of charcoal in Port-au-Prince, when sold by the bag, is about 1 Gourde/k: each bag costs 40 Gourdes and weighs on average 38 kilograms. When sold in small quantities ('lotse) with unit value varying between 0.5 and 1.5 Gourdes, charcoal costw on average 25% more (1.25 Gourdes/kg). In pots or baskets, the cost of better quality charcoal ("gayak! wood) can even reach 2 Gourdes/kg (see Annex X). In the other cities, charcoal is a litde less expensive: sold by the bag, it costs about 0.7 Gourde/kg in main cities wh as Les Cayes or Cap Haltien, 0.5 Gourdes in smaller town such as Gros Morne or Camp Perrin (southwest p)ma.  *14- 130 The charcoa producers share of the rt price has decreased Unremittingr, this is related to the general sagnation of income In Haits rural areas. However, a Oprofessional producer places a value of 1.5 Gourde/hour on his work, which is compaable to the income of a peasant (8 to 9 Gourdes/day) and sround 30% higher than the rate of a day laborer (6 Gourdes for a S to 6 hour day, see Annex 10). 131 Although some wholesalers and hanporters have rather lre sales figures from marketing cha baicaly because of the lag volumes concned (a truck making 100 trips annually can geerate a net annual income of $10,000 to $15,000), the margs for tansprte, wholesalers and retaile conform to the standards generally found in networks for wood and charcoaL 132 On the basis of information collected during vadous surveys, the transport price for dharoal is about 0.1 Gourde/day/kilometer (regression analysis performed on a dozen loads delivered by road, and about 50 statements made by tan ). Taking into account the tansport costs in Hati, whih fluctuate between $1.1 and $2 per km for a large trucks according to the state - paved or unpaved - and topography of the roadway (see Annex X), the trawporter ears a net margin of approimate 30% to 40%. The urban intermediaries (wholesalers and retailers) earn a gross cumulative margin of 25% of the reta price (see Table 1.8). Tabtl 1.8: Charcol Price Structwr Ooureu/Ra (d) Sorde/Ton 1. Produoer/charcoaler/collector priee 14-17 368-446 35.42.5 2. Tranwort to Port-au-Prince (a) 6-9 158-236 15-22.5 3. Tax 1 26 2.5 4. Intermsdiariaa/other mrgins 5-6 131-158 12.3-15 S. Wholtster purcde price (1.2+3+4) 28-30 736-788 70-75 6. Wholesaler margin (b) 5-6 131-15S 12.5-15 7. Retaler mrgin (sale by boo) 5-6 131-158 12.5-15 8. Averae prfce per b 40 1050 100 9. Retailer's additial margin (e) 8-10 210-263 20-25 iJl: (a) Including loadingunloding cost. (b) UWoltler's marin can be 6+7 Wimen the wholesaler sells directly to cornsrs. (c) Sale fn mlt qwntities ("lot", basket, pots etc.). (d) N of 38 in ver. WM: Surves EIP/8NE/PNU/OMD of 67 charcl transporters In the North-wet, South-west and Vat region. Wood SIpp e,r 133 Because transport costs are high relative to the sales price of fuciwood, the supply networks for this fuel are conideraby shor than hose for charcoaL Thus, a large part of the suppl for bakeries in Port-au-Prince comes from nearby "mornes" (small Isolated hills) and from  .15. areas close to the Central Plateau; In Gonalvo, wood I sled from the mangroves. Frwood tuck headed for the capil usually follow routes which alo thom to avoid official controls. The guidives and essential oil factories usually are suppled with wood from fut tres (manges, avocadoes), as these tre often are Juded useless or too numerous by the peasants since the eradication of pig4 134 Liquid perleum produsb are imported by Srlt=O (Shel Interaional), Exxon and Taoo Trader from refineries operating in the resio Kerosene normally comes from uncao from a fomer Shel refinery whih now i operted by a ubsidiay of P6toleos de Venezuela. Products an Utnspored by Genral Purpo tankers with 36,000 to 40,000 ton cpacity, which then continue on towards the Bahom 135 Propane Vs fnhed by Sll 0 to two disribon companies, Tropipz and Hal Gsz comes frm Punta Cardon (Venezua), Mexico, or the United States. Gas is tranqsorted to Haii by tankers owned by Tropi;z fiv times per trimestr. The trport cos re d b 30 cents/glo, equivalent to about $150/ton (see par. 3.47). 136 With roegrd to propne strge, Tropigaz has three tanks, one with 30,000 gllon capacity, bought from Es in 1953, and two others with 63,000 glon capacity, insaed in 1978. in addition, Haiti Gaz owns a unit with 30,000 glon capacity. Thus the total storage capacity is low (400 tons in water, ie. 320 tons of gas), and is on sufficient to guarantee a litt less than 15 das of sup, wit frequent ortaes of 1 or 2 days. Tropigaz also has a botting plant with three Ming lines, with a total capacity of 8000 g per day (16 tons/d), also quicly g saturtion. 137 Butane gs also is spped by SlTCO, mainly from Venezuela, and resold to Sodigz by Shell Haiti I 1988, Shel made ivestments of $800,000 for diuting Bip doves and butane. The invstments Included: (a) reoeption fa and two storame units with a total acta capacity of 130 tons (150 nomial) ($600,000) (b) a fMing cent with capacity of 900 to 1000 botdes per day (per cyce of 8 hours), or 3000 botdes daily fr tre cyce ($100,000). (c) various preliasq marketing studies, icluing a tudy of the taget client (see pars 3.53) ($100,000).  - 16 - 138 Both bottling plants for propane and butane are located in Port-au-Prince, whih absorbs the quasi-totality of butane consumpton and most of propane consumption. Propane and butane bottles are tanorted to other main cities by trucs and sold by distributors at a price reflecting additional transport cost (for instance the refiling of a 12 kg propane bottle costs about 40% more In Cap Haitien than in Port-au-Prince). 139 The Shell group plans to restructure its storage and bottling system in the medium term to face forecast demand of butane and propane at their current prices. It is possible that marketing of propane then would be dropped in favor of increased marketing of butane as part of a Caribbean butane strategy designed by headquarters, which projects the creation of a major storage center of butane in the area (possibly in Puerto Rico). Insituioal ramwok and Ongoig Actions lwnstitudn F-ramew-ak 1.40 Among the major publc institutions concerned with the sector, the Bureau of Mines and Energy (BME), through its Directorate of Energy Resources (DER), is responsible for (i) promotion of national energy resources and, more generally, of all forms of energy which could contribute to the country's development; (ii) rational energy use; and (iii) environmental protection in relation to mining and quarries. The DER is staffed by nine professionals and had an annual budget of 1 mlllion Gourdes in 1988. Its activities indlude management of various energy projects in cooperation with French assistance and OLADE: energy statistics, development of solar and wind energy, improved stoves, and use of bamse. The DER currently is the recipient of a program to stengthen its operations, financed by PNUD and OLADE (Energy Sector Strengthening Program), of whic this study is a component One of the main objectives of the program will be to prepare a National Energy Plan. 1.41 At the Ministry of Agriulre, Natural Resources and Rural Development (MARNDR), the Forest Resources Service (RS) is responsible for protecting, managing and developing national forest resources. In particular, it is responsibe for implementing the National Forest Project, which wil enter its second phase in 1992 (Forest and Environmental Protection Project). However, as with many rural development activities in Haiti, most of the reforestation activities are carried out by numerous non-governmental organizadons (NGO), both Haitian and