Some of the Determinants of the Consumption of firewood energy in developing countries: The case of Rural Botswana by J.W. Arntzen and D.L. Kgathi L Introduction For the majority of the world population, firewood is the main energy source. Fritz estimates that around 60% of the world's population depends mamly. on firewood to satisfy its energy requirements. I Moss and Morgan quote FAO figures (J 974) stating that firewood contributes upto 58.4% of the energy consumptIOn in Africa; for developing countries as a group this figure is 22.4% compared with only 5-7% for the whole world.2 In Botswana firewood consumption amounts to slightly over half of the total energy consumption. j In many developing countries firewood IS depleting rapidly. Due to the rapid population growth, the demand for wood biomass often exceeds ItS production, resulting in a decrease in the amount of wood left. Furthermore, the arable area expands stripping large tracts of land of woody vegetation. On the other' hand, widespread bush-encroachment due to overgrazing increases the amount of wood In the grazing areas. Fritz emphasises that firewood shortage is a locational problem: it occurs mainly around the population concentrations as a result of Incr:ased demand and reduced supply (due to clearing for arable or reSidential purposes). To counteract the latter, reafforestation programmes have been initiated In many countries. 5 People do not remain passive under conditions of firewood scarcity but adapt themselves in various ways, depending on the socio-economic position of individual households. Often these adaptations have a direct impact on other activities. For example, peopl~ may have less time left for agricultural production because of the Increased time needed to collect firewood. This may have an adverse effect on crop production since research has revealed that labour shortage is a constraint on crop productlon.6 It implies that firewood problems need to be studied in the wider context of rural development. An additional reason to study firewood in this wider context IS that the production of wood competes with some other forms of land use, m partl~ular crop production. Only fairly recently the concept of agro-forestry, consldermg trees as an agricultural crop and not as a separate form of land use, has gamed more support.? Given the. increased interest in firewood problems, strikingly little data exists so far; certalnl\ m Botswana. So far the available data refers mainly to case studies. of villages. Indeed very httle IS known about the complexity of factors determmmg demand, supply and ultimately the actual consumption of firewood in rural areas. We see an urgent. need for a more systematic approach to solving rural energy problems, 51nce the. malor problems at village level have been studied. Section three is a first =: attempt In thiS direction. A similar attempt towards a more systematic approach has made for Indonesia.9.. It should be made clear from the start that lack of data not allow for .quantlf1<:atlon as yet. It would be useful, therefore, to identify e~ntlaJ data and. In combination with data mentioned in the subsequent SeCtions, to poant out ppllI) data ~ hence, veas for future research. Section three is followed by two categories of case studies, which examine the country's present firewood situation in rural villages (section 4). The second one discusses rural-urban relations in firewood demand and supply (Sec.5). Both of these sections highlight the spatial aspects of firewood scarcity: lack of wood close to human settlements. Although at present the firewood shortage may be less serious in Botswana than in some other countries, continuing rapid population growth and urbanisation will aggravate the firewood problem in the near future. Present research on firewood issues is needed in order to prepare for this. A classification of energy sources, in particular the ones available in rural areas, precedes the sections discussed above. Different characteristics of energy sources are summarised. They show that firewood cannot be studied in isolation from other energy sources. z. Energy Sources Botswana has a number of energy sources within the country such as the sun, wind, coal, wood, cowdung and crop residues. In addition, oil and gas are used. The so-called primary energy sources can be used to generate a secondary energy source, i.e. electricity. Generally, two classifications of energy sources are widely used based on different criteria. The first one uses renewability as a criterion, hence distinguishing non-renewable from renewable energy sources. Renewable sources can be considered as ~ to the society from the environment (sun, wind, wood and agricultural residues). Present use of these energy sources does not affect future use at all (in case of sun and wind) or affects it only when their use exceeds the regrowth rate (wood, agricultural waste). In contrast, the present consumption of non-renewable energy sources (oil, gas, coal and material for nuclear power) directly affects opportunities for future use as it reduces the stock available. This classification throws some light on the possibilities to sustain economic development on the long run, be it national, regional or global. The second