INTRODUCTION

This paper is concerned with the impacts of non-wood forest products (NWFPs) on people. Probably the majority of rural households in developing countries, and a large proportion of urban households, depend on forest products to meet some part of their nutritional, health, house construction, or other needs. Very large numbers of households also generate some of their income from selling forest products. The paper attempts to outline the extent and nature of these patterns of usage and dependency, how they vary spatially and over time, and the factors that appear to influence these differences and changes.

Most people are continuously adapting their livelihood strategies to changing circumstances, and this can mean that the role of NWFPs is changing, often very rapidly; some people turning to greater use of such products as their circumstances change, others moving to use of alternative products or materials, or to different activities. Knowledge just about the patterns of present use is therefore likely to be of only limited value in determining what interventions might be needed in order to maintain NWFP supplies, or to support important household-level subsistence and commercial activities in the future. It is necessary to be also able to identify the directions of these changes and the factors underlying them. The paper therefore pays particular attention to the dynamics of the relationships between people and the NWFPs they use.

Such an exercise is necessarily shaped by the extent and quality of the data available. As NWFP uses and activities in the subsistence and small enterprise sectors escape the attentions of statistical recording systems, quantitative information on their magnitude and structure is very sparse. There is a great deal of descriptive information, generally concentrated in narrowly situation-specific accounts. Few analytical studies relate the use of forest products to household livelihood strategies, and even fewer have attempted to synthesize the information available. Any review of this kind must therefore be taken as indicative rather than conclusive.

An added constraint is the difficulty in defining boundaries around the subject. A forest products activity usually constitutes just one activity within an agricultural household. It then becomes difficult to separate out that part of household time, costs, returns, etc., that is attributable to just that activity. Similarly, many NWFPs are gathered or harvested outside "forests" as normally defined; being produced in managed fallow or farm bush, or from trees managed as farm crops. This can make it difficult to identify the consequences of NWFP use for forest management. In this paper we have examined use regardless of origin.

A related boundary problem arises from the definition of NWFPs. Many of the forest products that people produce and use from the forest are wood products. Many surveys and studies consequently look at the whole range of forest products being used, and it becomes difficult to separate out just those that are non-wood. Where necessary, therefore, some of the analysis in this paper takes account of all forest products that are important in understanding people's dependence on the forest.

The paper is organised as follows. Following a section that briefly considers the importance of NWFP use at the level of the national economy, the third section examines the significance of the main NWFP end uses (food, medicine, construction, income) at the household level. This explores differential degrees of dependence on the part of different sections of the community, and general trends in levels of use and dependency. In the fourth section, the nature of change, and the factors influencing change, in household incomes derived from NWFPs, are explored further. Four main issues that arise are then examined in more detail in the penultimate section: equity issues, adequacy of access to raw material supplies, tenure and control issues, and support programmes.

NWFP USE AND THE NATIONAL ECONOMY

The importance of NWFPs at the national level lies in the huge numbers of people involved in gathering, hunting, processing, trading and other aspects of their production and use 3/. As noted earlier, most rural people use some forest products, and many obtain part of their income from forest-product activities.

The lack of information about the numbers involved in subsistence use, or about the value of that use at the household level, make it impossible to arrive at even rough estimates of the economic contribution of that component of production and use. Somewhat more information is available about employment in income-generating activities. Results from small enterprise surveys in six countries in southern and eastern Africa indicate that an estimated 408,000 forest product enterprise activities provide employment for 763,000 persons an average density of 16 persons per thousand in the population (Arnold et al., 1994) 4/. Household surveys in high forest zones show much higher densities e.g. 68 percent of households in areas surveyed southern Ghana (Falconer, 1994), and all households in selected areas from which the city of Iquitos in the Peruvian Amazon draws its supplies of NWFPs. Table 1 reproduces estimates of numbers involved in a selection of other activities and situations.

Even where absolute densities of forest product-based employment are low, they usually account for a high proportion of overall non-farm employment. In a review of small enterprise studies, Liedholm and Mead (1992) reported as follows:

Small forest product enterprises can also account for a large part of overall employment in forest industry sector. Thus the estimate of 237,000 persons employed in small forest products enterprises in Zimbabwe in 1991 compares with a reported 16,000 employed in forestry and forest industries in the country in that year (Arnold et al., 1994).

NWFPS AND HOUSEHOLD NEEDS

Introduction

Most households exist in a situation within which their activities are influenced by an array of interrelated objectives, constraints, and other factors. Household livelihood strategies are likely to include pursuit of secure provision of food and other essential subsistence goods, cash for purchase of outside goods and services, savings, and social security. Other factors are likely to include concern to reduce critical risk factors, and local social, cultural and spiritual considerations.

Social and Cultural Context

Forests and forest products are linked to household livelihood systems in a variety of different ways. Forest products commonly contribute to meeting food and other basic needs, are a source of income and of inputs into the agricultural system, help households control exposure to risk of various kinds, and constitute an integral part of the habitat and of the social and cultural structure of those living within that environment. In the present paper we are examining just some of these goods and services from the forest. However, as Levin (1992) has pointed out, in writing about southern Thailand:

In many situations the strength and extent of cultural considerations has diminished, but many have persisted, despite the prevalence of "Western" ideas, practices and products. For instance, Table 2 shows how people in southern Ghana include both physical and conceptual values in assessing the benefits to them to be obtained from the forest. They therefore remain a potent factor in decisions about NWFP uses and values.

The differences from village to village reflected in that table underline another characteristic of NWFP use that should be kept in mind in what follows. That is the large variation in needs and practices that can and often does exist between different communities, among households within a community, and within a household between men and women and members of different ages. This limits the extent to which conclusions that can be drawn from studies on this subject are likely to be of general application.

Table 2: The highest valued forest benefits in surveyed villages, South Ghana
 
 
 

Village
Product	Banso	Betinasi	Essamang	Nkwanta	Essuowin	Koniyao	Kwapanin	Nanhini	Number of people  Ranking product first  in all villages
	(Percentage of people who rank the benefit first)
Pestle	28	9	27	33	45	31	38	24	71
Bushmeat	40	9	27	38	37	26	36	13	68
Canes	48	18	33	48	29	15	15	10	56
Building mat	24	18	13	33	8	15	30	10	43
Chewstick	40	9	13	38	18	5	15	7	39
Timber	20	9	27	-	32	21	19	-	39
Water	4	9	7	5	11	33	6	10	27
Medicines	16	9	13	10	5	5	6	27	24
Sponge	16	-	27	5	18	3	15	-	24
Gods	-	-	-	-	16	-	-	50	21
Land bank	-	-	-	24	3	10	13	7	18
Wrappin L	-	-	-	5	3	-	32	-	17
Fuelwood	20	-	13	-	3	8	6	7	16
Mortar	12	-	-	-	5	10	11	7	16
Fertility	8	9	-	5	3	13	13	-	16
Rains	4	18	-	-	5	5	11	10	15
Forest food	16	-	-	14	3	3	11	-	14
Raffia	16	9	7	10	3	3	4	-	12
Others	8	18	7	10	24	18	17	3	25
Total number of people interviewed	25	11	15	21	38	39	47	30	226
Notes: Some people named more than one product as the most important Source: Falconer 1994

NWFPs and Household Food Security

Another feature that needs to be recognised at this stage is the way that different NWFP activities are linked as components of livelihood strategies that households employ. The concept of food security for rural households in developing countries encompasses all factors affecting a household's access to an adequate year round supply of food. Thus it is concerned not just with the household's production of food crops, but with the availability of income to the household with which to purchase food, where this is necessary.

In examining household use of NWFPs, we therefore need to identify their effectiveness in both providing gathered foods that contribute to food self-sufficiency, and saleable products that could supplement income needed to purchase food. In doing so it is also necessary to consider whether, and if so how, income-generating activities based on forest products affect other aspects of a household's capacity to contribute to its food self-sufficiency.

