Gerald Foley
An assessment of the impact on domestic fuelwood demand of replacing fuelwood with conventional fuels
Gerald Foley is a well-known writer on forestry, environmental and energy topics and is affiliated with Earthscan and the International Institute for Environment and Development He is a co-author of two recent books on fuelwood: Stoves and trees, and Fuelwood: the energy crisis that won't go away. Both were published by Earthscan in 1984.
Conventional fuels such as kerosene and bottled gas (LPG) could be used to meet cooking and heating demands in developing countries. To what extent, however, it is feasible or desirable to substitute such products for wood-fuel - and in which countries - is a matter requiring close economic and social analysis, argues Gerald Foley.
Five years ago, in a similar article ("Substitutes for wood", Unasylva 32 [130]: 11-24), Foley and co-author Ariane van Buren, writing against the backdrop of the second world crisis in oil prices, placed the most emphasis upon coal as a potential wood-fuel substitute. The present article, written at a time of drastically reduced oil prices, offers a fresh perspective on the still-widening fuelwood crisis. It is based upon a new study commissioned by the FAO Forestry Department under the direction of M.R. de Montalembert, Coordinator of the Department's Wood Energy Programme.
· As traditional wood-fuel resources become depleted throughout the developing world, the issue of how to meet people's future fuel needs becomes more urgent. The impact of conventional fuel substitution on future wood-fuel demands is a vitally important part of this concern. It will heavily determine the balance that has to be struck between planting and managing trees for wood-fuel and investing in the supply and distribution of conventional fuels.
The questions raised are complex. The substitution of conventional fuels for wood-fuels is driven by both the push of scarcity and the pull of convenience, efficiency and modernity. The switch away from wood-fuel may be a voluntary or an enforced one; it may be seen by those who make it as an improvement in living conditions, or it may be seen as a new and grievous financial burden.
The task of analysing the relationship between future wood-fuel and conventional fuel demands involves investigation of a broad and interacting set of factors. Progress! to date has been very limited indeed, and a great deal more needs to be done before there is a secure foundation for policy-making in this area.
RURAL HOME IN SICHUAN conversion from wood to biogas (F. BOTTS/FAO)
The use of conventional fuels has been steadily growing in the developing world in recent decades. The oil-price rises of the 1970s slowed this trend but certainly did not bring it to a complete halt. The average annual rate of increase in the consumption of conventional fuels - primarily petroleum products in the low-income countries in the period 1974-81 was 3.7 percent (World Bank, 1984).
The most widely used conventional fuel at the household level is kerosene (also called paraffin, or paraffin oil). It is a product of the oil industry, obtained by distillation or refinery cracking of the heavier oil fractions. In the scale of refinery products if fits between gasoline and diesel oil. It has a calorific value about three times that of air-dry wood (43 MJ/kg compared with 15 MJ/kg).
Bottled gas, or liquid petroleum gas (LPG), is also widely used. It consists of butane or propane, or a mixture of both. These gases are found in association with natural gas or oil; they are also produced as by-products of oil refining, particularly cracking. They have similar properties as domestic fuels. At normal temperature they liquefy under pressure, and because they take up much less volume as liquids, they are invariably distributed under pressure in strong metal cointainers. They are usually sold by weight; a common size of LPG cylinder for family use is 20 kg, but smaller containers down to 0.5 kg are also used, especially for lighting. The calorific value of LPG is slightly higher than that of kerosene.
The lack of adequate transport and distribution systems completely precludes a significant shift to conventional fuels by rural consumers in the majority of the poorer developing countries at present.
Coal is used on a substantial scale in a number of countries, including India, the Republic of Korea and China. The costs of mining and distributing coal make it extremely expensive as a domestic fuel, unless it is already in use on a large scale for electric power generation or other industrial purposes. Its impact on domestic wood-fuel consumption is therefore likely to be limited, at least in the medium term, to a relatively small number of countries.
The developing world possesses vast resources of hydroelectricity which have the potential to transform the energy consumption patterns of many countries at some time in the future. But the transition from wood-fuels to such a widespread use of electricity is likely to be long and slow.