classification takes the need for cash as the base for distinction; commercial versus non-commercial energy sources. Commercial energy sources (oil, gas, coal and nuclear power) generally have a market price. In contrast, non-commercial sources are usually not traded and hence have no market value. They have an economic value. UsuaJJy, firewood, dung and crop residues are considered as non-commercial sources. In general, firewood is the most important accounting for 83% of the non-commercial energy consumption, followed by cowdung (J 5%) and crop reSIdues (2%).10 However this distinction is not clear-cut. Wood has become, locally, an object of trade, direc'tly or indirectly (via the production of charcoaJ) and hence then becomes a commercial energy source. Furthermore, energy sources like the sun and wmd are difficult to classify. Although the sources as such are non-commercIal, theIr utilisation requires the purchase of mostly commercial equipment hke wmdmills and sun collectors. The last classification is important particularly in rural areas, with mostly very low incomes, as it indicates the degree of monetarisation of the energy consumption, i.e. the extent to which households have to pay cash for energy. On a national level, the relevancy is that commercial sources, in the absence of national reserves, for example oil in Botswana, have to. be imported. The Posslblhties to import depend on the country's foreign exchange sItuatIon and external factors (e.g. political events). Non-commercial energy sources are usually produced/collected within the country. They do not require foreign exchange and are not dependant on external factors. However, their local production or collectIon may have a negative impact, such as deforestation, on the environment and other 2S activities such as extraction of nutrients in case of use of agricultural waste. In gene~, commercial energy sources are more frequently used in urban areas than in rural areas. For example, Pachauri reports that non-commercial energy sources contribute uf: to 51% of the urban energy consumption in India compared to 80~ m the rural areas. 1 Finally, it is useful to distinguish between domestic and mdustrIal use. The latter comprises the manufacturing sector, services and trade •. Whereas the former category is closely related to population development and the SOClo-economlC development of households, industrial use is more linked to national economic development, and at a rural level, to opportunities to start rural industries. 3. Some of the Detenninants of Rural Firewood Consumption a. Introduction Firewood consumption cannot be studied in isolation from other forms .of energy.Depending on the degree of access to other sources of energy and the specific conditions of households, people decide on the type(s) and amount of energy used. Furthermore, the rural sector cannot be studied in isolation from the urban sector. Firstly, there are energy- flows between these two sectors, for example, trade in wood, oil, and gas. Secondly, urbanisation, a widespread phenomenon in developing countries, influences both the rural energy consumption and the energy situation in and around towns. Both. aspects are incorporated in diagram 1. This diagram atteml?ts to d~scrIbe the most relevant factors explaining the ultimate consumption of f~rewood. Obviously, both the demand and the supply side are distinguished. Discrepancies between the two result in commercialisation of a so far .non-commercial energy source or may induce price fluctuations of commerCial sources. Finally, households decide on the actual energy ~onsum~tion (types and amount). This may involve changes in the mterac~lon between demand and supply. Government may influence the rural flrew~ c~nsumption. This can be done intentionaJly as part of forestry pollcy, price poJicy for energy sources or even unintentionaJly. In the latter Case, changes in the rural firewood consumption are indirect effect~ of government policies or actions (for instance, better access to paraffm . and ~as by i.mprovement of the road system). The demand and supply .Slde w1l1 be diSCUSsed in section 3.b and 3.c. Subsequently, the adaptation strategies of people are reviewed in 3.d. In section 3.e the role of government will be dealt with briefly. b. Determinants of firewood demand i. Population Generally, population growth is associated with an increase in the energy . demand of households. In most developing countries, ~=~~ ~atlon growth rates are considerably high. Africa's population Increased by an average of 2.7% in the period 1970-1979.12 household is the USUal consumption unit of energy; hence more accurate to link energy demand for domestic pw"pCliSes to the number of households rather than to the overall 1 1 "0 U c o c ell'" ... 0 o 0 III ell .9';: ~;: _U :::! ~ "OU bO 1!o L. o :J 0"0 L."O '" :J ~ 0 ::J 0 ... c ~ -::O~ c 0.'" o o ....ell L. ell L. U '" - 1 1 ... o'" =0 ell o "0 "0 o~ ;: '" *c 0:>- 0- I I ...£ ~o. ell 0. E :::! .-u.. '" L. :::! L. '" ell ell U .s::L. III ... :J C o 0 rtl.~ o III U >. ->. "0 o ~~ ~ o :J ell I ~ ell 0'" '"C ell 00. L. ;;:; ~ E ell :::! ....o -y .~ ~ .... 0 ...c III , u , '" c ~... E L. '"ell III "0 C ... :>- DO E '" ... '" ~ L. III U'" III Ql Ql L. ~ '" ... 