foreign, because of difficulties encountered at the MARNDR such as: limited operating budget, lack of coordination, lack of poliy continuity and lack of motivation among the employees 1.42 With the exception of public installations for security stocks of diesel and kerosene, the Haitian petroleum sector is totally private. Three companies share the market: Taco, Esso and Shell. The latter is the only company which markets LWG. Within t te Ministy of Industry and Commerce, the Bureau for Petroleum Product Supply (BAPP) supavises the sector and  - 17- regulates the price of liquid fuel; in pratice, although BAPP is conscious of the distortions in the price of gas (as shown in the third chapter), its resources and means to act are limited. Chaal Read nd ITaxao 1.43 A decree dated October 1987 grants MARNDR the ability to authorize and control the cutting of wood and also gives wood-consuming companies a period of 6 months to subsdtute wood by other fuels. In practice, this decree has not been applied. On the contrary, all control over forest exploitation - and charcoal exploitation in partcular - has disappeared since 1986. This engenders two main negative consequences for the sector: the universalization of charcoal production throughout the country and the slowdown of price increases for charcoal 1.44 There is a tax on the transport of firewood and charcoaL levied at the district leveL in addition, the forestry service has a number of checkpoints, which operate with unequal degree of effectiveness (see map no. 2 for Port-au-Prince in Annex XII). Transport permits certifying payment of the tranwort tax and the number of corresponding bag (often underestimated) are verified by forestry agents at the checkpoints. The official tax on charcoal, originally 025 Gourdes, has risen since October 1989 to I Gourde per bag. The forestry service records show that 368,000 and 513,000 bag were checked during the fiscal years (October to September) 1987/88 and 1988/88, respectively. This amounts to respectively 14,000 and 20,000 tons of charcoal, ie. less than 10% of the charcoal which in fact was marketed in Port-au-Prince. 1.45 Special effort is being made to improve the monitoring of the volumes transported and to improve collection of the charcoal tax, with positive results as shown in Table 1.9. Despite these improvements, there is room for much more progress: an estimated $800,000 additional revenues in charoal taxes could be collected by the State if control was fully eflcientL The firewood tax theoretically is 1 gourde per piece of wood, in general 1 gourde per stere; in practice, it rarey is coecte Iabtl 1.9: Chgrcosl Tax RevWrs Total 1bnthy % of theoretic (x1000 Gwurdes) revw rwwe revw. 29 Augut - 15 Decber 1918 48.8 13.9 2 3Jms mary - 30 larch 1989 75.0 25.0 4 10 April - 11 July 1989 146.2 48.7 8 11 july- 31 October 1989 150.9 40.8 7 31 October - 29 Decedbr 1989 141.4 70.7 12 source: lASSO, Forestry Project. 1.46 Besides the official taxes, there is an entire unofficial ytem of levies, the revenues from which do not go to State Treasuay. This was true before 1986 in the case of charcoal eploitation, where the merchant often had to pay local authorities more than 3 times the value of  * 18. the official tax. Ibis s stil the case for charcoal tansport; the ransq must make a contribution, either in charcoal or money, at various checpoints along his route. Thus his real not margin is diminished. 1.47 Numerous reforestation projects are underway in Hait mainly as part of programs for non-governmental asstce to rural areas: an enionmental study (IDA/BDPA) catalogued more tban 80 project aimed at limidng deforestation (soil conservation, seedling plntations, etc.) A Tropical Forestry Action Plan (TFAP) wil be formuated during 1991/92 under the suevision of PNUD and FAO in order to coordinate the different projects under a common poliqy frameworkL With regard to plantations, two approaches have been attempted, with very different results: large plantations and agroforesry. 1.48 The use of large plantations as a strateW for fghting deforestadon was attempted under mainly the Operation Double Harvest (ODH) financed by CARE, one of four components in the Agroforstry Outreach Project (AOP) financed by USAID, which has met largely with failure; as a result, the large plantaton option has been completely abandoned at present The ODH sought to implement large per-urban plantations on land belonging to major landholders: this project was halted in 1987, mainly because of a lack of interest on the part of the landholders Moreover, the National Foraty Project (IDA Credit) included an enerV plantation pflot component, which was based on creation of planted perimeters at three sites (covering a total 350 ha) and on management of part of the Pins forest (for an area covering nearly 6,000 ha). Despite positive results with regard to sgthen the FRS and preparation of the second phase of the project, the results were relatively di nting for the pilot plantations, largely because of the lack of counterpart funds, negtive reactions from the rral population, and the influence of eal factors (climate, livestodc). Because of the lack of financial profitability of such plantations, the second phase of this project is therefore oriented towards agro-sylute formulas established at the level of nul communiies and mement of foresty resources. 1.49 The objective of the agroforesty approach is to promote various types of tree planting integrated into Haitian cultivation xprces. The AOP Project in the northwest of the country and the 'Pyebwa' Project which was ilemented by a network of more than 400 NGOs in the rest of the country, have been quite sucesful with very small-scale rural forestry projects. According to the BDPA, apprcAimately 8 million treesu have been planted annually as part of these projects over the past 10 years. Tnhere have been sevewal evaluatkos of the success rate for planting schemes of this tpe, with varying resulb - the trees' survival rate is 80% according to CARE, 33% accrding to a more rigoroui evaluation carried out by the Pyebwa Project, and between 50 and 60% accrding to other project evaluations. A realistic tree suvival rate after one year could be conidered 40%: thus about 3 mllion planted trees uvive each year. By comparing the cost of the AOP Project (US$ 3 million annually) and the other main proects aged in this type of planting with the number of surviving trees, the cost per suviving tree in these agroforestry plantatons is estimated at US$ 15 (however, some estimtes go as high as $10 per surviving tree). e  * 19. Tnese fiWes only indude the cost of interational fore aid, and the cost of the land and the labor costs shoWd be added to obtain totad cost proved staves 150 Several small projects to dbitriute metal iWoved charcod stoves have been hunched in Hati, but only with vety limited results: (a) Between 1983 and 1988, the first National Forestry Project included a $90,000 component for distributing mproved stoves financed by the Canadian Center for International Research and Development (CIRD). This project, implemented by BME, conied of evaluation of eisting traditional cookstoves, development of adapted improved stoves (BMFB models) and trial dination in two small cities on the central plateau, Lascahobas and Beladere (30 stoves were distnbuted). (b) In 1989, the BME reeived asane from OLADE for a new pilot program to distribute nearly 600 portable cookstoves (BME model), with total financing of $15,000. A preliminary phase induded developing a working set for the craftsmen (tools, templates) and trainig a dozen craftsmen. Then, in association with two NGOs and with tecnical asitance from BIT for training, the BME opened three workshops in Port-au-Prince and reuited six craftsmen. Accordiug to xpected rel of the pilot program, an anticipated 7200 stoves - sufficient to sply about 5% of the households in the capital - wM be ditibuted by mid-1991, under additional financing of $15,000. (c) At the beginning of 1989, the French Associadon of Volunteers for Progress (AFVP) opened two training wrshops trgeted at the 'manje6 kwit market (small restaurants) and trained ten stoves craftsmen. The traiees began by producing a commercial line consisting of four model stoves of varying sizes (the BME model with feet): 500 stoves haoe been sold since September 1989, but mainly for domestic use. (d) Finally, some larger wood-fueled models were implemented in restaurants in public institutions, with the tecal assistance of the Renewable Energy Development Institute (REDI). 