The most common domestic use of electricity at present is for lighting. Cooking by electricity is relatively rare in the developing world; and where it occurs, it is usually among the richer members of society. Where it is adopted for cooking, it tends to displace gas or kerosene rather than wood-fuels; its direct impact on wood-fuel demand in the reasonably near future is therefore likely to be relatively minor.
As far as resource availability is concerned, the constraints on a major shift by domestic consumers in the developing world to conventional fuels are surprisingly slight., This can be illustrated by calculating the effect if all the estimated 2 000 million wood-fuel consumers in the world were to shift to using 50 kg of kerosene each a year. This is a fairly generous assumption; it is, for example, slightly over twice the annual consumption of end-use energy for cooking for a family in the United Kingdom (Leach et al., 1979).
The total additional consumption of oil resulting from this switch would amount to about 100 million tonnes a year. This is just about 3.5 percent of global oil consumption in 1983 (BP, 1984). Clearly, there is no major resource constraint on increased use of kerosene for cooking at the level of world oil availability.
At the national level, also, the impact of a shift on the part of considerable numbers of domestic consumers to the use of conventional, fuels would be relatively limited in many countries. This is because of the quite small quantities of conventional fuels required to meet cooking needs. The major users of petroleum tend to be transport which often consumes more than 50 percent of the total - and industry. Typically, these account for 75-80 percent of total petroleum consumption, with business and commercial enterprises usually accounting for another 10-15 percent.
Appropriate calculations based on 1980 figures show that in Nicaragua, Sri Lanka and Kenya, for example, a shift to the use of kerosene fuels for cooking by all present wood-fuel consumers would result in an increase of less than 10 percent in those countries' total merchandise import bills. In other countries, particularly those at the lowest income levels, however, shifts of such magnitude are inconceivable. A detailed calculation is therefore necessary for each country before any judgement can be made on the feasibility of large scale switching to conventional fuels.
Table 1. Comparison between cooking fuels
|
Fuel |
Net calorific value |
Relative fuel requirement in energy units |
Relative fuel requirement by weight |
|
LPG |
45 |
1.0 |
1.0 |
|
Kerosene |
43 |
1.0 |
1.0 |
|
Charcoal |
30 |
2.0 |
3.0 |
|
Air-dry wood |
15 |
40 |
12.0 |
Table 2. Relative prices of different fuels in selected areas
|
Location(date) |
Relative prices by weight |
|||
|
Wood |
Charcoal |
Kerosene |
LPG |
|
|
Bamako, Mali |
1-1.7 |
5.3-6.0 |
- |
- |
|
Bangalore India |
1 |
2.2 |
5.5 |
7.8 |
|
New Delhi. India |
1 |
|
3.6 (subsidized) |
- |
|
Indonesia |
1 |
18.4 |
5.2 |
21.9 |
|
Niamey. Niger |
1 |
2.5 |
8.7 |
- |
|
Nigeria |
1-3.8 |
4.9 |
2.7-5.0 |
7.6 |
|
Senegal |
1 |
2.5 |
8.7 |
1.5-3.1 |
|
Uganda |
1 |
3.2 |
14.7 |
20.6 |
|
Yemen Arab Republic |
1-2 |
1.6 |
1.0 |
1.0 |
The fact that a major shift to kerosene is theoretically possible does not, however, imply that it is either feasible or desirable in any particular country. The lack of adequate transport and distribution systems completely precludes a significant shift to conventional fuels by rural consumers in the majority i of the poorer developing countries at present. But it is also clear that in certain countries, substantial numbers of wood-fuel users - those in the urban areas, for example - could well switch to the use of conventional fuels without causing a major disruption of national trade balances. This is a policy dimension that has often been ignored in analyses of the wood-fuel problem in recent years.
The relative costs of cooking with different fuels might be expected to have a bearing on the choice made between them by consumers. Exploration of whether this is so requires consideration not just of the market prices of fuels but of the efficiency with which energy is delivered to the cooking pot in conditions of normal use.
Although there has been a considerable amount of testing of the efficiency of wood and charcoal stoves and fires, there has been very little reported work on the comparative efficiencies of wood-fuels and conventional fuels. One of the few comprehensive sets of tests was carried out in India in 1959 (National Council of Applied Economic Research. 1959). Meals were cooked using wood, charcoal, kerosene and LPG. The total heat absorbed by the food and the pot, as well as that lost from them by radiation, was measured. The amount of heat accounted for in this way was then expressed as the "percentage heat utilized".