0 C U ell '" "'0. L. ell C L. ::J E o ." '" .D L. ::J 41 0 ... .s::'" o T c-=. o ...c .~ Q) ell ell bO E a. '"N ].'\ii o '" III :::! '3 r:r 0. ell ~ ~ .... O~ ell 0 1Il ... ~...L. .~ III Ql III E't E0 ::J "0 0'" U ::J c 0 "0 5 ,~.s:: population. Wiersum shows that energy-consumption per capita declines with increasing household size.13 One should realise however that members of one household may temporarily or permanently be at different places and have separate energy needs. Clearly, a careful definition of the household or consumption unit is required. For our purpose, population development has to be considered in relation to available land. Population densities (persons/ha) vary widely in developing countries. In general high popula1:.londensity results in high pressure on land and its related resources, including wood. In practice, people are not evenly distributed but are settled In other patterns (see section 3.b.vi). ii. UsaKe In rural areas, energy is generally used for domestic purposes such as cooking, heating and lighting. It is also used for rural ind!Jstnal activities like beer brewing and brick moulding. According to the National Academy of Science "roughly 80% of the fuelwood consumed in developing countries is used for domestic purposes",14 The type of use influences the choice of energy source. For example, paraffin IS more important for lighting than for cooking. For cooking, firewood IS still most popular. Kuiper and Mellink attribute this to the easy handling of firewood and the fact that no special provisions are required. I 5 More so, open fires allow for more pots at the same time, which is more difficult with stoves. iii. Equipment The most important equipment is the stove for cooking. Another, obvious equipment is the lamp used for lighting. For cooking, open fires are still commonly used in rural areas. These fires have a low energy-efficiency rate, estimated to range from 6-10%.16 This means that only a small fraction of the caloric value of wood is converted into erer~ for ax:ki~,17 The use of stoves may increase the efficiency of wood possibly to about 20-30%. However, it has been pointed out at the same time that "claims about efficient stove designs have seldom been Substantiated by unequivocal tests,,18 Stoves are at present not frequently used in rural areas. They require some adaptation in cooking methods since they are usually suited for, at most, two. pots. F~rthermore, the costs of buying them may be an obstacle. This constramt becomes less serious for stoves constructed fr0l'!" local materials whether natural or waste. People will be more Inclined to use stoves if wood becomes scarcer with accompanying greater eff~ts to coll~t it. Savings resulting from the use of stoves ~or.ne qwte attractive. Of course, the income position of ~ndlvldual house~olds plays a role in the acquisition of stoves. Finally, It has to be reahsed that a change in habits of people usually takes time. ~neased efficie!lCy does !1Ot ~~nd on equipment only, as can be For from the ~Ic:le range 1~ effiCiency rates of open fires and stoves. example, 51mply soaking beans before cooking reduces the cooking time substantially.l9 Protection of open fire against wind also Increases the efficiency. Clearly, by proper management of the fIre and cookmg, efficiency rates can improve considerably. iv. Access to Eneq~y Sources Access to energy sources has two different aspects: physical access, i.e. the distribution of energy sources in rural areas. 'economic' access, i.e. the prices of the energy sources in the rural areas. Even if energy sources are phsyically available, they may be too expensive for widespread utilisation. This, of course, depends on the income situation of individual households. Wood is usually collected locally. Accessibility decreases if wood is becoming scarce. Access to crop residues depends on the arable activities of households; cowdung can usually be collected freely. Phy:;ical access to other commercial energy sources is generally more restricted. The distribution of these sources depends, among others, on the size and quality of the road system and focuses on the bigger villages. dung and crop residues are normally free of charge. In contrast with commercial energy sources, the price of firewood increases near settlements and, supply factors being equal, i highest in the towns. r Commercial energy-sources, however, fetch higher prices with increased distances to the distribution points. usually the bigger villages or towns. v. Income Income level is an important factor; he acquisiton of cooking equipment (3.b.iiD as well as in the use commercial energy sources (3.b.iv). In developing countries, rural incomes are often very low. Formal job opportunities tend to be concentrated in urban areas or mining and sometimes agricultural enclaves in rural areas. Agriculture, the main source of rural income in many developing countries, usually does not provide people with adequate incomes. Apart from the average income level, the distribution of income has to be considered. Evidence suggests that various socio-economic strata can be distingUIshed at a rural level •.l:C1 Difference in SOCIO- economic positil\ns between households will usually be reflected In different energy consumption