Increased commercialization of forest food products could lead to over-exploitation of the resource, or to diversion of food needed for household consumption from local use to sale on the markets. Households may sell not only what is surplus to their requirements but also food needed in the household. Changes in use of agricultural or grazing land to favour production of NWFPs for sale could reduce the amount of land available to the household for producing its basic food crops. Introduction of time consuming NWFP gathering or processing activities could be at the expense of time women need to cook and look after their children (Longhurst, 1987).

Subsistence Uses of NWFPs

Forest Foods and Household Nutrition/

There are many different kinds of food gathered from forests. Forests also provide the habitat for many commonly consumed wild animals and fish. Forest foods may also be smoked, dried or fermented, making them available over extended periods of time.
 

Importance in diets

For the majority of rural people, forest foods add variety to diets, improve palatability, and provide essential vitamins, minerals, protein and calories. The quantities consumed may not be great in comparison to the main food staples, but they often form an essential part of otherwise bland and nutritionally poor diets. Diet diversity is an extremely important element of nutritional well-being, in part because more vitamins and minerals are consumed, and also because it improves the taste of staple foods. The most common supplementary foods are leaves and wild animals, both of which are generally added to sauces and soups which accompany staple foods.

Some species are noteworthy as particularly rich sources of vitamins, minerals, proteins and fats. For example, many forest fruits and leaves are good sources of Vitamin A, shortage of which is a common cause of blindness in many developing countries. Forest foods can thus be used to combat nutrient related health problems (Table 3).

Forest and farm tree products are also valued throughout the year as snack foods. They are commonly eaten on the job: while working in fields, while herding and while gathering fuelwood. Forest fruits and nuts are the most common snack foods, especially for children.

On the other hand, some favoured forest foods such as snails, mushrooms and honey have particular harvest seasons that do not necessarily correspond to food short periods. In these cases, foods are gathered for as long as they are available.

Home gardens (intensively managed farm systems combining tree and herbaceous crops) are widely designed to make use of variations in the timing of the harvest of different component tree crops, in order to supply foods and saleable produce during the period between harvests of staple crops. Another important feature of such gardens, and other systems incorporating trees, is that work on the latter can often be undertaken during the slack season, thus helping to even out the peaks and troughs in the demand for farm labour.

Forest foods and emergencies

Many studies indicate that forests have provided essential resources during emergency periods such as floods, droughts, famines, and wars. There is a wide range of forest resources used as emergency foods. Often they differ from resources exploited in other periods. In famine periods, roots, tubers, rhizomes and nuts are most sought after. They are characteristically energy rich, but often require lengthy processing.

Changes in consumption of forest foods

Changes in the role that forest food plays in household nutrition may reflect penetration of rural markets by new food products, changing tastes, or decreased availability. However, decreased availability may reflect changes in the availability or allocation of a household's supply of labour rather than physical shortage of the product. As the value of labour rises with increasing wealth, the opportunity cost of continuing to spend time gathering foods, rather than purchasing them, becomes increasingly unattractive. In areas where rural populations have achieved high incomes and easy access to purchased foods use of forest foods is likely to be very little.

Some studies indicate that emergency uses of forest resources are dwindling as people rely to a greater extent on food purchasing, or as famine relief programmes become more effective. In others improved supplies of food crops have diminished the need to depend on forest foods. In Vanuatu, for instance, the introduction of the sweet potato, which could be planted at any time and produce an edible crop within three months, and manioc, which can be left unharvested for up to two years, has made the traditional emergency foods of wild taro, arrowroot, wild yams and sago virtually obsolete (Olsson, 1991).

Many studies report that a decline in use of forest food accompanies reduced knowledge about its use. As children spend more time in school than in the fields and the bush, the opportunity to learn about wild foods is reduced. Sedentarization is another widespread change that distances people from the food sources they used to be familiar with (e.g. Melnyk, 1993). Poorer knowledge about which plants can be consumed, and which cannot, further constrains people's use of these foods, even when they are still available and important for dietary balance.

Another widespread trend affecting rural consumption patterns and levels is the diversion of foods available to rural households to burgeoning urban markets. This has been reported to be the case with bushmeat in the forest zone of Ghana (Falconer, 1994); in many cases, the loss of regular supply of bushmeat has not been replaced by domestic meat, but has simply meant that less animal protein is consumed. Forest fruits in parts of the Amazon form another example (Melnyk, 1993). It is often the poorest, forced to exploit this source of income because of lack of alternative opportunities, who are most likely to suffer nutritionally from this diversion of food supplies.

The impact of declining consumption of forest food is not clear. In some cases these changes do appear to have led to a poorer quality diet; most notably as greater reliance on purchased food reduces dietary diversity. But greater reliance on the latter does not necessarily result in adverse nutritional consequences. Perhaps the worst impact is that poorer people's food options are being progressively reduced, especially during seasonal and emergency hardship periods (Falconer, 1989).

In some areas forests still supply a readily available source of foods and fodder. Also, particular foods continue to be consumed for their traditional social value, or for their medicinal qualities. But even in these cases the diversity of gathered foods consumed may have decreased.
 

NWFPs and Health Uses

The use of forest products for medicinal and other health purposes is very widespread; often in urban as well as rural households. Very large number of forest plants, and often some animal products, are frequently used within a single community. For example, 214 instances of medicinal use of plants were reported in a community in Sierra Leone (Davies and Richards, 1991), and 150 medicinal plants were observed in a location in Vanuatu (Olsson, 1991). Even in the relatively species-poor sal forests of West Bengal, 47 species have been recorded as being used in 42 villages (Malhotra et al., 1993).

Medicinal usage tends to overlap with that of forest foods; indeed particular items added to foods serve both to improve palatability and act as a health tonic or prophylactic. There are also often strong links between medicinal use and cultural values; for example, where illnesses are thought to be due to the spirits, or plants have acquired symbolic importance as treatments.

Such values often underlie the division between use of traditional and "Western-style" medicines that is widely observed at the present time. For example, it was found in southern Ghana that choices between the two were influenced by users perceptions of their effectiveness, but that the principal factor was the following:

NWFP Use in House Construction

Much house construction in the rural areas (and in many urban areas) in developing countries involves use of forest products. Although much of this is use of timber for door frames and poles for the frame of mud walls, this frame is typically bound together using canes, lianas, raffias or twines made from other fibrous plants, and roofs are commonly covered with grass, bamboo, reeds or leaves. Where it is available, bamboo is heavily used. It is reported that in Bangladesh over 70 percent of rural dwellings use bamboo as the prime building material (Dunham, 1992, cited in Wells et al., 1994).

A recent review study (Wells et al., 1994) notes that such usage is closely associated with poverty:

Nevertheless, there is evidence of practices that clearly do represent a loss of quality, or a deterioration in housing standards for those concerned as biomass construction materials become more difficult to obtain. These include reduction in size of the dwelling as materials supplies tighten, use of less durable species so that dwellings have to be replaced more frequently, and less frequent re-thatching and maintenance of existing structures so that their condition deteriorates (Wells et al., 1994).
 

Forest Products and Household Incomes

Employment and income from small-scale non-farm enterprise activities are nearly everywhere becoming of growing importance in the rural economy of developing countries. In stagnant or slowly growing agricultural areas small enterprise activities provide employment to surplus labour; in conditions of growing agricultural incomes they contribute to the process of growth, diversification and the shift to more productive uses of rural resources (Haggblade and Liedholm, 1991).

It has been estimated that rural non-farm work provides 20-45 percent of full-time employment in rural areas and 30-50 percent of rural household income (Kilby and Liedholm, 1986; Haggblade and Hazell, 1989). As was noted earlier, results of surveys of the small enterprise sector have shown that small forest-products activities everywhere account for a substantial proportion of the total (Liedholm and Mead, 1992).

