Key notes

Smallholders and Community-Based Management
of Farm Animal Genetic Resources

Wolfgang Bayer,^[1] Annette von Lossau^[2] and Antje Feldmann^[3]

Abstract

In this paper, it is argued that smallholder farming is just as rational as large-scale farming, but subsistence rather than production for markets is of greater importance. Smallholder farms are often multisectoral, animals have multiple functions, and enterprises are often severely resource constrained. A review of productivity assessment presents different approaches (productivity per animal, per unit weight of dam, per unit of available forage, multifunctionality and productivity) and concludes that there is no established methodology to assess multifunctional productivity. An attempt to relate the different functions to breeding objectives showed that, for some functions, such as manure production, mere survival of the animals is of prime importance and that breeding objectives depend on the mix of functions. Four approaches to managing animal genetic resources are outlined: 1) the acquisition of new breeds or species, which depends strongly on outside supply; 2) opportunistic breeding, where nature and consumption habits act as selection pressures; 3) deliberate breeding of large stock, where reproduction may be more important than selective breeding; and 4) selective breeding for particular markets, where the market demands can change rapidly. It is concluded that management of AnGR needs to be seen in the wider context of livelihood systems, that marketing tools, such as carcass grading, need to take adaptation to the environment into account, and that the possibilities to market smallholder livestock as a speciality need to be explored.

Introduction

The first question that needs to be answered is: why a keynote address on smallholders when another keynote deals with "community-based management of animal genetic resources, with special reference to pastoralists" (Köhler-Rollefson, this volume)?

One reason is the great importance of smallholders. In southern Africa more than 90 percent of animal keepers are classified as smallholders. Another reason is that smallholders have to work under specific circumstances, which differ from those of pastoralists, whose management of AnGR is well documented by Köhler-Rollefson. Smallholders by definition do not rely exclusively or predominantly on livestock and therefore have to organize the management of AnGR in a different way. A third reason is that, right from the beginning of agricultural science in the early nineteenth century, agricultural science and smallholders had a rather ambivalent relationship. This ambivalent relationship is still evident in policy documents in southern Africa and does not facilitate meaningful support for smallholders by extensionists and agricultural scientists. In recent years, great progress has been made in some countries such as Zimbabwe with respect to supporting smallholders, but approaches such as the participatory extension approaches (PEA) or participatory technology development (PTD) concentrate on soil and water conservation and crop variety development. Animal husbandry is hardly tackled.

To be sure, smallholder farming should not be glorified. It is appreciated that smallholders have a harsh life and that their importance will decrease if there is a broad increase of the standard of living of people in a country. In times of economic crisis, however, the importance of smallholder agriculture increases, as shown recently in Eastern Europe and the former Soviet Union. Supporting smallholder agriculture may not increase food production for export markets to the same extent as large-scale commercial farming, but smallholder agriculture plays an important role in giving rural people access to food, and its continuous neglect can have severe negative social consequences.

In this paper we discuss:

the characteristics of smallholder farming systems;

the functions of animal husbandry in smallholdings;

problems with measuring animal productivity under smallholder conditions;

the strategies of smallholders for managing AnGR; and

conclusions with respect to AnGR projects and programmes in smallholder areas.

Smallholder farming systems

Personal experience reveals that some extensionists, policy-makers and animal scientists in southern Africa still are convinced that smallholder subsistence-oriented animal husbandry is backward and unproductive, that the traditional farmers are irrational and that their way of animal farming has to be replaced by modern, intensive, market-oriented production systems.

Ascertained through various social, anthropological and economic studies of smallholder farmers in many parts of the world, it is now widely accepted that:

Smallholder farmers are not less efficient than modern dairy farmers or ranchers in using natural resources, including AnGR. However, their aims are not to produce as much as possible for the market and to maximize profits. Smallholders rather aim at meeting the needs of their social group, above all, to ensure subsistence. Often they try to achieve this by keeping a multispecies herd. For example, specialization in chicken fattening or dairy farming is reasonable for them only if the production risks are low and if the family’s needs can easily be met on the market.

The household economy of smallholders is often multisectoral and the farm income is supplemented by income from handicrafts, trade, wage labour, remittances or pensions. Family members are bound to help relatives and neighbours in need, and can expect the same in return. The demands of such a social security network on the farm family’s time and resources may prevent specialization, e.g. in cattle fattening.