patterns. vi. Spatial distribution of 'ewood demand The spatial distribution of firewood demand reflects human settlement patterns. This relates to both the distinction between urban and rural areas and the settlement pattern within rural areas. Owing to difference in income level, access to energy sources, e.g. the energy demand in urban areas differs significantly from that in rural areas. We saw already that commercial sources are more frequently used In urban areas, both for domestic and industrial purposes. Obviously, large differences exist also within the urban sector to the above-mentioned factors. Migration to urban areas may reduce rural energy demand. It implies, however, additional demand in rural areas around the towns. Within the rural areas permanent and seasonal settlement patterns need to be distinguished. Und~r the assumption of equal supply, firewood problems tend to increase with the size of settlements. A rural population concentrated in a small number of villages will normally face more problems related to firewood than one which is more evenly spread over the rural areas. The latter IS more likely to be close to firewood, still disregarding inequalities in supply. For the same reason, seasonal movements may alleviate firewood problems. The movements from the village to the lands or grazing areas reduces the concentration of population in certain periods of the year. c. Determinants of firewood supply i. Wood from Natural Ve~etation Usually, natural vegetation provides the bulk of firewood consumed. The amount of wood available depends on the forms of actual "land use and natu~al factors (soil, climate), determining the potential wood production per hectare. Production of wood to meet the demand of firewood is at best considered as a secondary form of land use, if not ~eglected at all. In fact, agriculture is the dominant land use activity In rural areas of most of the developing countries. "~to firewood also depends on land tenure. Hence both land use and land tenure have to be considered. Land Use For crop production land is stripped of most of its trees and therefore this form of land use is incompatible with wood production (unless wood is regarded as a crop, i.e. agroforestry). Population growth and the extensive nature of agriculture (e.g. shlftm~ cultivation) in many African countries, results in an expansion of the arable areas, particularly around the villages. Con~equently, wood is produced further away from villages. In grazIng. areas, trees and hence wood production, are part of the vegetation. Overgrazing usually causes bush-encroachment, I.e. an mcr~"ase i~ wood vegetation at the expense of grasses. Some of the InvadIng species' are highly appreciated as firewood.21 In sor:ne countries like Botswana, no pure crop production areas eXist. Crop production is mixed with livestock production in so-ealled mlxe women used to 3S collect wood from the Combretum imberbe and Combretum apiculatum trees, but in 1981 people generally collected different species of firewood such as D. cinerea and A. erubescens because preferred ones were no longer easily available.36 Oki and Majaha-Jartby revealed that the points of firewood collection were as far as 5km in Ditshegwane, a small village in Kweneng District with a population of 821 and about 85 km north-west of Molepolole, the capItal of Kweneng District. On the other hand, in the larger village of Shoshong (pop.ulation = 4,600) in Central district, these points could be as far as 12 km.37 In fact, around most large villages, the cutting of live wood IS becoming common because dead wood is no longer easily available. For instance, Jelenic and Van Vegten observe)8 Around Molepolole the situation has become desperate enough to force people to chop live wood as energy source. Similar cause for live wood destruction is found around the major towns, but the wood shortage forces people to buy firewood from traders. As the above quotation notes, scarcity of firewood has led to commercialisation of firewood. In most villages, particularly, the large ones, ,commercialisation of firewood is becoming increasingly Important. Even In rela~ively smaller villages, firewood trade seems to be emergm~ In Gabane vIllage, a single axle donkey cart load cost about P6-10. Another aspect of fiPewood trade (rural-urban) will be discussed in the next section. 5. Rwal-Urban Relations: Botswana a. Patterns of energy consumption The e<:0~omies of most developing countries are compartmentalised into two dIStInct sectors, the so-called "traditional" rural sector and the "modern" urban sector. Patterns of the various aspects of the economies of these countnes, Including energy use, also reflect this dualistic structure. The urban sector consumes more commercial energy whereas the rural sector depends more on non-commercial sources. Unfortunately, data on the breakdown of energy consumption into domestic and industrial use is not aVailabl~. Table 4 shows that the per capita consumption of hnon-commer~IaIenergy in the rural sector of Botswana was about 2 times Igher than In the urban sector in 1982.1he table further shows that the per c~lta <:onsumption of commercial energy in the urban sector was about 49 =s higher than in the rural sector. In the urban