 |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Our entire approach to the problem of deforestation has evolved radically away from that adopted by this project initially. Given the success of the ongoing programme, we are confident that we now have a number of positive recommendations to make, but we feel readers will be better able to Judge the merits of our conclusions if we present the whole process whereby we arrived at our present approach, rather than a straightforward listing of project achievements.
Figure
The UNEP/Bellerive project was based, throughout the period 1984 to 1987, in Githunguri, a small village near the town of Ruiru in Kiambu District. Ruiru, a recently industrialised town about 30km north of Nairobi on the main Nairobi-Thika road, is the focal point of the immediate project area.
Map of Kenya indicating key project
locations
It is an arid zone of relatively poor soils which in the past was devoted to large-scale sisal plantations. These plantations were sub-divided into smallholdings in the late 1960s, which are now cultivated primarily by ex-plantation-workers, workers on neighbouring coffee estates, industrial employees and their families.
Kiambu district is a relatively rich area of Kenya, its wealth being derived from coffee. Githunguri location and the environs of Ruiru, on the other hand, receive too little rainfall and have the wrong soil conditions for coffee cultivation. The population is thus substantially poorer than in the remainder of Kiambu, and dependent primarily on food crops - particularly maize - and on cash derived from family members working on the coffee estates and/or in non-agricultural employment in Ruiru and Nairobi.
A combined nursery and tree-planting demonstration centre was established on the project site in the opening months of the project. Officials from the Ministry of Environment and Natural Resources (MENR - Forest Department) were of considerable assistance to the project staff in establishing the nursery, in the following fields:
· selection of appropriate fuelwood species for the region· layout and design of the nursery
· arid zone planting techniques
· supply of initial seedlings from established Forest Department tree nurseries.
Project activities promoting fuelwood production can be divided into two general areas: the introduction of agroforestry techniques, and the dissemination of fuelwood tree seedlings.
a) Agroforestry
The pilot-scale use of inter-planting and food-fuel crops on the project site was entirely successful, as a demonstration of the potential of these agroforestry techniques. Yields for the crops planted between fuelwood trees were higher than the average for these soil and climatic conditions, with fuelwood yields also very satisfactory.
The project demonstrated clearly that it was feasible for an average smallholding to yield enough fuel to satisfy the family’s cooking needs for the year, without reducing food-crop yields.
Despite this success, the uptake of these techniques by the community has been disappointing. The general pattern of response seems to have been initial enthusiasm, sometimes translated into action on the part of the individual farmers, but little long-term alteration in land-use practices in the project area.
Several reasons may be put forward for this. Our main finding is that such agroforestry extension projects must identify more clearly defined and accessible project goals, from the point of view of the target community, than the straightforward “need for more trees”. This need might be perceived by the community, but still provide too weak an incentive to motivate a sustained reformation of farming techniques. We discuss this point further in sub-section 3.4 below.
A second point relates to the way in which the agroforestry component of this project was developed. Agroforestry techniques are, in a sense, an improved technology.
This project has demonstrated conclusively that, whatever the potential benefits of such a technology might be, the technical innovation cannot be expected to “stand alone”: it must be introduced as a component of a broader intervention addressing, in this case, all related aspects of land use.
This theme, the need to see improved technology as one component only of a programme aimed at reforming the overall pattern of resource-utilisation, will be very familiar by the end of this report, We mention it here in the context of agroforestry as an indication that our general conclusions on this point may have some application beyond that of improved stove projects.
b) Seedling dissemination
As a component of project efforts to promote tree planting among the inhabitants of the project area, the project nursery was used to cultivate fuelwood species to an age of 2 to 3 months. The young trees were then distributed to the local population together with instruction on planting and cultivation.
Although the seedlings proved enormously popular, with people travelling considerable distances to collect them, the survival rate of trees planted by individuals in the area was generally poor. Several reasons may be advanced for this, the very adverse climatic conditions which obtained in the period 1984-85 being the most obvious. But it is not enough just to blame the drought, since other projects have had very similar experiences under much more favourable conditions.
Nor is it enough to say that local awareness of the need for reafforestation was simply too low: the lengths people were prepared to go to in obtaining the seedlings suggests that they were more than adequately aware of the need to plant trees. Credit for this level of awareness must be given to the project’s very limited extension services, and also to the promotional activities of the Forest Department.
Why, then, was the survival rate so poor? It is tempting to blame short-sightedness and/or lack of persistance on the part of the local people responsible for cultivating the trees, but it might equally be said to reflect the fact that the villagers had a better idea of the real economic value of these trees than did the project management: this point is discussed in more detail below.