Smallholder farmers often live in marginal areas with poor infrastructure and difficult access to the market; therefore, external inputs such as feed or fertilizer are more expensive than in more easily accessible areas.

Smallholder farming is, almost by definition, strongly resource-constrained. Feeding and breeding strategies have to take these constraints into account.

Animals are often multifunctional. Besides producing food (milk, meat, eggs) or raw material (wool, hair, skins) and generating cash income, other functions of livestock keeping include:

- providing draught power and transport;

- providing manure as fertilizer, building material or fuel;

- controlling insects, snakes and other pests;

- providing a means of saving;

- fulfilling social obligations.

Livestock can contribute to group identity or prestige, as shown in the cowboy culture of the North American West, the dairy culture in the European Alps, sheep and goat milk producers in the Mediterranean countries or traditional pastoralists, agropastoralists or smallholders in Africa. With respect to sociocultural functions, the Thanksgiving Turkey in the United States, the St Martins’ Goose in Germany and the various social functions in African societies for which livestock is needed can be listed.

The typical smallholder does not exist. Smallholders and their breeding strategies depend not only on natural and socio-economic conditions, but also on the abilities and interests of the members of a farming family. Therefore, a great variability among smallholder farmers with respect to animal breeding, even within a single village, can be expected.

Assessing productivity of smallholder animal husbandry -
a contribution to breed evaluation and breeding objectives

Practising animal breeders have always made productivity assessments of their animals. "Scientific" animal breeding started about 200 years ago, when animal breeders, concerned with the "poor quality" of breeding stock, sought ways to improve the performance per animal and became part of the then emerging agricultural sciences. This meant that meat-producing animals were bred to grow more quickly. The idea was that faster-growing animals would use a greater part of feed for production and less for their maintenance requirements. In many ways, this approach still prevails in intensive modern agriculture (see, e.g., Arthus-Bertrand and Raveneau, 1994) and there are still many animal breeders who like large animals, despite the problems that arise with dystochia, susceptibility to stress, imbalance of muscle and skeleton growth, high nutrient and veterinary requirements, and poor meat quality.

Some 50 or 60 years ago, Bosma in South Africa pointed out that smaller animals can be equally and sometimes even more productive than large animals, if productivity is calculated not per animal but rather per 100 kg live weight of the mother. This argument was in favour of smaller breeds. Fertility and mortality were found to be the main determining factors for productivity measured in this way. In the late 1970s, the International Livestock Centre for Africa (ILCA, now part of ILRI, the International Livestock Research Institute) developed a three-stage productivity index based on a model for meat (beef) production in a ranching system (ILCA, 1990). It was first claimed that weaned calf weight was a good indicator for total herd productivity and then the weaning weight was calculated per potential mother, per kg post-partum weight of the mother, and finally per kg metabolic weight of the mother (kg^0.75).

When these methods were applied under village conditions in herds kept by smallholders or pastoralists, it was soon realized that this comparison was not fair, since animals on a ranch are rarely milked, whereas milk can be an important part of the diet of smallholder or pastoral families. ILCA therefore modified the productivity index and converted milk offtake into calf weight. Then an intense debate started about which factors to use: does a calf need 11 kg of milk in order to put on an extra kg of live weight, or does it need only 8 kg? Should we take into account that a calf growing 1 kg per day needs one day of milk to cover maintenance requirements, whereas a calf growing 200 g per day uses up 5 days of milk for maintenance to gain a kg of live weight? Or should we calculate the nutritive value of meat and milk produced per mother or per herd and express it as total food production?

These arguments were never completely resolved, and perhaps for good reasons. It depends whether one wants to compare "biological" productivity (breed comparison as a basis for a breeding strategy), which requires a standard procedure, or whether one wants to compare livestock systems, which requires the accommodation of differences resulting from different strategies for managing and using livestock. In Botswana, de Ridder and Wagenaar (1986) compared smallholder mixed farming with ranching and found that smallholder cattle keeping is more productive, not only per hectare but also per cow. This is because smallholders use their cattle not only to produce meat and animals for sale, but also to produce milk and draught power and operate at a higher stocking rate than ranches.