151 lhus, approwimatey US$ 120,000 has been invested in improved stoves (not counting the vaious msions by experts), and about 600 portable cooktoves acually have been d ted over the past 7 years. total may seem smal, but nonetheless there have been certain positi outcomes to these trials The most important of these is the design of a model adapted to the needs of more modest Haitian households. his model competes with the traditional round stove, which is used by 2/3 of a households. It has a ns ory efficiency (30% in theory, and about 25% in actual use - repesentig a 30% theoretic povement in efficiency over the traditional  -20 - stove), and It appears to be well accepted by the consumers (reasonable level of sales by the AFVP, positive reacdons from groups during market testing caried out for the study). 1.52 Another Important outcome, obtained through the activities of BME and AFVP, is the result of comparisons of various methods for man ng the stoves and organizing their production and sales. The BME's approach was to employ three groups of two artisans and to manufacture the stoves out of sheets of new galvanized steel, for a net production cost of $7 (in fact the stoves are marketed for $5 to $6 a piece). The AFVP decided to use sheet metal raade from material recovered from drums and junked cars, and to let the artisans operate autonomously once they were trained. The sales pr'ce of the stoves thus constructed is $4 for the producer, $5 for the retailer. Given experiences with this type of project in other countries, the AFVP approach, which depends on the autonomous initiative of the craftsmen and an initial market penetration using low cost products (possible through use of rerycled manufcturig materials), seems preferable. However, if a shortage of recoverable material arises (likely to cause the price of recovered materials to increase), the use of new sheet metal would be necessaqy.2/ Besides, a demand for improved stoves constructed from new sheet metal is likely to exist already because of their superior durability or for aesthetic reasons. Substitution 153 Efforts to substitute other fuels for wood and chaoal namely kerosene and LPG, concern mainly the domesti. sector. A program also has been undertaken for small busineses and rumal enterprises, although with limited success. 1354 The domestic use of kerosene for cooking is relatively limited. No promotion progrms have been undertaken, especially by the Government, and thus there is only a limited but dynamic marketing of the product by the private sector (approximately 1500 stoves sold annually). One company, Haiti Mtal, manufactures and markets the stoves, which are sold for between $20 (one burner, no stand) and $43 (two burner with a stand). In 1989, this market totaled 1140 units, with more than 90% of the two burner modeL Other companies market stoves imported from Venezuela (Haiti Gaz) and from southeast Asia (Electrotostore, with 200 stoves sold in 1989). 1.55 With regard to the small businesses and industries, the former National Bank for Agricdultural and Industrial Development (NBAID), with assistance from the European Community), introduced a program to substitute wood with petroleum products Any company which wanted a loan had to switch to a fuel other than wood. About twenty bakeries switched to kerosene, and a few others to propane, under this plan. ZI The current price of new sheet metal appears excessive (sales price nearly double the CIF rice); nonetheless, given the experiences in other countries (Senegal, for example), allowing e metal workshops to directy import sheet metal in order to mass produce and sell improved stoves does not seem viable, as distribudon of the stoves (which is diffct to organze) certainly would cost more than the resuting savings on material costs.  -21 - 156 .oane gaa Tropigaz inports propane, and distrbutes it along with another company, Haiti Gaz, in four sizes of bottles -5 kg (a vety linited number), 12.5 kg bottles (most of the market), 25 kg and 50 kg - and in bulk (1000 kg) for large companies. In 1989, Tropigaz came under the control of Shell Halti, when the Caribbean and Latin American operations of Tropigaz were acquired by ShelL 1.57 In 1986, Tropigaz made an inventory of its bottles, which numbered 66,000 in the 12.5 kg size and 25,000 in the 50 kg size. Taking into account all sales since 1986, equivalent estimates for 1990 would total 85,000 and a little less than 30,000 respectively. This averages out to about 2.1 bottles per user (assuming 50,000 propane consumers, of whom just under 80% reside In the capital). 1.58 Various local firms distribute imported cookers fueled by propane gas. The least expensive come from Latin America (the 'Llama azur from Venezuela, for example) and vary in price from $50 to $70. Given the purchase cost for a botde ($73A for a 12.5 kg bottle, without the pressure regulator), the minimal initial investment is $120 to $150, a very high cost for most urban households. Only several hundred cookers at the lower end of the scale (cooker without an oven) are sold annually. Tropipz also has commercial cookers for institutional restaurants and small food vendors (between $400 and $500, plus the regulator and the 50 kg bottle, ie. an initial investment of $550 to $800). 159 BnGazs, In March 1989, Shell began marketing the 'Bip Trcheri stove through the Sodipz company. Tis is a single burner butane cookstove which already is distributed in vanous West African countries. The stoves and containers are bought in France at ADO (Camping Gaz) and the stand is manufactured in Haiti Shell has received government support through a 'balance of payments assistance agreement", signed in October 1989, which allows it to import materials under an exchange rate of 5 Gourdes per dollar. The agreement provides a total _athorization of FF 30 million in 1990 (US$ 5.2 million) for importing botdes and stoves, renewable in subsequent year An initial authorization was granted in December 1989 (FF 1,655 million, or about $280,000) to import 14,000 units, followed by a second authorization of FF 3 milion (US$510,000) in May 1990 to import 25,000 units. If the 40% devaluation of the Gourde is taken into account, the subsidy accruing to the Bip program from the Haitian Government amounts to about US$300,000, equivalent to $8 pcr stove (see Table 1.10).  C22 . labt 1.10: liP Price lmport Iport Excae 400 over- Price Price I for 5 valuatfon Ff U#$ SNaitfan Sa1tian Contafner 80 13.79 13.79 19.31 Burner 31 5.34 5.34 7.48 Custom and Ahilnist- retive ies 0. 