The second major component of the project was the introduction of fuel-saving stoves, as a means of reducing fuelwood consumption to a sustainable level. Considerable doubt has been cast in recent years on the fundamental validity of this approach: “does fuelwood consumption directly determine the rate of deforestation?” Probably not, in many instances. “Does the use of improved technology affect the overall rate of fuelwood consumption?” Perhaps more efficient utilisation of fuelwood for cooking means that fuelwood is used for other purposes, or that the transition to other fuels is postponed, or...
None of these questions has a straightforward answer, applicable everywhere. We will discuss these issues in depth as we consider specific aspects of woodfuel demand reduction programmes below. In a report of this nature, however, we must begin by considering the project on its own terms.
We set out, in 1984, to disseminate domestic fuel-saving stoves. That component of the project was effectively wound up in 1986 as completely unpromising: why?
In considering the initial activities relating to woodstove dissemination, the key point to note is that the dissemination strategy, as is the case in the majority of such projects, was largely determined by features of the technical design of the stoves under dissemination. These stoves were:
· the Kanini Kega Stove (or protected open fire): a single-pot, chimneyless “stove” consisting of an air-dried clay wall built around three brick pot supports and a rudimentary grate.· the “Pogbi” stove: a double-pot stove also built of air-dried clay, with a chimney and sheet-metal cover. The construction method required for both these designs involves manually compressing a clay/straw mix into a heavy mould made of sheet-metal or wood.
Building a “Pogbi”
stove
The following should be noted:
· the stoves are extremely heavy: they were designed to be built on the site of use.· special equipment is required for construction
· the craftsmen who build them must master a number of unusual skills (albeit not very high-level ones): in particular, how to work the clay. Clay is used in the project area as a building material, but for this application it does not have to be worked in the same way as it does for the production of durable clay stoves.
The only feasible dissemination strategy for such designs was to establish a network of specifically trained, professional stove-producers. The nature of the work was such that only relatively unskilled artisans were prepared to take it up - working clay being comparable to the most strenuous tasks in the agricultural or building sectors. The requirement that a mould should be carried to the construction site restricted the prospective stove producer to a relatively small accessible market.
This approach had been adopted previously by Bellerive Foundation in Pakistan, and had worked. By the end of 1985, it was dear it was not working in Kiambu. Two groups had been trained, totalling 21 artisans, and all but two of the trainees had stopped producing stoves within a few weeks of completing their courses. Even those two had not established themselves independently, as intended, but had returned to their former employment, and were producing stoves only on an ad hoc, part-time basis.
The key reason given for this failure at the time was the high cost of the stoves produced. But this cannot be regarded as the only factor: the training centre never experienced any difficulty in finding customers for the stoves produced by the trainees, with these costed to reflect the prices which would be charged by a private independent producer. The most expensive design - the two-pot stove with a chimney sold at prices ranging to over Ksh. 1.000/- (US $60) -was also the most popular; while the cheapest stove - the basic protected open fire costing KSh 50/- (US $3) - was clearly the least popular, despite the fact that its performance, in terms of fuel consumption, was as good as the more elaborate models.
Whatever may have been thought initially, reducing wood-fuel consumption at minimum cost was clearly not a priority among the inhabitants of the project area. Other considerations, such as eliminating smoke from the kitchen and introducing a general air of modernity into cooking, probably took precedence.
Thus there clearly was a market for the stoves, albeit one made up of potential purchasers motivated in a different way than originally thought. Why, then, did the “graduating” trainees find it so difficult to establish themselves to supply this market?
The marketing infrastructure required to provide such products to the consumer was unfamiliar and inappropriate to the project target area. In other words, there was nothing wrong with the designs of the stoves from the technical point of view, it was simply that these designs forced upon us the wrong dissemination strategy. It is a clear example of technical considerations exerting a disproportionate influence over project development.
There are two ways in which “consumer durables” are obtained by the inhabitants of the project target area: either they are bought from a central selling point, or they are built on-site by a non-specialist, usually a relative or family member who has some training in building skills. The specialist plumber, who does nothing but plumbing and does that for cash, is virtually unknown, never mind the specialist stovemaker.
The products introduced by the project needed to be built on-site by a craftsman who had received a course of training specific to those products, and who was confined to a relatively small geographical area owing to the difficulty of transporting the equipment involved. Dissemination was therefore doomed from the outset: the craftsmen built stoves for their immediate families and relatives living nearby, and then stopped.
The importance of tailoring improved technology to the needs of the consumer in the project target area is generally appreciated. The emphasis placed on the consumer may however, lead to the producer being neglected: the product which the housewife wants may not coincide with the product which the local artisan finds easiest to produce and market.