Characteristically, forest-products activities form only one part of a household enterprise. In Zambia, 64 percent of those enterprises operated by persons previously in farming were found to be run in conjunction with farming, and 30 percent of them with one or more other small enterprise activities. Of those where enterprise activity is the principal one, 56 percent also farm and 65 percent have other activities in addition to the forest based one (Fisseha and Milimo, 1986). As was noted earlier, this close integration with other household activities makes it very difficult to separate out income from the forest-products activity, and even more difficult to define how it contributes to rural household livelihoods. Few if any studies focus on how forest-products income is spent, nor on who spends it.
 

Importance of Forest Products in Meeting Household Income Needs

For some, the forest based activity may be the sole or principal source of income. Rubber tappers in the Brazilian Amazon provide an example where this was until recently the case with a gathering activity. Sale of forest produce is also the principal source of income in many engaged in hunting. In the villages around the Korup National Park in Cameroon it was found that hunting is the single most important source of cash income for the majority of village households and for the village as a whole (Infield, 1988). The larger and more profitable processing activities, such as wood working, can also become the dominant household activity. In a recent study of six countries in Africa it was found that nearly half of those engaged in woodworking reported it as being their sole source of income, and for more than 60 percent it was their main source (Arnold et al., 1994).

More widely, forest based activities are engaged in part-time by farm households which cannot raise enough to be food self-sufficient year round. A study of a lowland village in the Philippines found that 73 percent of the households could not generate enough food or cash income from agriculture to meet their basic needs. All village households collected forest products for supplementary and emergency income; with more than half depending on rattan collection and timber wage-labour as a primary source of livelihood (Table 4). The average earnings from rattan collection was greater than the average wage for agriculture or timber production labour (Siebert and Belsky, 1985).

 
 
 
 

Forest product activity2	Use by household rice self-sufficiency status3
	Low (n-14)	Middle (n-27)	High (n-22)	Total (n-63)
Rattan gathering	57	37	9	32
Employment as timber labourer	43	33	36	37
Either rattan or timber	79	48	41	52
1 Household rice self-sufficiency refers to the ability of a household to meet its rice consumption demands through rice farming (whether irrigated or rainfed, or as owner-cultivators and sharecroppers).  2 At lease one adult household member (15 years and older) gathers rattan on a weekly basis or accepts work as a timber wage labourer whenever work is available.  3 Low: no rice production; Middle: up to 50% self-sufficiency; High: more than 50% self-sufficiency. Rice self-sufficiency is a measure of a household's economic situation.  Source: Siebert and Belsky, 1985

The evidence from the few studies that indicate levels of income indicate that returns to the practitioners varies greatly, with some activities providing only minimal returns to producers. A study in the Rufiji District of Tanzania (Havnevik, 1980) found returns from carpentry and carving activities to be almost twice the level of the minimum wage, while those from mat and basket making amounted to only a third of that wage (Table 5). However, although basketry and mat-making tend to generate low returns to the producers in many situations, they are still important in terms of contribution to overall household income. In the six African countries surveyed, in which cane-based processing is the largest NWFP activity, over 40 percent of the entrepreneurs engaged in these activities stated that the enterprise accounted for 50 percent or more of their household's income, and nearly a quarter reported that it was their sole source (Arnold et al., 1994)./
 
 

 
 
 
 

	Labour return categories and crafts
	I  Up to Shs. 1/=	II  Shs. 1-2	III  Shs. 2-3	IV  Shs. 3/= and above
Crafts:	Wooden beds  Mortar making Mat making  Basketry	Sawmilling  Canoe making  Masonry  Pottery  Salt production  Shoe repair  Fishnet production	Blacksmith  Carpentry  Carving  Canoe  Tailoring  Coconut-oil  Bicycle repair	Jaggery  Watch repair
Average hourly return for the category  Monthly potential income 26 days (208 hrs/month)  Number of nominal participants  % of total nominal participants	Shs. 0.48    Shs. 99.84      16,072    56.8	Shs. 1.51    Shs. 314.00      5,866    20.7	Shs. 2.45    Shs. 510.00      6,305    22.3	Shs. 5.91    Shs. 1,229.00    60    0.2
Source: Adapted from Havnevik, 1980

The seasonality of other forest product activities is directed by the seasonality of other activities, and the fluctuations in availability of labour. Many activities therefore decline in agricultural and planting seasons, or are phased to take advantage of slack periods. Others are governed by seasonally induced cash needs, such as school fees, or the need for income to buy food during the "hungry period" between harvests, or to purchase farm inputs.

Often these pressures work in conjunction one with another. A study in Sierra Leone found that fuelwood selling provided the first cash income from land cleared for rice production. Subsequently fuelwood collection for the market was concentrated during the off-peak agriculture period, providing cash income in a period when food supplies are generally at their lowest (Kamara, 1986).

The collection and processing of babassu palm kernels in northeastern Brazil is another example of an activity geared to the seasonal slack period in agriculture, which is also the period of most severe cash needs. Income from sale of kernels was found to account for 39 percent of cash income and 34 percent of total household income during this season (May, 1985). The majority of farmers in the area are poor tenants, and kernel collection is one of the few ways they can supplement their cash income. Both men and women are involved with the collection of babassu fruit though it is primarily the women who are involved with oil extraction from the kernels. Many of the poorer farmers were dependent on this cash for purchasing seed and other inputs for the new season's planting.

Small manufacturing enterprises that produce for rural consumers are exposed to the additional seasonal constraint that incomes, being agriculture-based, have a short peak. Rural demand may also be linked to particular seasonal agricultural needs. Much of the demand for baskets in rural Ghana, for example, is to store particular agricultural products; and is therefore concentrated at the time of harvest of those products (Falconer, 1994).

The Buffer Role of Forest Income

The third role of forest-products activities, as a source of income that people can fall back on in times of crop failure or shortfall, or in order to cope with some other form of emergency, can be very important. Many reports recount how more people engage in the gathering and selling, or processing, of whatever forest product is traded locally during difficult times. It has been reported from south Ghana, for example, that:

NWFPs, the Poor, Women and Children

Where people have had relatively unrestricted access to forests, forest food is often particularly important for poorer groups within the community. While forest gathering activities are not restricted to the poor, the latter depend on these activities to a greater extent. They are also an important source of income to the poor. With generally easy access to the forest resource, poorer people have been able to exploit forests for marketable products. As most small-scale processing and trading activities require little in the way of capital or skills to enter, they also are usually accessible to the poor.

Numerous studies of different communities and situations confirm this greater degree of involvement and dependence by the poor. One of the largest bodies of information on the subject comes from Jodha's studies of the dryland plain areas of India. Table 6 shows how the poor in each of the areas studied depend much more heavily than their wealthier neighbours on both saleable produce and fodder and fuel for own use from adjacent common lands. At the case study level, dependence on NWFPs has been shown to be related to size of landholdings in Orissa, India, (Fernandes and Menon, 1987) and in Brazil (Hecht et al., 1988), with family incomes in Sri Lanka (Gunatilake et al., 1993), and with levels of household rice self-sufficiency in the Philippines (Siebert and Belsky, 1985).