1.07 Pot support 3.25 3.25 Stove support 4.00 4.00 Gas 2.12 2.12 Equipment cost 29.27 37.23 Wodiga sale price 2.23 -5.73 Sodigs margin 31.5 31.5 Retail prfee 37.5 37.5 Dfstribution mergin 6.00 6.00 Exchang rate: U181 a fF 5.8 S3urcsa SHELL HaNti, June 1990. 1.60 Promotion of butane has been vey sccessfiul: 32,400 portable stoves, 44,500 bottles and 640 tons of gas (311,000 gallons) were sold in 14 months (between March 1989 and April 1990). The rhythm of portable stove has slowed down in May 1990 (only 500 stoves sold, as compared to a previous average of 2000/month) for two main reasons, according to project promoters: the price incase for the stove from $29.S to $37.S in Feruy 1990, and also a cetain exhaustion of the market, which initially was sustained by purchases made by institutions and businesses for their employees. However, butane penetration started again in 1991 with a 10% increase in the sales of gas when comparing the first semesters of 1991 and 1990, and a 20% increase when comparing the firstiwo quarters of 1991. In addiion, Shell intends to introduce the 6 kg butane bottle on the market, which should lead to a further increase in gas sales 1.61 In the ioitial start-up phase, the company sold the stove-container combination to customers for a price $4 below its net cot at wholesale level lThis was done to promote the distribution of the product; the loss was recovered by iceing the margin on the sales of butane by $0.50 per galon This inflated margin (it is difficult to label it a cro subsidy in the absence of reguLation of gs pricing) was a shrewd marketing decision for the companies: currently, evaery sale of a portable stove provides a granteed market of about 18 gapons annually, based on an average of 12 recharges per month. ITI represents a profit of $9/year, more than twice the $4 lost on each stove sale. Access to a advantageous ehange rate in addition to the increased price of the Bip portale stove, now selling for $37.5, has allowed the company to balance its sales price. Sti, the price of butane remains at $3 per recharge.  -23 - Legislation and Pricing Policy for Petroleum Products 1.62 The Haitian State sets the price of liquid petroleum products through its Petroleum Technical Committee, which consists of representatives from the Ministry of Industry and Commerce (through the BAPP), the Ministiy of Finance, the Central Bank, and the BME. The margins for kerosene (price at the pump) are negotiated between the State, the petroleum companies, and the National Association of Petroleum Product Distributors (ANADIP). In April 1990, the margins were set at about 20% for the companies and 5% for the distributors; they appear reasonable compared to countries with similar distribution networks. The companies receive a preferential exchange rate (5 Gourdes/US$1) when importing products, equivalent to a state subsidy of about 40% of the CIF value of the products. This preferential arrangement was only applied to LPG in January 1987; LPG is now imported at an echange rate of 5 to 1. 1.63 Kerosene had not been taxed for many years in order to promote its use in the productive sector (bakeries, for example). However, this has given rise to problems related to competition with diesel fuel (fraudulent distribution and usage of kerosene). In October 1985, new taxes were applied to kerosene, totalling nearly 35% of the -retail price. It should be noted, however, that in order to decrease the price to the consumers located in the provinces and in rural areas, kerosene is sold duty-free outside Port-au-Prince, in "yellow" gallon containers. The 50 cent/gallon value of detaxation pays in effect for the reusable empty container. About 500,000 glons are sold in this manner each year, equivalent to a litde less than 10% of the national consumption. 1.64 The prices of gas (propane and butane) are fixed by the two distribution companies, Shell for butane and Tropigaz (100% Shell owned) for propane, with formal notification given to the Ministry of Commerce. Tbis de facto monopoly in the gas market leads to cmulative margins for the petroleum company which seem to be high in relation with similar markets in other countries: in April 1990, the cumulative margins were 57% of the retail price of propane and 71% of the retail price for butane (47% for Shell Haiti and Sodigaz, and 24% for the retailer). The LPG tax was ended in March 1987. This decontrol and the drop in price, as well as the preferential exchange rate, accompanied by an intensive campaign launched by distributors, all incited a massive increase in gas sales (more than 75% since 1987), and in the sales of cooking equipment.  .24u IL MAW ISSUES 2.1 Tbe main issues affecting the household energy sector in Haiti must be taken into account when developing a household energy strategy. These are: (a) the environmental impact of woodfuel utiliztion on forest resources; (b) the inpact on the balance of payments of encouraging the use of imported fuels as substitutes; (c) exsting distortions in the gas market; f(d) the essential role of charcoal production in the rural economy; (e) the weakness of the institutonal frameworL IMpact on Forest Resorce and th nvrnmn 2.2 lTe current estimate of forest resources drawn from available data (see para. 1.23) was established using projected growth rates for forest formations and wood consumption. Several fictors make this approach very imprecse and limit the etent to which the rate of deforestation can be evaluated: (a) there are only extremely fragmentary data on the growth of natural forest formations (the evaluation of yields from the secondary srub formations used in most charcoal production is based on unverifiable hypotheses in the current situation) and even on the growth of planted trees. (b) the impacts of reforestation using isolated trees or rural micro-plantation is not quantified; (c) the exact role of woodfuel consumption in the deforestation process within Haiti has not been identified clearly and quantitatively. hi is particularly the case for self consumption in rural households: it is difficut to quantify wood consumption in supply/demand balances when the source of this consumption is felled wood or wood gthered from the wastes of land clearings or tree pruning.  -25 - (d) the impact of other phenomena such as the subdivision of agricultural plots Into small parcels, the increase in the culthation of food crops, and the reduction in cultivation of cash crop is difficult to evaluate and quantify.2/ 23 Although in certain cases environmental damage caused by intensive charcoal exploitation cao clearly be demonstrated, charcoal exploitation is only one of the causes, and probably not the principal one, of deforestation in Haiti Among other causes of deterioration of Haiti's environment are: (a) increased demographic pressure and its corollaries: the subdivision of agricultural plots, farming on lands not appropriate for agriculture; (b) poorly adapted and harmful agricultural techniques combined to extensive tree felling; (c) insecurity regarding land tenure discourages the use of cultivation practices which maintain soil fertility, and on the contrary encourages overexploitation of the land and discourages investments; (d) the lack of capital and income for the small peasants: they generally must fall back on subsistence farming of food crops, using very erosive production techniques, and have recourse to the production of charcoal as a complementary source of monetary income. 2.4 Degradation of the environment has immediately perceptible effects, especially with regard to erosion, loss of sediments, and decreased agricultural yields. Thus, overexloitation of the Haiti's ligneous potential does not constitute only a decapitaliation of available wood: It also ha secondary effects on erosion and alters local hydrology. 