With effective promotion, we can often persuade people to buy a novel product. But to persuade producers to adopt a production and marketing system different from that to which they are accustomed is virtually impossible. Think of the risks involved: the husband who buys his wife an improved charcoal stove is risking KSh. 90/= (US$5); the artisan who takes up stove-making as a career is risking his livelihood.
Given these problems, we decided in 1986 not to pursue the domestic stove dissemination programme further, in order to concentrate on the (radically different) institutional sector.
With the institutional fuelwood saving programme now fully established and operational, we have, in the course of 1988, begun to look again at the problem of domestic fuelwood consumption. Some of the ideas to emerge from this are given in subsection 4.2.2 below.
A recurrent theme throughout this report is the need for projects addressing the problem of deforestation to be tailored to the specific features of the problem relevant to the area in which they are operating. And the key to that is to establish what the problem actually is in the first place: it may not be obvious.
A problem has been identified, which relates to deforestation - hence the initiative to establish a programme to combat it. But before we do anything else certain basic questions need to be answered. Is the problem one of excessive fuelwood demand? Perhaps the observed increasing scarcity of fuelwood is due to clearing of forests for agriculture. If it is a problem of unsustainable demand, is it an economic problem (fuelwood prices rising uncontrollably) or a social/humanitarian one (women being subjected to an increasingly degrading and time-consuming task: the collection of fuelwood)?
Is the problem that, despite relatively efficient use of resources, demand simply outstrips supply? Or that fuelwood is being utilised inefficiently? Or that the wrong fuel is being used, with viable and environmentally sound alternatives to fuelwood being available? Or that the fuelwood is there, but in the wrong place, and at the wrong price, due to inefficiencies in the marketing and distribution infrastructure?
The answers to these questions are always built into the design of a woodfuel conservation project, at least implicitly in the strategies adopted.
The problem is that the questions themselves axe very seldom asked, and still more seldom are the answers arrived at in a manner which could be said to guarantee that they reflect the conditions and aspirations of the communities which the project is to deal with.
The root cause of this problem seems to be that donors assume (or at least, donors are assumed to assume) that any implementing agency worth its salt would have the answers to such basic questions at its fingertips, and therefore that the inclusion of an exploratory phase in a project proposal, during which these questions are investigated, must be a sign of inexperience on the part of the project implementors. Those responsible for drafting and submitting proposals feel obliged, often against their better Judgement, to make the answers to such questions implicit in the strategy they propose, in the fear that if these things are left too vague it will be thought that they don’t know what they’re talking about.
But this is precisely the point. We often don’t know what we’re talking about at the initial proposal-submission stage. The diversity of the fuelwood problem is such that experience gained in another country, or even in another district of the same country, will never provide us with the answers to these questions. Such experience may assist us to find out what the answers are (not least in suggesting questions we may not otherwise have considered). But until an adequate problem-definition exercise has been carried out on the ground, in the project target area involving local participation, no one, least of all the expert on fuelwood conservation, should claim to be very sure of anything at alt.
Consider the design of the UNEP/Bellerive project in Kiambu: as a result of pressure for an “action-oriented” project with a minimal research component, no formal socio-economic or demographic surveys were built into the original project design. The project took a straightforward economic approach, assuming that the use of improved stoves and the cultivation of fuelwood trees could be presented to the target community as economically viable activities.
Problems emerged: although the fuelwood economy in the project area was partially monetised, the degree of monetisation was substantially lower than originally thought. Owing to the availability of waste from the coffee estates, the inhabitants of the project area were obliged to obtain only a relatively small proportion of their fuel needs through the purchase of cut firewood. Coffee waste often had to be paid for, but the price varied so considerably that it was virtually impossible to place even an order-of-magnitude figure upon it
Without a formal survey built into the project, informal inquiries in the project area suggested a substantially higher degree of monetisation of firewood, and a much higher cost, than was probably the case in practice.
This was entirety understandable, and a frequent feature of projects of this nature: the reason is simply that the proportion of firewood which has to be purchased with hard cash is by far the most painful for the consumer, and so verbal reports of firewood prices tend to exaggerate it.
Verbal reports of firewood
prices
The collection of waste from coffee estates consumed a considerable amount of time, the burden being born principally by the women of the project area, The cost of this activity in social and humanitarian terms might be considered to be quite large, but it to important to note that in purely economic terms this cost was relatively low: the Opportunity cost of firewood (resulting from the time taken up in collecting it) being reduced by the lack of income-generating activities available to the women who had to undertake this task. The high population density and small size of the agricultural holdings meant there was general excess of labour available.