 
 
 
 

State	CPR contribution to
	Household  Category	Income  (%)	Fuel supplies  (%)	Animal grazing  (%)	Days of employment per household
Andhra  Pradesh  Gujarat    Karnataka    Madhya  Pradesh  Maharashtra    Rajasthan    Tamil Nadu	Poor  Wealthy  Poor  Wealthy  Poor  Wealthy Poor  Wealthy  Poor  Wealthy  Poor  Wealthy  Poor  Wealthy	17  1  18  1  20  3  22  2  14  1  23  2  22  2	84  13  66  8  -  -  74  32  75  12  71  23  -  -	-  -  82  14  83  29  79  34  69  27  84  38  -  -	139  35  196  80  185  34  183  52  128  43  165  61  137  31
Source: Jodha, 1990

The characteristics of easy access to the resource and low entry thresholds enable many women to also generate income from forest-products activities. Forest product processing may often be performed at or near home, allowing women to combine these income earning activities with other household chores (i.e. child care). In addition, as women traditionally use forest products to meet some of their household's basic needs (e.g. fuelwood, medicines, and foods), gathering of forest products for the market can often be accomplished in conjunction with other collecting activities.

Such activities are often an important source of the income that women need to meet the costs of feeding and clothing the family, and their other needs for cash. They therefore tend to rely more frequently than men on forest-products activities for the generation of income. This, and the need to fit these activities in with their other responsibilities, means that there can be marked differences in the kinds of forest product activities engaged in by men and women, as is demonstrated from the information from southern Ghana in Table 7 (Falconer, 1994).

Table 7: Income-earning forest products activities amongst men and women, south Ghana
 
 

Women	Men
Gathering
Food-wrapping leaves  Snails  Mushrooms  Other forest foods  Seed collection  Fuelwood  Pestle  Palm fruit (some wild)  Plant medicine  Resin	Hunting (game)  Honey  Chewstick logs  Canes  Building poles  Roofing materials  Pestle  Palm wine
Processing
Sponge production  Smoked snails  Plant medicines  Chewsticks  Palm oil  Charcoal	Basket weaving  Charcoal  Akpeteshie distilling  Carving stools/furniture  Mortars  Drying mats  Sleeping mats  Roof tiles  Cane furniture  Fish traps  Canoes/paddles  Crop guards
Trade
Food wrapping leaves  Plant medicines  Mortars  Pestles  Sleeping mats (urban markets)  Charcoal  Akpeteshie  Cane baskets  Chewsticks  Bushmeat (game)  Fuelwood  Sponge  Snails  Mushrooms  Other forest foods  Spices  Other household goods	Food wrapping leaves (Northerners)  Fetish medicines (eg hides)  Mortars  Pestles  Sleeping mats (rural markets)  Charcoal  Stools  Cane furniture
Source: Falconer 1994

The dependence of children on snack foods has already been mentioned. Income from forest products activities can also be important for them. It has been reported from Ghana, for example, that school children are widely involved in basket making outside school hours in order to make the money they need in order to meet school fees and other expenses (Falconer, 1994).
 

STRUCTURE AND CHANGE IN NWFP-BASED ACTIVITIES

People's dependence on, or involvement in, NWFP income-generating activities is affected over time by a whole complex of factors. Some of these are connected with the market, and include market growth or decline, and changing patterns of demand, access to markets, competition and shifts in prices. Others are more related to supply, the products collected varying with the composition and conditions of the forest resource. Much change is associated with changing opportunities as economies prosper, and the growing opportunity cost of engaging in labour intensive activities such as are associated with many NWFP activities.

The discussion is broken down into two parts ? one dealing with activities that produce and trade into local and domestic urban markets, the other dealing with those activities concerned with products destined for industrial and often external markets. The categories overlap in many respects, but are distinguished by certain features of the respective markets and end-uses.
 

Local and Urban Markets

Although most studies focus on products gathered for urban and export markets, these may not be the most important in terms of contribution to rural income and employment, or of quantities involved. Case study material suggests that the bulk of trade in NWFPs is local ? being sold between households or in village or other rural markets. Baskets, mats, household and farm implements, and forest foods tend to feature heavily at this level.

The concentration of small forest products processing enterprises in the rural areas also reflects this geographical orientation. Small processing enterprises predominate where there are factors that favour local processing, such as dispersed raw materials, small markets or high transport costs; where there are economies of small scale, such as in handicraft production; or where subcontracting is more efficient than are integrated operations. The large component of forest products activities in the rural sector reflects the size of rural markets for forest products, and the dispersion of these markets across large areas with a relatively poor transport infrastructure, so that they are more effectively supplied locally. They provide many of the products that farm households require at lower cost than can be achieved with supplies from the modern sector (FAO, 1987).

As they are tied to the rate of change in agricultural activity, rural markets for most non-timber tree products, though very large, in aggregate tend to grow only slowly. Market transactions in forest products grow as use of products that were not previously sold in rural areas, such as fuelwood and forest fruits, becomes increasingly commercialised. Most growth, however, is usually associated with expansion of urban demand. This tends to be based on a number of staple products that formed part of rural use patterns, and which continue to be consumed as people move to the towns ? wood fuels, certain foods, medicinal products, building materials and furniture, packaging, etc.

The expanding domestic trade flows to supply urban markets have given rise to often complex structures of producers, traders, transporters, wholesalers and retailers, that employ very large numbers of people. A study of forest product markets in Iquitos in Peru identified approximately 5,000 vendors of various forest products in the city in 1986, with the number having grown by nearly a quarter over the previous year. These vendors were supplied by a network of wholesalers, large merchants and several levels of buyers and sellers down to the village. In 14 villages in the region that were surveyed, nearly all households gathered and sold some forest products; most selling to trading intermediaries in the village. The main products traded included fruits, leaves and palm hearts; bags, baskets and other handicrafts; thatch and other building materials; meat and skins; charcoal and fuelwood; medicinal plants and fish (Padoch, 1988 and 1990).

Similar patterns are found elsewhere. Table 8 shows the situation in Kumasi, in the Ashante region of Ghana, in which there were on average 650 traders selling forest products in the city's main market during the period it was being studied, with an average of 70 traders bringing supplies in to the city daily. The Kumasi market serves as a hub for trade throughout the region, drawing goods into a central point and redistributing them to other markets. It was found that 68 percent of those interviewed in villages in the region got part of their income from forest-products activities (Falconer, 1994).

As was noted earlier, commercialization can result in the diversion of supplies of saleable forest products from use by the collecting household to the market, and a decline in their rural subsistence use. Growth in trade of forest products also alters relationships and rights. As pressures on a resource rise, traditional rights of use tend to become circumscribed or removed. Some of the longer established trades, that were earlier based on barter and credit-based personal ties of mutual obligation, are increasingly based on short term competitively established relationships of expediency (Beer and McDermott, 1989).

The improvement in rural infrastructure that facilitates the flow from rural to urban areas also exposes rural producers to competition in rural markets from urban producers. Factory made furniture tends increasingly to replace the local artisanal alternative, and plastic mats and basketware displace similar products made from grass, canes and bamboo. A study in Indonesia, for example, found that home-made bamboo umbrellas and wooden clogs were rapidly displaced by mass-produced products using metal and synthetic materials once these became available in rural markets (Hadi, 1986).

The seasonal nature of rural demand, tied to the availability of agriculture-based income, further disadvantages small manufacturing units, which are less able to meet these surges in purchasing than larger enterprises with sufficient working capital to be able to hold stocks. Their practice of producing on a "one-off" basis in response to individual orders also hinders small enterprises in coping with a seasonal flush of demand that would require organised batch or flow-line production (Arnold et al., 1987).

The position of many small enterprise activities tends to be further eroded by internal competition within the sub-sector. Because of very low capital and skill requirements for entry into many forest based activities, it is all too common for many more production units to exist than can be supported locally. The sheer smallness of most enterprises (see Table 9) not only renders them vulnerable to competition but hampers their transition to larger, more viable size. They depend heavily on inputs from the entrepreneur and his or her family. Too small to be able to draw on most rural credit facilities they are seldom able to absorb the costs of moving from the household to larger separate premises and powered equipment. Studies have shown that very small, generally one-person, activities are the least efficient; even small increases in size are often associated with significant increases in economic efficiency (Liedholm and Mead, 1987).