2.5 It is diffdcult to estimate the rate of deforestation in Haiti given the lack of recent data and reliable historical comparisons. Aocording to the BDPA study, the estimated forest cover (degraded ot dense broad leaf forests and pine forests) was only 220,000 hectares in 1988, amounting to less than 10% of the productive biomass surfaces and less than 30% of the wood capitaL ihe greater uncertainty surrounds evaluation of wood resources available from non-forest sources: shrub formations, tree crops, herbaceous crops, savamnas, etc - Le. 90% of the surfce concerned. lTe sensitivity to error for these projections is rather large (see Annex I): all other things being equal, if the 1978 figur represent an overestimate of 25%, the forest corer would 2, Furthermore, thesw phenomena can have opposite effecta For example, the reduction in the number of coffee plantations can deprve some areas of an important wood resource (pruned cuttings from coffee and ade trees or wood obtained from felling old plantations), whereas the subdiision of large parcels usuall increase the leDt of tree hedges which are also an important, and regular source of firewood and timber.  -26. disappear entirey by 1992 (an event which, fortunately, seems unlikel); if on the other hand they were underestimated by 25%, there are suffient trees In Haiti to satisfy the firewood and charcoal needs for the not 20 years without jeopardizig renewal of the resource (independently of other causes of degradation of the resource). 2.6 It woud prove unfounded to question the consensus on deforestation in Haiti, or even to reject the conclusions concerning the gradual disppearine of resources. Yet it is unfortunate that the lack of reliable information on wood resources and on the impacts of firewood and carcoal consumpdon on deforestation makes it Impossble to precise design and measure interventions needed in this domain, especilely for charcoal substitution. There are serious economic constrits assocated with partially dismanling the national charcoal sector, as with incasig petroleum Imports; thus environmental jutification of such actions must be caref supported. 2.7 Experience with other projects - mainly those Implemented in West Afiica - proves that detded inmentories often provide data which is vey different from the first estimates of the resources condition and availabilty, pecia for the non-forest resources (isolated trees, wood from fallow land, wood from pruning) whih provide the core source of wood energy. This brings to fore the impotance and urgenc of undertking an inventory of wood resources, based on aeral photos and on-site crosachecks as planned under the IDA Forestry and Environmental Protection Project (probable start in 1992). 2.8 Nonetheless, it is clear that deforesdon in Haiti is an uncontestable reality, even if the scope of its causes and effects large exceeds the basic problem of wood energy supply and demand. Thus one of the fundamental objectives of an energy str for households is to slow the deforestation phenomenon by reducing the pressure caused by woodfuels cnumption. Impac on the Balanc of Pmayet of Promoting Petroleum Substitutes for Wood andChrl 2.9 The large enery plantations have encountered so many obstacles that this option basically has been abandoned in favor of the agroforestry approach. However, wi the latter be able to enure n of forest resources sufficient for preserving the national forest endowment? Under the current assumptions about the resouce, it appears doubful: the current rhythm of agroforestry tree planig, already quite sutained, would have to be multiplied by a factor of 7 or 8 in order to obtain nearly 23 mllon suriing pts each year (see figures in Annex I). This efst woul cost at$30 million pe year, based on the current costs of such 2.10 husactiowhich target on i of woodfuils, essentially in urban areas would be needed: eneg cetion, substituion by petrolm fuels Under the same asmpzaons as cited above, a drastic reduction charcoal cn o would be needed to prevent the obliteration of Hatian fores resources. Even if charcoal conmption were cut in half by 2000 - .a c ae task - with continued reduction effrtsfte 2000, this would only delay the  -27- theoretic disappearance of the forest by about 15 yeas In order for the resource to be maintained in its current state, an almost complete termination of charcoal consumption in the capital and main cities of Haiti within the neot ten years would be needed (Annex I). Therefore it is clear that interventions are needed not only to conserve dharcoal but also to substitute it. 2.11 It is possible to estimate the impact on the trade balance in terms of imported gas or kerosene at current CIF prices if petroleum fuels were to be substituted for an equivalent level of charcoal conCmption/ Consumption of charcoal in Port-au-Prince is estimated at 200,000 tons (consumption by households and the small productive sector) in 1990, and at 100,000 additional tons in other urban areas. The annual cost to the country of importing gas or kerosene in quantities the equivalent of the current urban consumption of 300,000 tons of charcoal would amount to about US$35 million (with theoretic CIF prices, see Table 32), equivalent to about 25% of Haiti's export receipts in 1990. Thus promotion of charcoal substitution potentially could be a very costly solution for the country, and would only be justified if charcoal exploitation presented an irrevemrble threat to the environment. lAe aitin On Maret A Diso4d Situation 2.12 Gas imports and most of gas marketing in Haiti - for both propane and butane - - are carried out by Shell and its various subsidiaries (Sodigaz, Tropigaz). Although other petroleum companies appeared interested in this market, they have not epicitly voiced their interest, probably because of the country's difficut political situation. The de facto monopoly ierited by Shel presents several problems for the country and constitutes a major handicap to the development of gas as a substitute for charcoaL 2.13 One of these problems is the current storage capacity. In the case of propane, it is extremely limited: a Lttle less than 15 days of storage, which often leads to depletion of stocks. Because storage is linited, delveries must be made in small quantities, at considerable transport costs which burden the product's CIF price (in 1990, the transport cot practically as much as the gs). In fact, although the Haitian Government has invested in storage for other strategic products such as diesel and fuel oil, it has no sewrity stock facilities for as. Nor do the petroleum companies plan in the short term to invest in new storage or new bottling facilities. 2.14 The main problem involves the establishment of gas prices (see paragraph 3.46 for more details). In a monopoly sktuaion and in the absence of state regulation, the margins estabshed by the companies seem to be vety high: between 60% and 70% of the retail price, split between the Importers and the distributors. By compaison, the combined margins for gas in a Q/ Taking into account the fuels calorific values and the effciencies of related stoves, it is estimated that 1 kilo of kerosene is equivalent to about 3 kg of chacoal, and I kg of gas is equivalent to 3.7 kg of charcoaAt current CF prices (repectively 0.68 and 0.77 Gourdes per gallon), kerosene or gas rts to substitute for dharcoal would contribute a nearly identical weigt to the trade balance.  -28-. number of countries rarely represent more than 30% of the sales price, even in cases where the markets are very limited (West Africa, for instance). In the case of a landlocked country such as Burkina Faso, which has a very small market (nearly 800 tons, versus 7000 tons in Haiti) and where gas is transported by road from the coast for a distance of more than 1000 kilometers, butane gas taxed at a rate of more than 20% sold for about US$ 1/kg in 1989. This is equivalent to the untaxed retail price of the butane sold along Haiti's coast. 2.15 Throughout many countries, the State used taxes and perequation to absorb the much of the profit produced by the drop in international oil prices (up to August 1990). In the case of Haiti, however, it was the importer who apparently benefitted most.4/ This distorted situation is not a fact which can be separated from the analysis of the sector, as it has direct implications for the problems of comparing costs of household fuels and on taxation of petroleum products. Charcoal Production: An Essential Economic Activit for Rural Areas 2.16 Tlree important factors characterize HaitPs rural milieu: (a) increasing demographic pressure on agricultural lands, due to increases in the popWulation density from 300 to 400 inhabitants/km2 between 1965 and 1985 (whereas in some areas, estate lands and large agricultural domains have been abandoned), (b) a decrese in the size of cultivated land parcels: the percentage of cultivated parcels measuring one "carreau" (1.29 ha) or less increased from 40 % to 70% between 1950 and 1971, when the last agricultural census was taken, (c) a decrease in the value added of agriculture per unit from $185 in 1965 to $172 in 1985 (1976 constant prices), a decrease which undoubtedly has intensified in the last year 2.17 In a situation like this, any energy strategy which has as an objective the reduction charcoal production activities must take into account the economic importarce of this activity in rural areas and the social repercussions which could be brought about due to the loss of income for some producers. Annual revenues in the charcoal sector amount to about $50 million, more than in the electricity sector, more than half those of the petroleum sector and about one third the industrial value added in HaitL This also represents an annual injection of $15 million by the urban consumers into Haiti's rural economy (at the level of production and of rural marketing of charcoal). A/ Under current conditions, with few investments and a quite precarious supply system, the small propane market in Haiti provides almost as much margin to the petroleum company and the related distributors as the combined propane sales provide to the different importrs and distributors operating in the Dominican Republic, which has a market 12 times as lre (about 100,000 tons/year).  -29 - 2.18 In terms of employment, it is estimated that more than 150,000 people are Involved in charcoal supply from rural production up to retafl marketing; at least 67,000 of these are charcoal producers J/. Charcoal production is an income source for a large number of urban distributors: for the small neighborhood retailer, charcoal marketing constiutes a non-negligible complementary income source, and for certain wholesalers and transporters the income generated is substantiaL 2.19 Charcoal production also has become a crucial source of income for the economic su"ival of some areas which are especially disadvantaged, and more and more, for a substantial portion of HaitPs rural areas. The subdivision of farm land, the eradication of pig husbandry, the abandonment of coffee cultivation in favor of food crops are all factors which encourage the peasant to turn towards one of the few activities from which monetary income is assured: charcoal production. 2.20 In the district of Anse Rouge, one of the areas in which charcoal production has long ben established, the forestry servc estimate that there are 4000 charcoal producers and 2000 merchants (there often is overlap between the two activities). Most of the rural households and a large number of the urban households in that district are engaged in producing and marketing charcoal which, along with salt production, is the only possible activity. In other dtricts of the Artibonite or in the north of the country, swh as Saint Mkbel rAttalaye or Dondon, between 50% and 80% of the active mral population is engaged in carcoal production, and in other areas with better agricultural possibilities (northeast, southwest), between 20% and 30% of the popuation. 2.21 According to surveys of charcoal producers in the Artibonite and in the northwest of the country, crcoal income has become more important to the peasant than agricultural revenues (even though it is not possible to preciely estimate the very variable ratio between income from the two activities). In similar surveys underaken in the south of the country, the money from charcoa generally only appears as an occasional source of income, emcept during drought periods. It is sinfit that this activity, geneay the work of small peasants (as is still the cise in the south) now is done in the north by medium and large peasants Mhis appropriation of charcoal production and marketing activities by the local petit4bourgeoisie (perons of influence, merchants) in the Artibonite and in the north of the country occurs at all levels, indluding charcoal collection and ransport to consumption centers, whereas these activities are carried out by peasants in the south. 0/ EEstimate obtained by dividing the total charcoal consumption by the average estimatd production of a single charcoal producer.  .30 - 2.22 The efficency of the Haitian private sector varies aording to the pnts The private sector has proven quite cesfl in case of charcoal marketing, which functions in a satlsactoy manner witLout large nvestents, and also in the case of butane promotion, which in three months had achieved its initial objectives for dissemination. Its efficen has been less obvious in the case of kerosene; its promotion as a cooking fuel has not been sucesL In most cases the private sector has adequately fulied its role: the responsbit for distributing fuels and cookstoves is incumbent upon the private sector alone, with the as of and under the regulation of public authorities. 2.23 Various indicators Mustrate the specific instutiona difficulties in Haiti, and above al the budgetay constraints faced by Haitian institutions. In order to operate satiactorily, a central admiistration should have a central operating budget (minus sabies) equivalent to about half the total salarieL However, the 1988 operating budget for the central adminition was situated at only one third of normal needs The two mistries Involved in the different aspects of the enery sector, the MARNDR (spervising miisty for the forestry sector) and the Ministry of Public Works, Trnsport and Communications (supevisng ministry for the eneru sector) fnd themselves in partularly critical conditions: these two miistries were forced to operate in 1987/88 with expendituretalicatons equivalent respectiv*elyto only 10% and 12 % of their actual operating needs; while the BME, with a 7% ratio between operating expenses and salaries, was ganted ony 20% of its budgetary needs to work with. 2.24 Withoutfundstofinanceopatingand maexpendituresthe adminttion canot operate nrmally. The limited impact of various laws implemented in the enerD sector (sch as the requirement that busiesses switch from wood to other fues), the lack of ordination between the various projects, the delays and the negligible resl acbieved in implementin progams