sector, commercial , g>: accounts for about 88.9% of total energy used This form of energy ~maInlY consumed by the mineral sector. For i~stance, in 1980, the GWHngwato CO~~sions Limited (BCL) Copper Mine consumed about 278 ,~f electn4cOlty, "which accounts for about 70% of Botswana's elect nClty sales." In ddit' abo than 30% of the a Ion" ut 130,000 tones of coal, which is more total production In Botswana, were consumed by BCl,41 In Botswana's cap'tal ' ellpensi~ form: City, . Gaborone, like in other urban centres, the mosJ caI in l?aO. •~ energy IS electricity which cost about P 141;.00 per 10 umt. However the other tw.nd. ~ood cost about P 73.00 per the same , one should take mto consideration the fart that the TABLE 2 I Type of Settlement Non-Commercial Energy Commercial Energy Total I Urban Rural Total 290-330 560-680 520-620 utilisation efficiency of electrIcity 2430 50 410 2720-2760 610- 730 930-1030 (90%) is higher than that of wood (6-20%). The prices of electricity and wood per usable calorific value (J 07 k cal) were therefore about P1572 and P365-P1216 respectively In 1980. b. Eneq?;y flows (supply) The urban sector mainly depends on the rural sector for its supply of fIrewood whereas other forms of energy such as coal are distributed to both the rural and the urban sector. 1. Firewood Firewood is exported from the districts to the nearby urban centres where it IS mainly consumed by those who do not have access to other forms of energy. It is mainly needed for domestic purposes particularly by low-income households. The high income group also uses firewood but maInly for the purpose of heating. Many people In low income urban areas collect firewood either by bicycles or motor vehicles from the rural sector. Some purchase it from firewood traders who collect it from the rural sector by donkey carts or motor vehicles. Kgathi indicates that there were at least about 35 traders who operated between Kweneng district and Gaborone and they collected firewood on the average about 45 km from the consumption area in Gaborone.42 Most traders took about 12-I4 hours to travel from the harvesting points to the consumption area. The species preferred by consumers in Gaborone were those with long lasting embers such as Combretum apiculatum (mohudire) and Combretum imberbe (motswere). About 43 tons of firewood were exported by traders from Kweneng district to urban Gaborone per month. The figure is likely to rise in winter when wood has a higher demand due to cold weather. By-laws on the exportation of firewood out of the districts exist. For instance, Kgatleng district prohibits the exportation of firewood whereas Kweneng district imposes a tax on whoever removes firewood out of the district. A trader transporting firewood by a one axle donkey cart (carrying approximately about 350 kg of wood) is charged a tax of about P 1.00 whereas a trader with a two ton truck full of wood is charged a tax of about P2.00. The by-law also forbids people to cut Jive wood. However, there is clear evidence that people continue to cut live wood and also evade payIng firewood tax. It seems this by-law will only have an impact if its enforcement can be improved. 37 As the urban population increases, one would expect the demand for firewood to increase in the urban sector unless this resource IS considerably substituted by other forms of energy: During the intercensal decade of 1971 -81, the urban population in Botswana grew at an average of 10.7% as it can be seen in Table 3. TABLE 3 Urban Population and Growth Rates (1971-81) Town 1971 Census (I) 1981 Census(l) Growth Rates(2) Gaborone 17 713 59 657 12.9 Francistown 18613 31 065 5.3 Lobatse II 936 19 034 4.8 Selebi Phikwe 4 940 29 469 19.6 Orapa I 209 5 229 15.8 Jwaneng n/a 5 567 n/a Total 54 411 150 021 10.7 n/a Not avaIlable. Source I. 1981 Population and Housing Census: GUide to Villages and Towns of Botswana, Gaborone, Government Printer, 1983. 2. Calculated from the formula: Pn = Po (J +r)n where n Number of years (intercensal decade) Pn Population after n years Po Initial Population growth of population per annum New towns such .as Selibe-Phikwe, Gaborone and Orapa had very high growth rates dunng this period as it should be expected. Despite such an Increase In the urban population, very little has been done to ;~:abhsh Woodlots which can supply the urban population with wood fuel Since naturally produced wood seems to be scarce around urban centres. Woodlots established by the local government in ~waneng are a good example. These plantations do not depend on rain f~: m~ke use of sewage water. The Ministry of Agnculture has some est y plantatIOns In Gaborone and Lobatse but it no longer initiates any new progr ... this .. 4~mmes; Instead It SUpports organisations which engage In actiVity. Like In rural areas, most of the woodlots in the urban ~ector are those of Eucalyptus trees which are not suitable for use as IrewOOd as already mentioned. 