The results showed that for wood burned in brick stoves, the heat utilized was 14 percent; for charcoal used in a traditional metal stove it was 27.4 percent; for kerosene wick stoves it was 41.6 percent; and for kerosene pressure stoves and LPG stoves it was about 55 percent. Although these figures can be taken as only a very rough indication, they provide a ranking in efficiency which is broadly substantiated in the available literature on cooking stoves. As a working assumption, it can therefore be taken that cooking with wood in open fires requires twice as much energy as cooking with charcoal; and that cooking with charcoal requires twice as much energy as cooking with kerosene or LPG.
Table 1 shows how such assumptions translate into the quantities of different fuels needed to perform a given cooking task. The second column shows the effect of the differences of efficiency in use; the quantity of wood energy required is four times greater than that in the form of kerosene or LPG, and twice that in the form of charcoal. In the third column these figures are translated into the relative weights of fuels required, by taking into account the relative calorific values of each. This shows that for a particular cooking task, the weight of wood that must be purchased is 4 times that of charcoal, and 12 times that of kerosene or LPG. If the actual cost of cooking is a major determinant of the choice of fuel made by, consumers, then the relative market! prices, by weight, of the various fuels should broadly parallel these figures.
Table 2 provides a compilation of comparative price data for wood and conventional fuels expressed on an index basis for a number of countries. The price of wood is taken as the base-line in each case; in some countries, a range of wood-, fuel prices is given, with the lowest being taken as the base-line. The remaining columns show the price of the other fuels in relation to the lowest price for wood.
It can be seen that the price of charcoal is invariably greater than that of wood. There is also a certain tendency for the figures to cluster in the range of two to three times higher, as might be expected from Table 1. But the variations from this! are also extremely large.
As far as kerosene is concerned, the range of variation in price in relation to wood is even greater than that of charcoal. The price of LPG is generally higher than that of kerosene, but again there is no consistent pattern visible in the figures showing its relation to the price of wood.
Major policy decisions are being made in almost total ignorance of the factors determining people's choice of fuels.
The cost of kerosene, in comparison with that of wood, rarely reflects its advantage in energy efficiency.
Perhaps the most striking feature of the figures is that the cost of kerosene, in comparison with that of wood, rarely reflects its advantage in energy efficiency. If the relative efficiencies assumed in Table 1 are even approximately realistic, it appears that many people are paying considerably more for their cooking by using wood-fuels than if they were using kerosene.
There are numerous reasons why this might be so. Kerosene, for example, is not necessarily available for sale in many areas on a regular basis; consumers may therefore have no choice but to use wood-fuels. In some areas, wood and charcoal are important for the sake of the flavour they impart to certain traditional meals, and consumers may be willing to accept a cost penalty because of this. In other cases, it may be impossible to cook heavy stews on the flimsy appliances in which kerosene is usually burned.
The cost of the stove may also be a barrier to the use of conventional fuels. In Guatemala it has been found that very poor families often buy their wood daily in very small quantities. The US$12 required for the purchase of a kerosene pressure stove is beyond their means (Bogach, 1981).
In practice, however, very little is known at a detailed level about what determines the choice families make between different cooking fuels. But the apparently logical assumption that the cost of fuel delivered to the cooking pot has a major bearing is clearly inadequate. Other factors are obviously more influential for large numbers of people. This is an area in which a great deal more work needs to be carried out before any firm conclusions can be drawn.