The information from six countries in southern and eastern Africa that is summarised in the box (Arnold et al., 1994) illustrates some of the factors involved. It indicates a very high rate of attrition in forest product enterprises ? enterprise birth and closure (death) rates are both very high. However, as is indicated in the information summarised in the box, different sub-sectors are evolving at different rates.

Case study as well as survey data suggest that the grass, cane and bamboo sub-sector, the largest forest products group in that region, is characterized by low barriers to entry, requiring only limited or widely-available skills and only little fixed or working capital (Townson, 1994). This commonly results in large numbers of very small enterprises, mainly run by and engaging women, most of them single person units, operating from the home, and concentrated in the rural areas.

Rates of growth are slow, with most of the growth in the sub-sector accounted for by start-up of new enterprises. Apart from the small part of the sub-sector that has upgraded its product to craft baskets and related items for urban and external markets, the products of grass, cane and bamboo serve a predominantly low income rural market. This market is likely to show little growth, and be subject to invasion by low cost factory made alternatives such as plastic containers and mats that undercut and displace artisanal products. Internal competition among excessive numbers of producers makes it difficult for individual enterprises to improve their situation.

Manufacture of furniture and other forms of woodwork, in contrast, requires certain skills, and is therefore less easily entered. Acquisition of tools and small machines can permit incremental increases in output and improvements in quality. Access to investment and working capital may therefore impose a further condition of entry, or expansion. Small enterprises can compete with factory-made alternatives across a wide spectrum of situations, and demand for woodwork products exists in urban as well as rural markets. As a result, the sub-sector is commonly characterized by fewer but larger enterprises, dominated by men, many operating in premises outside the home, and constituting an important urban as well as rural activity.

 
 
 
 

Attribute	Jamaica	Honduras	Zambia	Egypt	Sierra  Leone	Bangladesh
Proportion of enterprises (%):
One-person operations	58	59	69	69	--	36
Production at home, not workshop	52	72	81	76	--	--
Rural location:  - enterprises  - employment	88  79	100  100	96  95	80  65	99  96	97  --
Women's share:  - ownership  - labour force	32  30	10  6	12  12	65  31	--  --	(3)  21
% family members in  - labour force (no.)  - hours worked	82  68	51  57	86  --	89  89	(41)  34	73  --
Mean Values:
No. of workers per enterprise	2.2	2.2	1.7	1.9	1.8	3.8
Total investment (US$)	3030	1055	--	--	431	255
Hrs worked annually per worker	990	1247	1205	1712	2004	836
Annual production value per firm (US$)	4979	2536	--	1501	1384	2362
Source: Fisseha, 1987

 
 
 

 
 
 
 

Forest products sector	Botswana	Kenya	Malawi	Swaziland	Zimbabwe	Total
Grass/cane/bamboo	0	25.8	16.7	11.5	11.5	14.4
Woodworking	43.4	61.6	43.0	25.3	32.1	47.8
Other forest products manufacturing	0	55.3	32.0	3.4	10.4	22.9
Forest products trade	8.1	28.2	16.7	30.8	7.6	24.0
All forest products	10.2	38.7	22.2	12.1	13.8	22.8
Source: Arnold et al. 1994

Industrial and Export Markets

A considerable number of gathered and traded forest products are materials for industrial use - e.g. natural rubber and other exudates, fruits of various palms, and various vegetable oils - often for use in external rather than domestic markets. In contrast to the major products traded and consumed domestically, which are mainly staples of everyday use, most forest products gathered for sale for industrial use face uncertain market prospects. The history of such "extractive" products records that, once a product achieves commercial importance, industry seeks to bring production and production costs under control by replacing supplies from wild sources by plantation sources or by synthetics. The greater the success of a product the more likely this is to happen. The market for the forest product then declines, often rapidly, and prices fall to levels at which production is no longer remunerative (Beer and McDermott, 1989; Richards, 1992).

Oil palm and rubber are but two of the many forest products that have been largely displaced by production from cultivated sources. The babassu oil industry, once the largest oilseed industry in the world based on harvest from a wild plant, has declined rapidly since the mid-1980s due to substitution by synthetic detergents and less fatty edible oils (Richards, 1992). Synthetics have also replaced or greatly reduced the market for a long list of other extractive products of the forest that at one time figured prominently among the products of one or other of the tropical forest zones. A history of forest products trade by Penan communities in Sarawak, for example, records a succession of product trades that have been abandoned because of declining prices and rising costs (Brosius, 1992).

Extractive supply systems can also exhibit inelasticities of supply that undermine their viability. When, in the 1970s, demand for babassu oilseed from northeast Brazil was growing rapidly, expansion in production was constrained by the farmers' need to give priority to their rice crop; labour for nut extraction becoming available only when weeding of rice had been completed. Supply therefore proved inelastic in the periods of growth in demand, causing processors to shift from manual to industrial processing. This led to men displacing women as the main income earners from babassu (May, 1992; Richards, 1992).

The main trades in extractive products typically involve extensive production and trading networks. In the case of rattan production in Indonesia, for example, collectors sell to village intermediaries, who are commonly merchants or shopkeepers who provide goods to the collectors, often on credit to be repaid in rattan and other forest products. Village middlemen sell the rattan to river middlemen or trade boat operators who in turn sell to urban buyers who function as the link with industrial or export buyers. Traditionally supply has thus involved networks of local collectors and intermediaries bound by long-term (often debt-based) trading relationships (Peluso, 1986 and 1991). Very large numbers of people are involved in aggregate in this system (see Table 1), and in 1987 Indonesia exported 130,000 tons of rattan, valued at about US$ 200 million. In addition large quantities are used domestically (Beer and McDermott, 1989).

___________________________________________________________________________
Patterns of growth in forest product enterprise activities in Africa

In six countries surveyed recently in southern and eastern Africa - Botswana, Kenya, Lesotho, Malawi, Swaziland and Zimbabwe - an estimated 763,000 persons were employed in 408,000 small enterprise units engaged in the following activities based on the transformation or commercialization of forest products:
 
 

activity	persons	units
grass, cane and bamboo products	321,600	203,300
woodworking	202,500	68,200
other forest product manufacturing	88,400	51,400
forest products trade	150,100	84,300

In the period covered by the surveys, the net number of new forest products enterprises was increasing.  Enterprise birth rates were very high, but so were closure rates, particularly in the early years of an enterprise.  Only about a half of forest products enterprise closures were reported as being for "bad business conditions".  Many of the entrepreneurs who closed enterprises, particularly among those engaged in woodworking, subsequently start new enterprises.

 Employment in those small forest products enterprises that had survived had been growing at 11.5 percent per year, with a much faster rate of growth for woodworking (30.6 percent) than in grass, cane and bamboo (3.1 percent), and in trade (18.5 percent).  At the time of the surveys, about 80 percent of jobs existing in grass, cane and bamboo, and 78 percent in forest products trade, came from new start-ups.  In woodworking, in contrast, 55 percent came from expansion of existing enterprises.

 However, only a minority of enterprises grew at all.  Of those that did grow by adding to the workforce, most grew only by small amounts.  Only in woodworking did a substantial share (30 percent) of the growth in employment come from enterprises that graduated from being very small to intermediate sized enterprises.

 Above average growth rates were more likely to be found in enterprises that are young, and those that started out small.  Enterprises located in commercial districts were more likely to have higher growth rates than their counterparts operating out of the home.  Female-headed forest products enterprises were likely to grow less rapidly than those operated by males.

 Source: Arnold et al., 1994.
____________________________________________________________________________

The evolution of production and trade in rattan in Kalimantan has proved more resilient than that for most extractive products. Collection is just one of the activities practised by producers, being usually combined with agriculture. Collector-trader relationships were traditionally less exploitative than in the Amazon, and access to land and the resource was based on established local rights rather than landlord-tenant relationships. Rattan lends itself to smallholder cultivation and has widely been raised in individual rattan gardens as well as being harvested from the wild resource. The output, in addition to being sold to industry and exported, provides the raw material for a substantial household-based small enterprise activity in the manufacture of rattan carpets and other products (Weinstock, 1983; Peluso, 1986).