38 II. Other Forms of Eneq~y: Coal and Electricity This section mainly discusses two government programmes which attempt to increase the use of electncity and coal in rural areas. The rural electrificatIOn and coal distribution programmes aim at providing the rural population with electriCity and coal since It depends almost entirely on wood'? A P260 millIOn central power statIOn 15 being constructed In order to supply some villages and towns WIth electricity. Small generators uSing domestIc coal may be used to supply remote areas With electriCity since It ml~ht be too expensIve to connect them to the central power system.1i The price of electricity in the rural sector will be the same as in the urban sector despite the fact that incomes are generally low 111 the rural sector. It IS, therefore, reasonable to conclude that thiS project will only benefit those who can afford to pay electriCity bills. The government IS also working out a strategy of marketing coal from Morupule, and other areas to the rural and urban population. It IS estimated that 111 these areas the mineable coal reserves known, amount to about 3,600 millIOn tonnes.45 The Morupule mme IS, however, the only coal mll1e which has started productIOn. Morupule coal IS relatively cheaper compared to Imported coal as well as other forms of energy such as fIrewood, gas and electnclty. For Il1stancp, Il1 1982, the Mll1lstry of Mmeral Resources and Water Affair, estimated that the price of Morupule coal 1t1 Gaborone wa, about 1'35.00 per J 07 k cal wherea, that of firewood, kero,cne ilnd electnclty were about P73.00, P588.00 and r l,It 15.00 respectively, per the same unIt of calonflc value.46 A major dJsadvantage of coal" that ItS users need cookll1g stoves which may be an obstacle lor many households. Moreover, the use of coal IS as',oClated With mor£' serlOU, air pollutIOn problems. Co Som~acb of the RuraJ-Urban Energy Flows The sale of firewood prOVIdes those who are unempJoyed or under-employed WJth Il1come. In some African and ASian countries Jt IS the main source of Income for the landless people.li7 In Botswana, thiS trade mall1Jy supplements agricultural Il1comes s1l1ce most rural households have access to both arable and graz1I1g land. Firewood trade between urban Gaborone and rural Kweneng prOVIded self-employment to about 35 people Jl1 early 1983. These people generated additional employment to other people who aSSIsted them Il1 harvestll1g and transporting firewood to the market. On the average, the traders earned at least P77.00 per month.li8 In low mcome areas of Gaborone such as 'White City' and Bontleng, expenditure on thiS fuel amounted to about P 10.00 to P 12.00 per month. ThiS Impl1es that thiS fuel accounted for about 10% to 12% of the budget of the lowest paid industrial class workers who earn about P4.77 per day.49 If firewood trade IS not properly monitored, it can lead to deforestation which can make the soil prone to wind and water. This may lead to the depletion of the sod fertilay whICh may lower agricultural productiVity. In the long run, rural underdevelopment may become even more seriouS If no attempt 15 made to 39 address this problem. In Botswana, deforestation is a problem particularly around urban centres and major villages where firewood is depleting very rapidly and most land is cleared of vegetation for crop production. In these areas, as already mentioned in section 4, preferred species of firewood are no longer easily available. About 80% of women interviewed in Ga-Kutwe lands area, in Kweneng District, revealed that firewood trade contributed to the depletion of the Combretum imberbe (motswere) trees. The few of them which were available about 10 years ago had drastically reduced In number. The elderly women mentioned that the growth of Gaborone had led to an increase in the demand for firewood in the surrounding distrICts such as Kweneng and KgatJeng. The increaSing scarcity of firewood in this town led to its commercialisation. The distribution and supply of coal and electrcity to the rural sector may reduce the consumption of firewood energy in the long run. This will, however,depend on many factors such as the coal marketing system and the levels of subsidies on these forms of energy. Experience of other developing countries indicates that only a few rural people benefit from rural electrification programmes. For example in 1975, it was estimated that even though 25% of the villages had electricity in India "the proportion of the villagers using electricity was low, the extent of use usually slowing greatly after 20% of the population becomes users,,50 As far as coal is concerned, one wonders as to whether it will really reach the remote areas considering the problems associated with its distribution. However, if a substantial number of people were to resort to using coal or electricity (or both) the demand for firewood would decrease, hence alleviating the deforestation process. 