|
Fuelwood and conventional, fuels in Mexico A detailed survey of two contrasting villages in Mexico made in 1982 revealed a shortage not only of domestic firewood but also of liquid petroleum gas, the main alternative fuel. Shortages were principally due to inadequate provisions for social forestry requirements on one hand and inadequate systems for the distribution of LPG in rural areas on the other. The village survey was subsequently placed in the context of the domestic fuel situation in central and southern Mexico. About 20 million people use firewood in Mexico. A transition was found from complete dependence upon firewood among the subsistence families who produce maize, through the combined use of both wood and LPG among the households having a mixed economy, to, finally, reliance upon gas only, on the part of wage-earning employees for whom the opportunity cost of obtaining firewood is too high. The largest single domestic use of firewood is the preparation of maize tortillas (which cannot be conveniently done using currently available gas stoves). Households that do not grow maize buy readymade tortillas and thus use less firewood than do the subsistence farmers. Scarcity of fuel is felt most, however, by the landless households that earn their livelihood from selling firewood and timber and can often only afford to burn twigs and crop residues for themselves. The cost of a gas stove (of the currently available models) and steel cylinders, and regular cash outlay for LPG refills is beyond the means of the poorer sector. Wood-only households were using an open hearth and consumed on average 16 kg of firewood (standardized to air-dry tropical pines) a head weekly (about 5.3 kg oil equivalent), 12 kg in cooking and 4 kg in water-heating, and more than 76 percent complained of having to burn green wood owing to the firewood shortage. An optimum amount would be nearer one tonne a head yearly, without energy-efficient wood stoves. Gas-only users consumed 1.4 kg of LPG (1 kg cooking and 0.4 kg water-heating). On the basis of the internal market price, firewood was found to be five times more expensive than LPG, apart from the fact that LPG is at least three times more efficient in terms of energy output even if a wood stove is used. In all, 80 percent of the total population had gas stoves, but an average of 8 kg of wood per caput weekly was also used. The need to have an alternative fuel supply was very evident, since not only is access to both fuels difficult, but entitlements to either fuel are complex and vary with both external economic factors and internal social and political pressures, including land tenure. Most wood-only users regarded firewood as a "free 9 god", in spite of the fact that 15 work-hours weekly were taken to collect sufficient for the average household of seven. From the point of view of national development strategy, these two work-days a week should be taken into account, especially as this "waste" of human resources in firewood collection is also linked to waste of environmental resources, namely forests, soil, and water, which combine further to reduce the nation's productivity. The opportunity cost of using wood rather than gas was very high for the women: 4 hours a day each for two women, or 8 hours as against 3.5 hours for one woman using LPG. It was found that inadequate fuel supply was having a detrimental effect on diet, both because fewer traditional staples like dried kidney beans (the principal source of protein) were being cooked, since there was not enough wood to keep the fire alight for long enough, and because changing daily timetables of children at school and adults working outside the village meant that food had to be heated up more times a day and at times different from those of the traditional two meals of the campesino. The role and cost of firewood is only just being recognized by the ministries concerned in Mexico. There is not at present a crisis in firewood supply, but if the rural domestic fuel situation in particular the provision of alternative fuels and suitable stoves - is not dealt with now, current scarcities could cause major social and environmental problems within 10 to 15 years, with subsequent impacts upon the national economy. Margaret Evans Adapted from Commonwealth Forestry Institute Occasional Paper 23 (1984), Firewood versus alternatives: domestic fuel in Mexico. |
In any attempt to apply formal economic methods of analysis to what determines the choice of fuels, the distinction between non-commercial and commercial wood-fuel consumers is a crucial one. Noncommercial consumers are those who obtain their wood-fuel without paying cash for it; they simply gather it from their surroundings. Commercial consumers, on the other hand, pay for the wood they use.
Non-commercial wood-fuel consumption can account for up to 90 percent of the wood-fuel used in the poorer developing countries. The vast majority of this wood is collected by families for their own use and does not pass through any market channels. This means that, for practical purposes, these activities lie outside the scope of conventional economic analysis. It can be highly misleading if analyses based upon commercial wood-fuel consumption are applied to noncommercial consumers.
Another major problem in the analysis of consumer behaviour is the lack of adequate data. The quality of the available information on domestic fuel consumption in the developing world is generally poor. Even where data exist, considerable problems can emerge in their interpretation. The use of figures based on family levels of consumption, for example, can give rise to major discrepancies when compared with those based on individual consumption levels (Leach, 1985).
Given these limitations, it is not surprising that the amount of analytic work on the economics of domestic fuel and interfuel substitution carried out to date is relatively small. Very few studies can be extended beyond their particular context except in the most tentative fashion.
On the crucial issue of whether relative price changes cause consumers to switch between wood-fuels and conventional fuels, the data are particularly meagre. There are anecdotal accounts of people switching to conventional fuels as a result of increases in the prices of wood-fuels. There are similar accounts of reverse substitution, in which people change back to wood-fuels as a result of increases in the prices of conventional fuels. But there is little information on the extent to which such changes are taking place.