in the past (such as the projects to dbitrbute improved stves), are eamples of this. 225 A substantial number of actvities financed by international aid tend to bypass public authorites, preferring the efficency of local orgaons, in paricular the dick network of NOOs (minimum of 400) operatng throughout HaitTi solutio, which is very advantageous for short. term and quick interventions and actives, gives rise to a cetain number of prbems when a long- term pepective is taken: dispersion of efforts, uncoordinated approaches, duplication of interventions, and difficulties with supeOvisiov, etc. 2.26 Because of these prbems, the goenment should make a renwed effort to fulfil its mandate by defning the directions and coodintin the actions taken for household enerV. }iven the weak Insituonal framework, it thus woud be necessa to the extent posible: (a) to concentrate stratelD on a small number of components, to avoid dispersion of public intervention on minor topics;  - 31 . (b) to take into account the currnt wealmess of internmtion means and plan for necesay strengthening of institutions (adequate coordination betwee the government and NGOs) and of logistics p=oWoof a Household ESn=rySta 2.27 Whatever the degree of reliability in the information on the resource and on its evolution, the preceding paragraphs showed that none of the planned solutions up to the present - forestry plantations, improved stoves, substitution fuels - would be able to provide in and of itself the solution to the problems of degradation of woodfuel resource 2.28 The choice with regrd to plantin&g man*ging edsting foest resources, conditions related to its exploitation, and Als to reducing specific consUp by the various wood and dwacoal conumers and substitution with other fuels all appear comple and interdependent. It thus i necessar to Incrporate al these items into a coordinted action progam, which fuly takes into account the various aspects, energy and forestry related, economic and sociaL Thus an energy strategy for the sector must be defined and implemented which smultaneously addresses the following needs (a) to decrease the pressure on Haitian forestry resources, and more generally on the environment, aused by woodfuel consmption (b) to supply at least cost stoves and energy fuel adapted to the needs and finanial means of different categories of consumers; (c) to improve the efficiency of production, distribution and utilization of household fus; (d) to presenre the interest of rurl woodfuleb produces, eseiaBy those for whom charcoal production is a main economic activity, by allowing them to ezercse their activit in a more efficient framework or by asuring their reconversion. 229 Analysi of the situation all several guiding priciples of the stategy to be defined, which are the following: (a) take into account the dcaraci of household energy demand; the strategy should be based on analysis of the actual market for fuds and stoves, take into account consumer preferences and financial means in order to define solutions which are diversified and adapted to their needs.  32- (b) define a pricing pohcy whih reflects the economic costs of supply and distribution of household fuels, in a least cost approach for the country; (c) gradually restore the financial balance of the sector through eliminadng exdsting distortions. Tbrough the means of preferential exange rates, the Haitian economy wbsidizes petroleum fuels and, since recently, portable gas stoves. In addition, under current pricing conditions for woodfuels, the smallholder owners of wood covered areas subsidize the price of charcoal to the benefit of the urban consumer by selling wood at a price generally lower than replacement costs: this is possible at the cost of forestry resource-mining production techniques, which induce degradation of the environment and of the rural economy in the medium term. (d) improve fiscal revenues from household fuels; this would allow at least partial self finaneing of investments in the energy sector. In fact, neither the current tax sucture, nor the existing methods and effciency to collect taxes are sufficient for the State to have resources to invest in improvements for enery supply and consumption. (e) focus interventions on a small number of essential themes: the importance of certain priority action themes, the urgency of an intervention at a large enough level and the difficulty of undertaking multiple interventions in the cufrent institutional context al incite to limitating the of strategy to some major components. (f) take advantage of all relevant organizations present in Haiti (NGOs, communal sections, institutions, associations) as co-implementors of the various strategy components.  *33 - IL THE STRATEGY 3.1 In countries like Haiti where woodfuels consumption is dominant, a household energy strategy generally should be centered on two principal axes: (a) interventions on woodfuel supply-side: improved production and marketing of wood and especialy charcoal; (b) interventions on fuel demand-side, mainly through (i) substituting wood and charcoal with other fuels, especially petroleum products; (ii) improving conditions for using wood and charcoal, in particular by the distribution of more efficient energy products. A. intr n on Woodfuel u 3.2 In order to define the options for supply-side intervention, the present economic situation of charcoal exploitation must be analyzed. Tlis anabsis demonstrates the need for complementarity between demand-based and dharcoal supply-based strategies, and allows the resources and actions needed to modernize the charcoal sector to be defined. Current_Eonomic Situation of Charcoal EDotto 3.3 The evolution of real charcoal prices can be examined using a relatively long timeseries (dating from 1953 to the present) whih was compiled by the IHSI as part of its price index series. Use of this type of statisical abstract, where charcoal is included in a basket of commodities, generaly does not yield themost reliable gauge, because measures of charcoal price relative to weight are not accurate and are generally made with a small sample. Nonetheless, this abstact undoubtedly is one of the better sets of timeseries data available in Haiti. hree principal periods of evolution are evident from the data (see Figure 3.1): (a) during the first period, up to 1967 (the year when cyclone Cl6o struck), charcoal prices decreased in real tems, (b) during the second period, from 1967 to 1985, charcoal prices rose rapidly, (c) during the third period, from 1987 to the present, charcoal prices grew more slowly.  .34 - 3.4 Svera eplanations are possile, for these successive trends. The relative drop in real charcoal prices from 1950 to 1967 undoubtedly is due to the relative abundance of wood during that time. In addion, the 1960swere puntae by a m n of cycloes (Flora in 1963, Inbs in 1966, and the strgaet Cl6o In 1967) during ih substantial quantities of wood were felled, thus providing 'cost-free materials to produce charcoal inepenively for the urban population. Fig=-.t1: CheroosL Prie - 1953 * 1990 TreOd 25 20 3.5 T'he opposite tendenc of prices after 1967 can be explained by the evolution of the resource supply: relatively abJundant befor 1967, wood became increasing scarc in subseqent years Thec development of major and secondaiy roadways in the north and south at the beginning of the 1970s opeed Hatis rural areas to the modern economy, temporarily producing a oownward effect on prices However other facor once agdai contriuted upward pressure,on prices, including inceased tranporttion coss and the progressively spreading practic by local authorities of Ilegally collecting part of the costs of charcoal transport and marketing. 