6. Concluding Remarks I~ most developing countries, mcluding Botswana, shortage of firewood has become a p oblem. The rapid population growth means not only an increased demand for ~rewood but also an additional demand for wood (e.g. construction of huts, fencing). urthermore, It contributes to the expansIOn of the arable area implYing a clearing of ~~e vegetation and hence a reduction of the supply of fire~ood. As a result," the ma~d for firewood may exceed Its supply. Shortage of firewood has been described as a OCatlonal problem, i.e. firewood is depleting around the human settlements. In i~:~:;r, In countries with a low population density like Botswana, shortage of mamly occurs around the towns and large villages In their effort to Overcome firewood shorta I ". " h f th h " "ges, peop e may resort to burnmg down of living trees, ence e~~).er b~~::enmg their wood supply in the future and the quality of the soli (erosion, food" adaptation strategies mclude travellmg longer distances to collect Irew ,usmg less preferred" "" d f substitut ed" f" speCIes, purchasmg wood, mcreasing the deman or cooki el~ssr ucmg Irew()()(/demand by cooking with more efficient stoves or by strat;::es are fr~~~t1y. Case studies In Botswana indicate that most of these meals or re r ~ t y . USed. Howeve[, there is no evidence that people cook less appears to: 0 usmg stoves as a result of firewood shortage. The situation The situation in~~: serious m towns (e.g. Gaborone) and large villages (e.g. Shoshong). ns IS aggravated by the rapid urbanisation. The analysis of determi t f developing countries, :::": 0 demand and. supply of firewood shows how, many characteristics, suffer Ir •. result of their demographic and socio-economic diagram 1). The fact thaor:' f~ba1anc~ betwee!l demand and supply of firewood (see Irt~wood IS the malO source of energy in most developing countries, underlines the fact that shortage of wood is not only a concern of the direct users but also a concern of government as well. We have already indicated how firewood users are affected by its shortage and how they might adapt themselves. Adaptations often take place at the expense of other activities, particularly in the case of low income households. For example, cash spent on energy cannot be used for food etc. Increased utilisation of household labour may affect other productive activities. Clearly, rural energy problems, particularly firewood, cannot be Isolated from overall rural development. Adaptation strategies may also aim to supplement the wood production of the natural vegetation. In Botswana, natural wood production is low due to semi-arid conditions. However, wood production tends to be higher in overgrazed areas due to bush encroachment. Woodlots and planting of trees around the villages have up to now contributed only very little to firewood supply. Woodlots are to a large extent meant to supply wood for construction and fencing in Botswana. This is illustrated bl the selection of Eucalyptus species, which are not highly appreciated as firewood. 5 Although the shortage of firewood in Botswana has not yet reached the level of many other countries, the present signs of scarcity suffice the launching of a comprehensive rural energy policy, comprising also programmes of substitutes for firewood. Special attentIOn should be paid to the areas with the most serious shortages, i.e. towns and viJJages. Firstly, firewood collection should be recognised as a separate form of land use. This would imply that changes and policies affecting land use should be evaluated taking into consideration their impact on firewood production. If the effects of the policy are not desirable, action should be taken to counteract this. Firewood collection should be one of the facts determining the management of land (others are, for example, crop and livestock production and wildlife). There is need for consideration as to whether areas around villages should be assigned an exclusive use by the villagers. This traditionally existing situation is threatened by the emerging shortages and related phenomenon like firewood trade. Traders should collect wood far away from villages so as to avoid interference WIth firewood coJJectlOn by villages. This is pressing, particularly around towns. Trade IS one way of overcoming shortfalls of "naturally" produced wood. Another option is the planting of woodlots. Second!}', access to substitutes for firewood should. be improved, partIcularly around towns and large villages. This implies that the distributIOn and Pricing system of the substitutes should be worked out in such a way that many people would have access to them. At present these forms of energy are beyond the reach of most households since only a limited number benefit from them because they cannot afford their prices. Future utilisation of these forms of energy will also depend on the development of the firewood price and the average income level of rural households. Thirdly, continued efforts to raise the energy utilisation effiCIency of fIreWOOd should be supported. Special emphasis should be given to the reductIOn of Investment costs for people and the facilitation of the acceptance of energy-saving equipment. It is expected that people will be more inclined to use stoves If the firewood shortage continues. This paper has shown that there is a shortage of data on various aspects of firewood corsumption. Therefore, more research on rural and urban firewood consumptIOn (In m and kg) needs to be carried out. Furthermore, a breakdown has to be made between domestic and industrial use. Trade in firewood, particularly around towns, should be monitored so as to be able to evaluate the impact of the eXisting regulations. Various forms of firewood substitutes should be studied so as to assESS a a realistic contribution of these sources to the rural energy supply in. future. Particular attention should be paid to the determinants of the .selectIOn of substitutes for firewood (income level, access to substitutes, price differences e.g.). 