Nevertheless, the system has recently come under severe pressure. As demand for other gathered forest products declined in face of competition from synthetics, collectors became increasingly dependent on sale of rattan - demand for which has been growing rapidly. With growing demand for rattan, and the opening up of the forests in order to exploit timber, short term traders entered the market, raising prices to producers and stimulating increased harvesting. The construction of industrial processing plants in Kalimantan has subsequently raised output to levels that observers consider are unlikely to be sustainable (Peluso, 1991).

Broader developments in the region have also contributed to the weakening of traditional production and trading systems for extractives in the Amazon basin. With improved links between the producer and market areas, trade in extractives became less dependent on networks of gatherers and intermediaries. Enhanced land values in areas opened up by roads lessened landowners' interest in extractive income, in favour of other land uses, so that gatherers were widely deprived of access and had earlier usufruct rights withdrawn. Increased commercialisation and access to markets also expanded the range of options open to gatherers. Agriculture became more attractive, and job opportunities in towns led to large-scale migration from the rural areas. The numbers engaged in extractive activities in the Brazilian Amazon have been steadily declining, as people reduce their dependence on low-margin activities and products facing unstable markets, and a way of life that was "often poorly remunerated, lonely and isolated" (Richards, 1992).

It has been argued that, as demands for new forest products are emerging at the same time as others are declining, forest dwellers can maintain incomes from forest products by shifting from one to another. Much attention in recent years has also focused on ways of making trade in these products more remunerative and stable to producers/. However, the balance of the available evidence suggests that the range of products marketable in any quantity is limited, and that, except in the also limited areas rich in exploitable resource, these markets are likely to provide at best short duration incomes, and that their low-input low-output nature makes them poor livelihood systems. Although the typical boom-and-bust sequence may provide significant employment and income initially, in the longer term it can be very disruptive for rural economies, particularly where the trade has encouraged people to move away from more diversified and less risky agriculture-based livelihoods (Browder, 1992; Afsah, 1992; Richards, 1992).
 

DISCUSSION

This section explores in more detail four sets of issues that appear to be of particular importance in relation to socioeconomic benefits and constraints stemming from use of NWFPs. These are (1) issues involving the ability of the poor and women to benefit equitably from NWFP use and activities; (2) issues surrounding diminishing access to supplies of NWFP raw materials; (3) the related issue of the impact of systems of tenure and control on people's access and use of these products; and (4) issues relating to government regulation and support programmes.
 

Equitable Access to Benefits

Access by the Poor

As was noted earlier, the characteristics of forest resources, and of many of the commercial activities based on them, render them accessible to and appropriate for the poor. While forest gathering activities are not restricted to the latter, they do tend to rely on these activities for meeting a greater share of their basic needs. Similarly, most small-scale forest-products enterprises appear to be relatively accessible to the poor, requiring little in the way of capital or skills to enter. As was documented in an earlier section, there is widespread evidence that they do rely more heavily on NWFPs than the less poor.

However, the poor are not always able to exploit the opportunities available from forest-products activities. In a study in an area of Orissa (India) where production and trade in forest products formed an important part of rural livelihood strategies, it was found that the wealthier categories of household collected and sold larger quantities of most products, because they could spend more time on collecting (Fernandes et al., 1988). The constraints the poor face in not being able to devote as much labour to gathering or trading NWFPs as their wealthier neighbours can has also emerged in studies in the Gambia (Madge, 1990) and Viet Nam (Nguyen Thi Yen et al., 1994).

In the Orissa case, the disadvantage of shortage of labour suffered by the poorer households was often compounded by restrictions on their access to the resource. Use of trees on village common land was monopolised by the rich and powerful, forcing the poor to collect from further afield (Fernandes et al., 1988). Declining access by collectors due to changes in land ownership and control has been reported from many situations.

Many authors have also noted that poor gatherers are often exploited by middlemen who control access to the market, or by those who control access to the resource. This is particularly a problem with extractive commodities where collectors depend on intermediaries both for access to markets and for supplies of goods. In the worst cases, such as the rubber tappers in the Brazilian Amazon, concessionaires traditionally took advantage of the virtual monopsony created by their control of the rubber stands to keep the collectors in a state approaching debt bondage (May, 1992; Richards, 1992). Even in less exploitative situations, collectors have little if any bargaining power in their transactions with merchants and traders, or with private owners and managers or public officials who control access to the resource.

However, the widely voiced criticism of intermediaries in many of these situations needs to be tempered by recognition of the difficulties, costs and risks associated with trade in often seasonal and perishable products, production of which is widely dispersed across rural areas with limited transport infrastructure, with little if any access to storage or credit facilities, and serving poor markets in which large numbers of people are seeking to derive a living from the trade. Such market systems may be inefficient and unstable, with sharply fluctuating prices, but they are not necessarily exploitative ? in the sense of intermediaries capturing an exorbitant share of the profit (Padoch, 1988 and 1990)/.

Nevertheless, prices to collectors are often a small fraction of the market value. Combined with unstable markets, this means that some forest product activities provide no more than marginal and uncertain returns to the producer. This is also the case with some of the overcrowded one-person, low-skill processing activities discussed earlier. Thus, while these forest based activities provide some means of existence to the poorest, they may not provide any means for future investment (either in forest gathering or agriculture), or for improving their quality of life. The concentration of the very poor in such low return activities evidently limits the extent to which they can benefit from NWFPs.
 

Participation by Women

The high proportion of women involved in NWFP activities - both as owners and employed - has already been noted. However, where they are concentrated in low return labour intensive cottage industry activities such as mat and basket making, there is a danger that their share will decline. As was noted earlier these very small enterprises have great difficulty expanding, or even surviving, and tend to lose their markets to factory made alternatives.

There are also instances of women being displaced in higher value forest products processing activities. As the lac-turnery cottage industry in India, for example, shifted increasingly from hand lathes to machine lathes women found it difficult to get access to the necessary training and have been progressively displaced and marginalized (Campbell, 1991). Where this is the case, particular gender-focused interventions may be called for in order to correct the discrimination.

On the other hand, women have widely been successful in such expanding activities as food preparation and vending, and in the trading of other NWFPs. There therefore does not seem to be a generic impediment to participation by women in NWFP activities/. Barriers to, and opportunities for, their involvement tend to be situation specific.
 

Access to Raw Materials

Nearly everywhere users of forest products are faced with a decline in the size or quality of the resource from which they obtain their supplies. Timber harvesting in the forest is likely to damage or destroy other components of the forest that provide products for small enterprises. Clearance of land for agriculture and pasture reduces the extent of the resource, distances users from remaining supplies, and is likely to result in more intensive use of what remains. Shortages of NWFPs are a real problem for many.

However, reduction in forest cover and disturbance of the forest structure does not necessarily mean reduced NWFP availability. Bush fallow or farm bush may be as, if not more, productive of foods and other products as the pristine forest; people may adapt by drawing on a wider range of edible plants and animals; or trees valued for food may be protected or planted as the forest is removed. A recent study in Sierra Leone found that the greater part of locally used NWFPs came not from the forest but from fallow and farm bush. The four species used most frequently for construction were all fallow not forest species. Only 14 percent of all hunted or collected foodstuffs derived from forest itself, and 32 percent of the medicinal plants. Moreover, the most used bushmeat species, the rodent "grasscutter", is found only under open cover; it does not occur in the closed forest (Davies and Richards, 1991).