3.6 The recet slowdown in the growth of real charcoal Prices can be explained by. (a) at the. demand level ieased competitio from new fuel. he threshold at which petroleum afels become competitive with wood-based fuels has been surpassed: in tems of useful ceerv, prpane s and kerosene have become less aPeeive than charcoal (see para 3.42 to 3.45).  0 3S - (b) veral factors on the supply side which are the more likly deteants In the downward pressu on charcoal price hyq incude: (1) the retiv dease In rural Icomes, and thus the decrase in the opportunity cost of pessantlabor, (I) and in partkula, from 1986 onwards, the susension of loca controls on fores aploon, which prompted the spread of :charoal production throughoui the country. 3.7 None of the price evolutions produced benefits for charcoal producers; in fact the contrary is true. The produceres dsare of the rtal carcod prie has decrased coniuous for the past ten years (see Figure 32). Charol price i pdmarly have been due to inreasing intermediaiy costs; the producer price has rmained ontant. Ibis means that the produces acu income has dropped, If the newly-recogized fact of inasin g monetazy value of the wood used for charcoal prodtion Is taken Into accomt. Elta 3s2 Prodwwr's Swa of CharAoml Price 1979 - 1990 Tred 3. 960 0.ad.13 be be RetaLl price Iso Producer's share 10 acM ..Uw v vo _ Producer price 5 1Gm 0, a SonI sa C) 64 .6 5 5 7 * a- e a 3i8 in fact, the time when charod production was based primarily on aoess to a ree rource is essnt past Neady 60% of caol produer opeating on land other dt their own buy the wood used for ion, ther thro diect purchamses or, more frequently due to insficiet cash flow, by spttig the charcool sales pdce with the wood suppl In the case whe wood is purhased directly, the producer pays an erage of 3 to 3.5 Gourdes for the quantity needed to produce one bag of charcoal, equvalent to 1$ 2/m' or HS3/ton of wood (suming a  -36. weighted carbonization efficiency of V0%). When the charcoal price is shared, the split most often is 40% to 50% for the wood supplier (about 4 Gourdes per bag) and 50% to 60% for the charcoal producer: in this case, the equivalent price of wood for the producer is nearly H$ 3/rn or H$ 4/ton. 3.9 Thus the 1990 price of wood used for carbonization is less than half the cost of planting wood, estimated by FAO at US$5.4/m3 or US$7.7/ton for the larger plantations. Another comparison to be made is with the present costs of planting trees under agroforestry projects: each surviving tree costs about US$1.5 and annually yields - independent of the environmental benefits to the rural economy of planting trees - about 3 US cents (based on productivity of 10 kg/year, calculated in terms of current prices of wood used in carbonization). Finaly, these figures can be compared with the estimated environmental costs of deforestation - namely loss of sediments, reduced soil productivity, and erosion (see Annex XII) - which in Haiti are about US$ 23/ton of wood (Hosier and Bernstein, 1989). 3.10 The fact that a monetary value has been attached to wood used in charcoal exploitation indicates an ongoing transition in the perception of deforestation in Hat Whereas deforestation previously was noted only by specialists, its impacts now directly affect the wood consuming population. However, the cost of wood is yet not enough to induce rational exploitation of wood resources or even the development of a specific economic rent associated with wood planting. Replanting efforts must be pursued, but in strictly financial terms, they are not profitable at present. Based on the FAO cost estimates, planting wood could be profitable if producer costs more than double to 18 Gourdes/m3 (split 6 Gourdes for the charcoal producer and 12 Gourdes to the wood supplier). lhis would imply a 20% increase in urban retail prices, which is acceptable if less expensive fuel alternatives also are available to consumers. Petroleum fuels already provide less costly alternatives for households able to invest at least $30 or $40 for a portable cookstove. In addition, the recommended improved stoves program would permit consumers to save up to 20% on fuel expenditures for a modest initial investment of only $4 to $5. 3.11 The evolution of charcoal prices presents a basic problem with regard to which stratey should be adopted. The strategies to be considered can be described in terms of two extremely different options, both of which carny strong potential environmental risks: (a) The first option is to retain the current policy: let charcoal consumption increase at its natural pace. Resource shortages gradualy will induce the cost of wood to rise until the associated economic rent develops. When the fact that Hait's environmental deterioration is due to multiple causes - not just wood offtake for dharcoal production - is taken into account, planting and cultivating trees for energy use becomes then another means of envionmental preservation, just as with the cultivation of coffee and banana plantations in other countries. However, will the environmental damage caused in the meantime not be irreversible?  a 37 - (b) Another option is to forego supply-side Intervention and focus strategy on substitution and energy conservation. The induced decrease in cacoal consumption would diminish the pressure of wood offtake on forest resources. However the policy of intervening only on the demand-side also carries certain risks: a downward pressure will develop on charcoal prices, and subsequently on the cost of wood. This wil tend to perpetuate the 'mining character of urrent forestry practices, in which the resource is exploited without concern for its renewability. 3.12 These considerations emphasizes the ned for a global strategy of intervention Incorporating both energy and forestry aspects: (a) demand-side interventions to induce energy substitudon and energy conservation; (b) supply-side interventions to establish conditions which wmi render wood and charcoal prices more ecpensive, namely through rational exploitation of wood resources due to a modernized charcoal sector. ModMrnM nthe Cbarcoal Sector 3.13 Intervendons on the conditions of charcoal marketing and production must be a princpal element of the strategy. This will convert exploitation of forestry resources from its present 'ming model into a cohesive system for rationally managing a renewable national resource. Two important factors which would hinder development of a tidonalized system must be taken into account: (a) Instiutional weaknesses in the sviso of the sector, and the foreseeable dulties this superviing entity wSil have when attempting to implement required improvements; (b) the lack of any systematic study of the charcoal sector, its role in the peasant economy, and its impacts on the environment in each exploitation zone. 3.14 These constraints shape and color to a large etent the contents and organation of supply-side intervention. As a result, it is recommended that a specialized, autonomously managed and financed unit tempoariny be establshed within the MARNDR Tbis unit would be responsible for analzing the different aspects of charcoal exploitation (including the conditions for enforcig laws and fiscal policy) and for coordinating and Wpvising the projects and programs to modemize the uharcoal sector.  .38. 3.15 The unit's operaions wil be developed In coordination with and as a complement to actions undertaken within the framework of the new Foresty and Envionmental Protection Project