Another approach to assessing animal productivity is to look not at productivity per animal, but rather at productivity per unit of forage. In conventional high-production systems, the nutritional needs of the animals are calculated, and then feed of the required quality and quantity is obtained. But what if feed resources are limited? This approach is followed by pasture and range scientists, e.g. when discussing carrying capacity of the pasture, and it was hypothesized for cut-and-carry systems by the Animal Production Systems Group at Wageningen Agricultural University in the Netherlands (e.g. Zemmelink, 1980; Zemmelink, Brouwer and Subagiyo, 1992; Bayer and Zemmelink, 1998). The reasoning for optimal production from limited feed resources goes as follows: at low stocking rate or when forage is abundant, animals tend to select the best parts of the forage only, and the feed intake and animal production are high. However, when the number of animals is low, overall production is also low. When more animals are kept, the animals eat more forage of lower quality, so that the performance per animal will be lower. If, for example, in the first case, only the best 10 percent of the forage is used and then the number of animals is doubled, then the performance per animal will decrease to 90 percent of the previous level but, because there are more animals, the overall production will be substantially higher.

A further increase in animal numbers will lead to a fairly linear decrease in performance per animal, whereas - in graphic terms - the performance per hectare or overall production follows a parabolic curve. The peak of that curve indicates the optimal production per hectare, in the case of grazing animals, or the optimal level of forage use, in the case of cut-and-carry systems.

In grazing systems, the optimal production of beef per hectare is achieved at a stocking rate that allows only half of the maximum live-weight gain possible per animal (Jones and Sandland, 1974). In other words, if a steer can gain 150 kg/year at a very low stocking rate, that same animal will gain only 75 kg/year at the optimal production per hectare. The individual animal may look thinner, but the overall production is greater with several thin animals than a few fat animals.

In the case of cut-and-carry forage, Zemmelink, Brouwer and Subagiyo (1992) presented an example from a smallholder cattle system in Indonesia. The optimal production for live-weight gain was achieved when the best 33 percent of the available forage was used. The maximum gain per animal that could be achieved was 450 g/day, whereas the gain at optimal forage use was roughly 250 g/day. When a second function - manure output - was included in the calculations, the optimal level of live-weight gain and manure production was achieved when 70 percent of the available forage was used. At this level of utilization, the daily gains per animal went down to 100 g.

The genotype of the animal was not considered in this research, but it is clear that the ideal genotype for a low degree of forage utilization and high daily weight gain would be quite different from the ideal genotype for a high degree of forage utilization and low daily gain. Productivity studies that incorporate multiple functions of animals are still rare, and even more rare are those studies that take the consequences of multiple functions into account when discussing the type of animal that fits best into such systems. An exception is the recently published thesis of Workneh Ayalew (2000).

Functions of animal keeping and related breeding objectives

As already pointed out, animals in smallholder farming have more functions than to produce food and raw material. For many families growing crops on marginal land and for specific types of crops, manure can be very important. For the purpose of manure production, mostly large animals (cattle, donkeys) are used. Animals don't need to grow fast, but they need to be able to survive on often poor-quality forage. If draught is an important function (provided primarily by cattle and donkeys), animals need to have sufficient power during the time they are used for traction but, at other times of the year, they only need to survive but not necessarily grow. To fulfil the function of savings account and capital formation (all animal species), it is important that the type of animal being kept does not require much management input or veterinary care and can be kept at low cost. If an important function is to produce food for the family, the choice of animal and breed will depend, among other things, on infrastructure. If no refrigeration is available, it is advantageous for a smallholder family to keep different sizes of animals for different purposes. A chicken is enough for a normal family meal (or perhaps two chickens or a goose or turkey for a large family). For larger gatherings, a local pig or sheep or goat may be large enough. Only for special occasions, such as a wedding, funeral or graduation ceremony, would an ox be slaughtered. Otherwise, there would be too much wastage.