This lower-than-anticipated real cost of fuelwood clearly had a considerable impact on the progress of both components of the project. It was assumed that people would see the improved stoves as an investment which would help them to save money on firewood. But if they were not spending money on firewood in the first place... Likewise it was assumed that people would see the value in growing more trees. But if trees are free...
We are not criticising these initial assumptions themselves: although they turned out to be wrong, they were perfectly reasonable given the information available to the project management at the time. Nor are we criticising the fact that the assumptions were made: such assumptions had to be made in order for the project to get started. It is completely unrealistic to think that we can determine objectively all aspects of our strategy through academic surveys before we begin the implementation phase.
What we are criticising is this: it was not acknowledged in the original project design that these were assumptions, and that, as such, they might turn out to be wrong. There was therefore no mechanism built into the project for correcting our course of action once implementation began.
If a project turns out to be based on a misguided assessment of the needs of the project target community, the usual response is to say that more research should have been carried out initially. In the course of this report, we frequently reiterate the need for studies to be carried out on the ground to clarify the problem, determine impact priorities, develop dissemination strategies and so on. Is this not an invitation for programmes to get bogged down in a morass of academic research?
Elaborate surveys and pilot studies which lead to no effective action are too familiar in the development world for us to ignore the danger. And the approach which we advocate here may well appear elaborate and academic: research must be undertaken initially simply to establish what the problem realty consists in, then to identify the priorities for any programme addressing it, then to develop and pilot locally appropriate designs and dissemination strategies... Anyone with experience in this field will already be beginning to smell the rot setting in: years of surveys and countless consultant man-months before anything actually gets done.
The rot setting in
We have two ways of avoiding this: focus and flexibility.
First of all, in the problem definition exercise, we should aim to identify specific components of the fuelwood problem in the project area on which to focus our efforts. Provided we restrict ourselves to a sufficiently well-defined and homogenous field, then we have some hope of developing a realistic and effective programme within a reasonable time-frame and at a finite cost
What we mean by homogenous may be understood by an example: almost all large-scale permanent institutions in Kenya have similar catering requirements and operate under similar conditions. It therefore makes sense to establish a single programme to address the problems of fuelwood supply and demand in this sector. On the other hand, the domestic fuelwood consumer in the Central Highlands is in a completely different situation to one in a lowland arid region. Thus a single, uniform, domestic fuelwood saving programme on a national scale is impossible: this is not a sufficiently homogenous aspect of the problem of fuelwood scarcity.
Not really the same problem
Our second point, the need for flexibility, may be more controversial. Donors seem to appreciate tidy, linear projects. One activity should be completed, then the next activity begun on the basis of the results obtained from the first activity. There are obvious advantages to structuring a project in this way: it facilitates project timetabling, provides a simple check on whether the project is on schedule, makes for clear reporting and so on.
There are two problems with adopting such a linear approach to the design of fuelwood conservation projects. First, as remarked above, if we aim to finalise the research phase completely before proceeding to implementation, then there will definitely be an unacceptable delay between the project launch and anything concrete being achieved on the ground. Pressure will be placed on project managers to minimise this delay, which usually results in inadequate attention being given to the crucial initial exercise of problem definition.
Secondly, it is questionable whether it is possible to adopt such a linear approach. There is no clear cause-and-effect progression between programme components. The progress of pilot dissemination may contribute to clarifying the problem; technical designs will suggest dissemination strategies which in turn will suggest modifications in the design and so on. Everything is interconnected.
The fuelwood problem is such that it is completely impossible for any activity in a programme addressing it to be completely and objectively determined by the results of previous activities.
There is probably no way, for example, that a survey involving nothing more than the collection of data on the project target community, could have forewarned us of the fundamental problem encountered in the initial dissemination of domestic stoves (see above). When questioned, it was clear that local artisans either had very little idea of the make-up of their clientele, or (more likely) were disinclined to reveal such sensitive information to a numerator. Thus only by trying it out, could we determine whether or not the proposed dissemination strategy was going to work.
Every afforestation project must therefore begin with a phase during which the project management is free to try out, with the community, different strategies and approaches: on the understanding, accepted by project supervisors and donors alike, that not every strategy tried will necessarily succeed.
The need for such flexibility undoubtedly places a considerable responsibility on the project implementors in the field, to ensure that opportunities are taken to try out new ideas, while tight control is still retained over project activities. But unless we accept such an exploratory approach we will always be imposing the priorities of the project onto the community. If we are extremely lucky, these may coincide with the community’s own priorities. But experience suggests that the chances of this are slim.
 |
 |