In many situations fallow land, farm bush and even the forest itself is actively managed by local users to conserve or encourage species of value/. The babassu palm in northeast Brazil has long been integrated into local farmers' shifting cultivation system (May et al., 1985), and farmers in the flood plain forests of the Amazon area manage them to favour the economically more valuable species they contain (Anderson and Ioris, 1992). Planted fruit trees appear everywhere at a very early stage in agricultural settlement, and as natural tree stocks diminish the amount and range of planting generally increases (Arnold and Dewees, 1995).

Though the main impetus for this domestication is to meet household needs, it can include species intended for non-farm enterprise activities. Recent work in west Kalimantan has documented a major transformation of forest areas adjacent to communities with improved access to markets, in favour of "orchards" of the wild fruit tree durian (Padoch, 1992; Peluso, 1994). The expansion of household level cultivation of rattan in Kalimantan, first as rattan gardens incorporated into shifting cultivation systems, and then as a smallholder crop, also illustrates the impact that commercialisation can have on encouraging production of products that are suitable for cultivation at this scale and level (Godoy and Feaw, 1991; Peluso, 1991).

People also display considerable adaptability in use of available forest products. As supplies of a particular product are depleted, users switch to using or selling another. For example, in Ghana people are continuously experimenting with and adapting their health practices to reflect new needs and changing availability of medicinal products (Falconer, 1994).

Consideration of whether or not reduction in forest resources is affecting people's dependence on products that those forests contain also needs to take account of the dynamics of change in these patterns of use and dependency. The changes that disturb or deplete the forest may themselves open up new livelihood options that reduce users' dependence on or interest in NWFPs. As logging roads were built in Kalimantan, alternative employment opportunities emerged, reducing interest in rattan collection (Peluso, 1988). The same happened in the Amazon, where the road infrastructure also dramatically enhanced the value of agricultural and pasture uses of forest land, undermining many extractive forest product activities - but stimulating those such as babassu oil-seed production that could be operated on an industrial scale.

Nevertheless, there are many situations in which forest reduction has reached the stage where supplies of needed NWFPs have been severely depleted or even lost. This is notably the case where use was narrowly focused on particular products and species ? hence the decline in so many extractive trades.

For similar reasons, raw material shortages can be a particular problem for processing enterprises. The results from the surveys of small enterprises referred to earlier nearly always show entrepreneur perceptions of raw material supply problems to be higher in the forest sector than in most other sectors (FAO, 1987; Arnold et al., 1994). For the countries reported on in Table 11 it is apparently a problem that is getting worse; with more entrepreneurs listing it as their main problem now than at start-up.

Table 11: Perceived problems faced by small forest products enterprises1
 

Their problems are often worsened by unfavourable forest policies and policy enforcement practices ? which can include unfavourable harvesting controls favouring timber production, exclusive allocation of timber to large users, complicated licensing or auctioning procedures, plus demands for heavy deposits or other insurmountable preconditions, high prices due to state monopolies, and monopoly distribution systems. In a survey among furniture makers in northeast Thailand, for example, problems with forestry regulations were cited by nearly half the enterprises as the main negative factor affecting their operations (Boomgard, 1983).

Some of these distortions reflect the pressures on forest services to produce revenue from forest products disposal; this is more easily achieved from a small number of larger enterprises than from many small ones. However, granting licences to harvest NWFPs to urban traders and other outsiders is likely to reduce availability to local people who depend on the forest for some of their immediate needs. Fees and royalties imposed on collection of NWFPs can make their use uneconomic for the poor; much harvesting is consequently carried out illegally, by those seeking to avoid payment.

The mandate of forest services to protect the forest resource can also interfere with NWFP uses. Many users of the forest are more difficult and costly to control and service, raising issues of how to ensure conservation of the resource and environmental stability. Countries often impose restrictions on private sale and/or transport of particular forest products for this reason. This can have the effect of discouraging investment in production of these products on non-forest land, as well as raising costs to those engaged in production and trade (and making them more dependent on intermediaries). Forest services may also be themselves producers and sellers of particular products; often competing with local producers and traders at what are effectively subsidised prices.

These conflicts of interest between the two sides contribute to the poor relations that so commonly characterizes relationships between forest service personnel and forest users. This in turn undermines efforts to involve users in more collaborative forms of sustainable management of the resource.
 

Tenure and Control

Controls exercised by forest services are paralleled by other changes that restrict or remove users' access and rights to harvest as pressures on a resource increase. The greater part of the raw material supplies that users of NWFPs draw on comes from land that they do not control - privately owned land or land controlled by private concessionaires, state land managed by forest services or other government departments, or common pool land operated under collective control or without any form of control at all ("open access"). They are therefore very vulnerable to changes in land use. Loss of use rights as land owners shifted to other land uses, and withdrew rights of access from the tappers and collectors, has been a major factor in undermining rubber tapping and babassu kernel collection in the Amazon. Lack of security of access to future forest product harvests influences household decisions in favour of the short term results to be obtained from shifting cultivation rather than extractive activities (Pinez-Vasquez et al., 1990).
 

Commercialization and the Decline in Local Controls

The commercialisation of forest products that were previously collected for own use puts pressure on collective management systems and encourages privatisation of the products in demand. The example from Orissa cited earlier (Fernandes et al., 1988) is but one of many where, even when the resource remains in the public domain, as its value rises access to the harvest is effectively captured for the benefit of just the wealthier and more powerful within the community. Though villages in Kalimantan still retain rights to rattan and other forest products on adjacent lands, in practice they are now often unable to exercise these rights in the face of gangs of collectors retained to procure supplies for the rattan processing factories, or uncooperative logging companies with rights to harvest timber on the same area (Peluso, 1991). There are also numerous instances where use or harvesting rights to particular products on private land - for example for fuelwood and post-harvest grazing - are withdrawn once those products acquire significant market value, and industrial rights usurp local rights (McElwee, 1994; Jodha, 1990; Falconer, 1994; Beer and McDermott, 1989).

More profound still have been shifts in control that have removed the resource from forest products production. Table 12 shows the reduction in village common land per capita in various areas of India, as a result of land reform (and encroachment) that has transferred land from communal control to private ownership, and thereby from forest product to agricultural use. A very large part of the village lands on which the poor depended for much of their income, and for fuelwood and other subsistence biomass products has in this way been removed from their use (Jodha, 1990).
 
 

 
 
 
 
 
 

State	Average area of CPRs per village  (ha)	Decline since  1950-52  (%)	Persons per hectare of CPR
			1951	1981
Andhra Pradesh  Gujarat  Karnataka  Madhya Pradesh  Maharashtra  Rajasthan  Tamil Nadu	827  589  1165  1435  918  1849  412	42  44  40  41  31  55  50	4.8  8.2  4.6  1.4  4.0  1.3  10.1	13.4  23.8  11.7  4.7  8.8  5.0  28.6
Source: Jodha, 1990

The erosion of traditional NWFP production and management systems has come about as a result of a long period of political, economic and physical changes. State assertion of control first over the forest resource and then over the land has widely reduced access and rights of usage. At best people were left with usufruct rights, application of which was subject to the whim of the State and its officials. In recent times the reduction in availability of common property resources has nearly everywhere been massively accelerated. Privatization and encroachment, as well as government appropriation, have been the main processes taking resources out of common use. Increasing pressures on what is left have frequently led to its progressive degradation.

Concurrently, traditional methods of access control, usufruct allocation, and conflict resolution have widely become ineffective or have disappeared, undermined by political, economic and social changes within the village and nation. Increasing population pressure and in-migration of outsiders, greater commercialization of the products of the resource, and technological changes that encourage alternative uses of the land, have all contributed to increased differentiation within communities that reduces communal cohesion and uniformity of interest. With the progressive transfer of responsibility for resource management decisions to the central state, many common pool resources are no longer managed in any meaningful sense of that term by those who use them. In his study in the dry regions of India, Jodha found that of the communities that in 1950 had exercised controls such as rotational grazing, seasonal restrictions, and use of watchmen, only 10 percent had such controls in 1980, and use of fines, taxes, and fees had ceased altogether. Much usage is now of an unregulated "open access" nature (Messerschmidt, 1993; Shepherd, 1992; Jodha, 1990; Poffenberger, 1990).
 