Smallholders are not averse to selling livestock if they need cash. However, to be able to understand how production for sale fits into smallholder livestock systems, it is important to understand local markets. If chickens are fattened purely for sale, i.e. on a commercial scale, a farmer with access only to the local market will find it difficult to sell 300-400 birds within a short period. It is better to take an approach that allows more staggered marketing in smaller quantities and, for this, it may be more appropriate to keep chickens that can brood, preferably also under extensive free-ranging conditions. Chicken breeds used as commercial broilers, which have the potential to grow faster than local chickens, will not show their potential under such conditions. In other words, they are usually poorly suited to smallholder animal production systems.

Numerous other functions of livestock may be specific to particular areas. In South Africa, for example, some smallholders keep geese as "watchdogs". For this purpose, different traits are required than fast growth. For social or cultural purposes, chickens, goats or cattle of a particular colour may be required. These are rational, non-commercial objectives that are reflected in smallholder practices of animal breeding. The functions of animals and the related breeding objectives can change with circumstances. Close to urban areas or major roads, free-ranging animals may become a less attractive proposition and stall keeping may offer opportunities (e.g. better control of animals) and create new constraints that need to be addressed, such as higher labour requirements, more organizational inputs to obtain feed, higher animal concentrations and higher incidence of certain parasites (worms, fleas, lice).

In smallholder farming areas - given the often poor quality or seasonally low quantity of feed, the high disease pressure, the poor infrastructure and the high costs of veterinary services and other external inputs - the animals need to be adapted to the environment and capable of coping with sometimes very adverse and low-input conditions. When the merits and demerits of breeds are assessed, these conditions must be taken into account - and not the high performance of certain breeds under intensive production conditions, where animal disease, feed quality and other environmental factors can be controlled. This reasoning should also have a strong bearing on breeding objectives and organization. A summary of functions, desired characteristics and breeding objectives is given in Table 1.

If the multifunctionality of animals for smallholders is accepted as valid, questions that should be pursued further are:

What are the consequences for breeding objectives and for the requirements for the organization of breeding operations?

How can the multiple objectives of animal keeping be measured/evaluated/expressed as a productivity index, so that they can be translated into breeding objectives?

How can policy-makers be influenced to accept subsistence as a valuable aim for animal keeping?

Breeding management strategies

With respect to breeding management, it appears reasonable to start by trying to understand the present practices rather than prescribing a scheme from above. Unfortunately, smallholder breeding practices have not been described to any great extent in the literature. This stands in contrast to commercial breeding plans or, to a lesser degree, the animal-breeding practices of traditional pastoralists. Some pastoralists keep herds that are so large that they can select breeding stock from within the herd and often favour particular breeds (see Köhler-Rollefson, this volume). In contrast, smallholders with their much smaller flocks or herds rely heavily on formal or informal exchange and transfer of breeding stock or genetic material between households, villages, government farms and their animal enterprise, or between the commercial and communal sector.

The following smallholder strategies for the management of AnGR by smallholders can be distinguished:

acquisition of new animal breeds or species;

opportunistic extensive breeding, mostly of small stock;

selective breeding, mostly of large stock;

selective breeding of livestock for more commercial purposes.

Table 1. Functions of animals, desired characteristics and breeding objectives

Acquisition of new animal breeds or species

On occasion, smallholders take up the keeping of new animal breeds or species. Examples are turkeys, rabbits, geese, introduced pigs or more traditional animals such as cattle, which they had not kept previously. They may introduce these animals into their farm as more speculative ventures for the market, to produce food for the household, as "watchdogs" or as a hobby and for pleasure. The keeping of these new animals may last for only a limited period of time, or may increase towards a more sustainable enterprise. Within any given village, such species are not common, at least initially.

Acquisition of new animals is normally the activity of an individual, who obtains the animals from a commercial breeder or a government farm. A prerequisite for the sustainability of such a venture is that appropriate types of animals remain available. A good example could be observed near Pieterburg (South Africa), where a woman had started turkey keeping about two years ago (Bayer, 2000). She had bought a pair of birds from different origins and now has about 20 birds. She is aware of the danger of inbreeding and, when the birds show signs of increased incidence of deformation or reduced thriftiness, she will buy new birds. She can proceed with this system as long as a landrace of turkeys is available and not only the overly heavy commercial birds bred for intensive systems. Trying out new animal species is an innovative venture, although innovations are more likely to be in the field of animal husbandry rather than animal breeding. The lady herds her turkeys on to green pastures, so that they get different types of food. Green turkey feed comes also from the greengrocer in town, where her husband (who works in town) collects leftover leaves and waste.