41 At present, data seems to be very fragmentar5' and needs suppleroentation bearing in mind the factors behmd firewood consumption. 2 Footnotes I. Fritz, M. Future Energy Consumption of the Third World, Pergamon Press, Oxford, 1981. 2. Moss, R.P. and W.B. Morgan, Fuel and Rural Energy: Production and Supply in Humid Tropics, 1971, p.20. 3. Botswana Government, Energy Development in Botswana, Ministry of Mineral Resources and Water Affairs, Gaborone, 1982. 4. Fritz, M. Op.cit., p.94 5. Kriek, W. "Bois de Village Project in Upper Volta in Village" in BOS Newsletter 1983; Rey, C. "Village Afforestation in Tanzania" in BOS Newsletter; Skutch, M. "Village Afforestation in Tanzania" in BOS Newsletter. 6. GUlbrandsen, O. Agropastoral Production and Communal Land Use, Gaborone, Mmlstry of Agriculture, 1980. Opschoor, J.B. Environmental Resources Utilisation in Communal Botswana (NIR Working Paper No.38), Gaborone, NIR, 1981. 7. Gielen, H. Report on Agro-forestry Survey in Kenya, ICRAF / Agricultural State University, Netherlands, 1983. 8. White, R. Handbook for Village Woodlot Planning and Management in Botswana Hukuntsi, 1979; Jelenic, N.E. and Van Vegten, J.A. ~ Pain in the Neck: the Firewood Situation m South-West Kgatle~, (NIR Research Notes No5) Gaborone, NIR, 1981; Oki, J and J. Majaha-Jartby Botswana Village Energy Survey Report, Gaborone BRET, 1983. ' Arntzen, 1983. No.ll), -""-"-=:.....:::=~~~~~~~~~~~ J. Firewood Collection in Mosomane: Kgatleng (NIR Research Notes 9. ~~~:fr,~~~ellink, W. Supply, Demand and Consumption of Fuelwood for Netherlands, 198~ In Central Java. Agricultural State University, The 10. Moss, R.P. and W.B. Morgan, Op.cit., pp. 19-20. ll. Pachauri, R.K. "Third W Id P I.. No.9, 1983, p.21S. or OlCles: the urban-rural divide". in Energy Policy, 12. ~~t ~ P.1~lerated DevelOpment in sub-Saharan Africa, Washington, 13. Kuyper, J. and W. Mellink, Op.cit., p.21. 14. Quotation of Arnold, Firewood Crops, National Academy of Science, Washington, U.S.A., 1980, p.164. 15. Kuyper, J. and W. Mellink, Op.cit., p.23. 16. Moss, R.P. and W.B. Morgan, Op.cit., p.28. Krishna Prasad et.al. state that open fIres can reach an output efficiency of 30%. 17. Obviously, the 'lost' heat has some value for heating and lighting. 18. National Academy of SCIence, 1980, Firewood crops. p.164. For Instance, Krishna Prasad et. al., find that open fires may be more efficient than a metal stove (27.5%) and a mud stove (15.4%) (personal comment Tabe Tietema). 19. Ibid, p.I65. 20. See for a bnef discussion J.B. Opschoor, 1981.Environmental Resource Utilisation In Communal Botswana, NIR Working Paper No.38. 21. Van Vegten, J.A., Man-made Vegetation changes: (NIR Working Paper No.40) Gaborone, NIR, 1981. pp. 54-56. 22. World Bank, Accelerated Development in sub-Saharan Africa. p.116. 23. Botswana Government, Energy: Briefing Note, Gaborone. Other estimates are usually lower (Jelenic and van Vegten, 1981 and White: Woodlot Management Plan for Matsheng villages): Insufficient details are provided to make a reliable companson. 24. White, Richard, Op.cit., p.3. 25. Wood, et.al., Socio-Economic Context of Fuelwood In Rural Communities, AID, 1980, p~ 26. Geller, H. Prototype Metal and Mud-burning Cook Stoves for Botswana, Gaborone, BRE T, 1983, p.24. Also: Koester, H. and V. Kohnhorst, Possibilities of Coal Utllisation (Initial Study Project Survey) Grunds To ff technik GMBH, Essen, 1983, p.14. 27. Geller, H.S. Op.cit., p.24 28. Oki, J. and J. Majaha-Jartby, Op.cit., pp. 19 and 35. 29. White, Richard, Op.cit. 30. Information from Ministry of Agriculture (Botswana) 31. Jelenic and Van Vegten, A Pain in the Neck: The Firewood SituatIOn In South-Western Kgatleng Botswana NIR Research Notes No.5, Gaborone, 1981, pp. 35-45; Also: Kgathi, D.L. Firewood Trade between Rural Kwenen and Urban GaboroneL Botswana; A Socio-Economic Perspective NIR Working Paper N 0.46), Gaborone, 1984, p.43. 43 32. Arntzen, J. Op.cit., p.lO. 33. Kgathi, D.L. Op.cit., p.38. 34. Arntzen, J. Op.cit., p.12. 35. Ibid., p.14. 36. Jelenic, N.E. and van Vegten, Op.cit., p.13. 37. Oki, J. and J. Majaha-Jartby, Op.cit., p.41. 38. Jelenic, N.E. and J.A. van Vegten, Op.cit., p.2. 39. Kgathi, D.L. ~., 1984, p.33. 40. Botswana Government (Ministry of Mineral Resources &: Water Affairs). Op.cit., p.9. 41. Ibid., p.9 42. Kgathi, D.L. Op.cit., pp.24 and 34 43. Kweneng District, KwenenR District Development Plan II 1983 - 86, Molepolole, p.68. 44. Botswana Government (Ministry of Mineral Resources and Water Affairs), Op.cit, p.13. 45. Koeser, H.J. and V. Kohnhorst, Op.cit., p.5 46. Botswana Government (Ministry of Mineral Resources and Water Affairs) Op.cit., p.21. 47. Moss, R.P. and W.B. Morgan, Op.cit., p.38 48. Kgathi, D.L. Op.cit., p.35. 49. Kgathi, D.L. Op.cit., p.31. 50. Ceceski, E. et aI, Household Energy and the Poor in the Third World, Washington, D.C., Resources for the, 1980, p.56•. 51. Another important reason for selecting Eucalyptus was the assumed high growth rate. 52. A rural energy survey presently being conducted by the Ministry of Mineral Resources and Water Affairs may fill various gaps in data.