Collective, Private or State Control

The shift away from collective local control is now so heavily entrenched in many countries as to make further privatization or appropriation seem either inevitable or desirable, or both. However, recognition of the continuing importance of common pool resources to many rural populations has stimulated re-examination of policies and practices that lead to the erosion and breakdown of local collective systems of resource management. The strong thrust towards bringing use of common resources under private or government control has been widely influenced by the thesis of the "tragedy of the commons", which argues that the increasing pressures on individual users prevent effective cooperation and group control (Hardin, 1968). However, much usage of common property resources is of an "open access" nature, characterised by unrestricted entry and unregulated use rather than by collective management. Verdicts of breakdown of common property resource management have all too often been erroneously passed on situations in which the deterioration actually came about due to unregulated use under an open access regime.

This misunderstanding has been compounded by a tendency to overlook the factors that encourage collective action, and the self regulating capabilities of groups of users, and the reasons why the alternatives of private or state control may themselves not be sustainable or efficient. Because exclusion from a common property resource is difficult, it may not be feasible to privatize it. Privatization, by transferring control of the resource to a limited number of individuals who thereby acquire the social and legal sanction to exclude others, is likely to exacerbate the problems of the excess of population without access to private property (Bromley and Cernea, 1989). Private use can also lead to overuse and degradation. Equally, the State may not be able to control, manage or prevent degradation to a resource it has expropriated (Berkes et al., 1989).

There are, therefore, good reasons for questioning the discrimination against collective management, and in favour of attempts that have been made to reverse this trend. However, interventions to strengthen or establish contemporary systems of collective local management of forest resources have to date had only limited success. When local institutions have broken down under the pressures of change, it is not to be expected that new village institutions capable of controlling resource allocation and use can be created easily. The low returns and high social cost associated with trying to control common property resources may prove unacceptable to users, to the point at which they prefer to leave it to the state to manage them.

Commercialisation puts particular pressures on collective control systems. Though the resource acquires a higher value as market demand emerges, which in principle could increase the incentive to conserve and manage it (Peters et al., 1989), in practice it usually heightens pressures to exploit it now. Given the relatively long production periods of many NWFPs, uncertainties about future prices, and the often uncertain tenure situation of the users, many will understandably opt for the short term returns that can be obtained from increasing current exploitation levels (Pinedo-Vasquez et al., 1990). In such situations, conflicts within the community, and with outsiders, about use of the resource are likely to increase, and become less easily resolved. As increased resource and product values make the transaction cost of privatising worthwhile, this is likely to undermine the incentive to maintain collective control further.

However, successful local control systems do exist or have evolved. In a recent review of information on the subject, McElwee (1994) argues that one of the key factors is being able to control the intensity of commercialization, so that it does not overwhelm the resource or the institutions governing the resource:

Some of the more successful adaptations are ones that combine features of both collective and individual control. The concept of separate land and tree tenure, whereby ownership of the produce of a tree can be separated from rights of use of the land on which the tree stands, is widespread where land is still communally controlled. Across much of Africa, for instance, this provides farmers with sufficient assurance of security of access to tree products to stimulate tree planting and management on land on which they only have usufruct rights (Fortmann, 1985; Warner, 1993, Shepherd, 1992). Peluso (1994) has described, for an area in west Kalimantan, how systems to control access to tree produce have changed over time in response to political ecology and market forces. Most recently, with a major expansion in the market demand for the fruit durian, and a consequent increase in its production, a system dominated by groups of descendants from the original planters of long-living fruit trees is being replaced by more individual control. This gives greater ease of management, and avoidance of potential disputes about distribution of proceeds, but diminishes the access to tree benefits of those in a family that do not have producing trees of their own. But the two systems co-exist, with choices between the two being determined within each descent group.

However, in most situations the more pronounced trend is towards shifting the focus of production of NWFPs on to individually controlled land. Managed fallow, farm bush, home and forest gardens and compound farms, and other trees incorporated into farm systems are becoming increasingly the main source of many NWFPs. The products most at risk of depletion are then the ones that are not easily domesticated in this way. Many medicinal plants fall into this category.
 

Regulations and Support Programmes

Small enterprises in general have to operate within a policy environment that is oriented towards large modern sector industry. Thus, small enterprises often face subsidized credit allocation regulations and tax concessions aimed at firms above a particular size. Even where such overt restrictions do not occur, licensing and other burdensome bureaucratic procedures tend to exclude them from access to available incentives or assistance (Liedholm and Mead, 1987). A widespread issue is therefore to achieve a more neutral policy environment, in order to remove inadvertent bias against the small enterprise sector.

In designing programmes to provide support to small forest product enterprise activities, it is important to recognise that there are different potential target groups with different needs and opportunities. Those in the process of starting up face different problems and constraints than those seeking to expand. Those new entrants driven by supply side forces, as people search for activities where they can sustain themselves, face different issues than those who are responding to market opportunities. Among those enterprises that are growing, those seeking to expand from a one-person beginning have different needs for assistance from those that aspire to graduate to larger scales of operation.

The large numbers entering many non-timber tree products activities ? and small enterprise activities generally ? suggests that there is no shortage of prospective entrepreneurs, and raises questions as to the volume of resources that should be allocated to encouraging yet more new entrants. The high rate of attrition among start-ups suggests that support at this stage should focus on helping potential entrepreneurs identify prospective lines of activity that are compatible with their capabilities and experience. It also raises the possibility that it could be more efficient to concentrate limited resources for support services mainly on helping established businesses.

The concentration of new entrants in many countries in low-return tree product activities that can offer no more than marginal, unsustainable livelihoods, presents particular issues. Support to such activities once higher return or less arduous alternatives emerge could impede the emergence of better livelihood systems for the participants. That being the case, it may be more fruitful to help people move into other more rewarding fields of endeavour rather than seeking to raise their productivity in their current line or work. Care needs to be taken in such a case to ensure that not only current income levels but also future growth prospects are indeed better in the alternative product lines to which people are being encouraged to move.

With regard to support of established enterprises, a credit programme that provides small amounts of working capital may be of great help to a very small enterprise seeking to grow a little. As discussed earlier, this type of incremental growth is often associated with substantial gains in efficiency. Enterprises seeking to graduate to a larger size, by contrast, would generally need more complex forms of assistance, including help in searching out new markets, in management skills and in production control. It is generally more expensive to offer this type of assistance, but benefit/cost ratios can still be favourable if the result is substantial growth in employment, in efficiency and in incomes earned by the clients.

With such diversity among the enterprises and those engaged in small no-timber forest products activities, support programmes need to be based on a sound understanding of the patterns of growth in particular sub-sectors and situations, and a clear focus on the needs and possibilities of the particular category being targeted. Generalised approaches, intended to assist everyone, are unlikely to be effective.
 

CONCLUSIONS

NWFPs constitute an important part of the livelihood system of huge numbers of rural, and many urban, households. Consideration of their role should form part of analysis and planning related to nutrition, health, food security, agriculture and rural development as well as forestry.

Given the often rapid and substantial shifts in patterns and levels of dependence on particular products and activities that occur, it is important to be able to identify the direction of change and the factors responsible for change, and to be able to distinguish between those products and activities that are likely to remain viable and grow, and those that will decline or disappear.

In pursuing this aim it is useful to distinguish three different basic situations within which small enterprise activities are prominent:

• in economies where population is growing faster than per capita incomes, growth in non-farm employment reflects its function as a sponge