MEXICAN EXTENSION AGENT AT WORK discussing a new and improved kitchen (F. BOTTS/FAO)
Cooking with charcoal requires twice as much energy as cooking with kerosene or LPG.
A study in Indonesia showed that kerosene consumption was responsive to changes in its own price but was little affected by changes in the price of wood.
What evidence there is, in fact, suggests that relative changes in the prices of wood-fuels and conventional fuels have little effect on consumer behaviour. A study in Indonesia showed that while kerosene consumption was responsive to changes in its own price, it was little affected by changes in the price of wood. Equally, wood-fuel consumption varied in accordance with its own price but very little when changes occurred in the price of kerosene (Newberry, 1984). This would indicate that, at least in the short term, domestic fuel consumption patterns are resistant to change.
But any such conclusions must be extremely tentative in the absence of reliable data. The difficulties faced by those trying to establish what determines consumer behaviour in this area have been graphically described in a report by the Energy Department of the World Bank:
For non-commercial sources, price data are by definition absent and quantity data may be obtained only from a combination of surveys and guesswork. Even for commercial fuels, retail data are usually difficult to obtain or unavailable, except for electricity, and these data may not easily be compared across regions of the country. Energy consumption overall, or of one fuel over another, may be determined more by availability than by price differences. For all these reasons, the analysis of household/residential energy consumption in LDCs is not very amenable to econometric modelling or other sophisticated techniques. (World Bank 1981)
One consistent pattern does, however, tend to emerge from most examinations of domestic fuel consumption patterns in the developing world. The choice of fuels appears to be strongly related to the income level of the consumer. At the upper income levels, conventional fuels are almost universally used to meet a substantial proportion of domestic needs; wood-fuels tend to dominate the fuel consumption patterns of the poor. The stratification of fuel use by income is broadly similar in most countries.
It is not difficult to understand why this should happen. The dirt, smoke and general inconvenience of wood-fuels have led to their decreased use or complete elimination for domestic cooking in virtually every place where living standards have risen substantially. In comparison with a wood fire, conventional fuels are clean and simple to use; their adoption is widely regarded as an indication of the modernity and higher social status of their users.
One commentator on India said the following:
The more affluent sections of society always prefer cleaner and more convenient cooking fuels to wood. The poor know that there is a hierarchy of cooking fuels and they view changes from fuelwood to charcoal to kerosene to electricity or gas as steps in the improvement of the quality of their lives. (FAO, 1983b)
This is confirmed by observations in other countries. In Malawi, for example, firewood is used by 75 percent of the low-income urban families, 25 percent of the middle-income families, and 8 percent of the high-income families. The proportions are almost exactly reversed for electricity, which is used by 80 percent of high-income families, 26 percent of middle-income families, and 2 percent of low-income families (Energy Studies Unit, 1984).
The position of poor people migrating to the city in Kenya has been described as follows, and could be repeated for many areas across the developing world:
The mass influx of population from low-income areas in the countryside implies that many urban households will initially be in a relatively weak economic i condition. Secondly they will be inclined toward energy modes associated with their earlier practices and customs.... In fact, the energy consumption of the lowest income urban families is almost entirely wood-based, much of it in the form of charcoal.... At the same time, those urban families which become more thoroughly integrated into the commercial economy show marked qualitative changes in their patterns of energy usage.... low to middle level income urban households use kerosene for lighting and to a more limited degree for heating. (O'Keefe et al, 1984)
In the rural areas, a similar stratification of energy use by income also tends to be found. At the lower-income levels, the most common fuels are non-commercial wood or other biomass. With rising income, kerosene comes increasingly into use for lighting. Higher on the income scale, bottled gas and electricity (if available) tend to be used in increasing quantities.
This brief review has clearly revealed the difficulties in making any quantitative assessment of the likely impact of conventional fuel substitution on wood-fuel demand. The fact is that there is an almost total lack of detailed and usable information on what is actually happening, and on the factors influencing the choice of fuel. It is not an exaggeration to say that major policy decisions on the levels of subsidies for conventional fuels, or on the implementation of programmes of wood-fuel plantation, are being made in almost total ignorance of the most crucial factors determining people's choice of fuels.!
At the same time, the need for action is becoming more urgent. Examination of the present patterns of wood-fuel consumption reveals that tree resources are already under pressure in many areas (FAO, 1983a). But with virtually no knowledge of the factors influencing the choice of fuels, it is impossible to say at this stage what emphasis should be placed on efforts to maintain the supply of wood rather than encourage the shift to conventional fuels.
A considerable amount of detailed analytic work on this issue is therefore urgently required. But this must be based upon actual local conditions and recognition of the fact that in each country wood-fuel consumption is heterogeneous. The total "demand" at any given time is a result of decisions by a mosaic of quite distinct groups of wood-fuel consumers, all of whom react differently to price changes and scarcities brought about by the depletion of traditional fuel resources.
Thus, in the rural areas, the response to wood-fuel scarcity will depend greatly on the patterns of landholding. Those with their own farms are under least pressure, since they can plant quick-growing trees to provide themselves with the wood-fuel they need; or they may be able to obtain most of their fuel requirements from live hedges, trimmings of fruit and shade trees, and other sources of burnable biomass from their farms. Those without landholdings, on the other hand, must find alternative supplies of fuel, be these biomass alternatives or conventional fuels.
Income levels are also crucially important. Those with very low incomes are unlikely to be in the market for conventional fuels, no matter how severe the scarcity of wood-fuels. On the other hand, upper-income rural dwellers may well shift to LPG or kerosene in the midst of an abundance of wood-fuel. The question of access to conventional fuels must also be taken into account. In many parts of the developing world, the supply of conventional fuels, and of appropriate services to use them, to the outlying rural areas is at best sporadic and quite often non-existent. Even if they wish to switch away from wood-fuels, many rural consumers are unable to do so.
In the urban areas, most upper-income dwellers are already using conventional fuels. The question of conventional fuel substitution therefore revolves around the decisions made by middle- and lower-income groups. These have a choice of i either buying conventional fuels or buying wood and reducing the amount of cooking if wood-fuel prices begin to rise as a result of scarcity. The substitution of bread and other pre-cooked foods for traditional staple dishes is an obvious though poorly documented feature of urban living which clearly has an impact on wood-fuel demand.
There is, therefore, no single answer to the question of what will be the impact of conventional fuel substitution on wood-fuel demand. The future balance between wood-fuels, and conventional fuels will be the result of a large number of decisions made by very different groups of consumers. It is only by identifying the key groups of consumers concerned in each area and patiently endeavouring to establish the chief factors that underlie their decisions in the choice of domestic fuels and their reactions to changes in the prices and availability of wood-fuels, that a coherent overall picture can be created.
BOGACH, S. 1981 A fuelwood policy for Guatemala. United Nations Development Programme
Report. Ottawa, Van Meurs and Associates & William O. Matthews Associates.
BRITISH PETROLEUM (BP). 1984 BP statistical review of world energy. London, British Petroleum Co.
ENERGY STUDIES UNIT. 1984 Malawi urban energy survey. Lilongwe, Ministry of Forestry and Natural Resources.
FAO. 1983a Fuelwood supplies in developing countries, by M.R. de Montalembert & J. Clement. Rome.
FAO. 1983b Rural fuelwood: significant relationships, by A.K.N. Reddy. In Woodfuel surveys. Rome.
LEACH, G. et al. 1979 A low energy strategy for the United Kingdom.
LEACH, G. 1985 A handbook on domestic energy consumption in developing countries. London, International Institute for Environment and Development. (Draft)
NATIONAL COUNCIL OF APPLIED ECONOMIC RESEARCH (India). 1959 Efficiency of utilisation of domestic fuels. Bombey, Asia Publishing House.
NEWBERRY, D.M.G. 1984 Equity, efficiency and the pricing of petroleum products in Asia. Cambridge, UK, Churchill College. (Draft)
O'KEEFE, P., RASKIN, P. & BERNOW, S. Energy and development in Kenya: opportunities and
constraints. Stockholm, Beijer Institute.
WORLD BANK. 1981 Energy pricing in developing countries, by A. de Julius. Washington, D.C., World Bank Energy Department.
WORLD BANK. 1984 World development report. Washington, D.C.