Starting to keep new animal breeds or species is a form of speculation, which can lead to economic success but also poses some risks, like any speculation. As a development activity in smallholder systems, such an approach should not be ruled out, but it must be kept in mind that, with the numbers of animals kept in smallholdings being so small, the management of genetic resources depends on an outside supply of animals, from either government farms or commercial suppliers. If these sources fail, it is difficult to imagine that these ventures will survive.

Opportunistic extensive breeding of small stock

Small stock such as chickens or goats are often kept as a low-input savings account and occasional source of food. They may also have a social function related to some cultural ceremonies, for which a particular type of animal is required. Nevertheless, breeding is often done in a rather opportunistic way, with little selection for higher growth rates or more eggs. Breeding is rarely deliberately controlled.

When the West African dwarf goats in rural areas of central Nigeria were studied, it was found that the flocks kept by the farmers (on average, five animals per household) contained hardly any entire males above one year of age, even though at birth and up to three months of age the sex ratio was balanced (Bayer, 1986). Sires were the very young males (aged six to nine months). Animals were frequently sold or slaughtered for food, and heavier young males were preferred. As adult males of these dwarf goats can have a very strong smell, they are not particularly popular. This practice served as a selection against fast growth but, at the same time, for early sexual maturity. As the son mates with mother and sister, our initial suspicion was that inbreeding would be a problem. As it turned out, frequent exchange of animals through buying and selling, gifts and transfers associated with marriages minimized this problem. Furthermore, after the cereal harvest, the animals were allowed to roam freely in the village area and this also gave a good chance for the small flocks of each household to mix. In this extensive system, mortality among young stock is high. This maintains a high degree of breed adaptation to a harsh and low-input environment.

In the case of chickens, cocks may be selected according to colours and, on occasion, smallholders will buy larger commercial cocks and cross them with the local chickens. However, because these cross-bred chickens are more susceptible to disease and have higher feed requirements, they usually do not survive for long under extensive free-ranging conditions.

In some areas, goats are milked and can be small and fairly efficient milk producers. Dairy goat keeping is often women’s business. The experience with cross-breeding of indigenous x commercial dairy goats, such as the Saanen goat, is sobering. Although cross-bred goats can - under improved management - produce substantially more milk than indigenous goats, the problems posed to smallholders by the higher feed requirements, higher disease susceptibility and scarcity of appropriate bucks are not easily solved. Rischkowsky (1996) reports on the failure of such a goat cross-breeding scheme in Malawi and Workneh Ayalew (2000) indicates that total net benefits (taking into account not only production of milk and meat but also other factors such as risks) are greater for indigenous goats under improved management than for cross-bred goats in Ethiopia.

For extensive keeping of small stock, promising interventions are not so much on the breeding side but rather on measures to increase survival of young stock, e.g. by somewhat improved shelter for chickens, by vaccination against some diseases (Newcastle disease in chickens, for instance, and peste des petits ruminants in goats) and by improving forage supply. These measures will have a much bigger impact on total productivity of livestock than improved breeds. Breeding activities will become beneficial only when animal health problems are solved and forage is available in sufficient quantity and quality.

Selective breeding of large stock

Large stock in southern Africa can be more or less equated with cattle. Camels or buffaloes, which are of great importance in other parts of the world, are almost completely absent, and horses are a special case for breeding. Donkeys - although they are very useful throughout southern Africa - are generally left to themselves for breeding. A stallion may be castrated only if he becomes a nuisance.

Cattle have a multitude of functions, which differ in their importance from site to site. Breeding objectives will differ accordingly. Cattle serve as a savings account - the large bills of livestock currency; they produce manure, provide draught power, are sold for income, are important for culture, provide meat and milk as food and skins and leather as raw materials.

Many peoples in southern Africa have a pastoral or agropastoral past and the ideal is that each kraal has its own bull. In the past, bulls were selected from within the herd. The bride price (lobola) is a way of exchanging genetic materials between herds. If herds are below the optimal size for a bull, it is not uncommon to use bulls not only for breeding but also for draught purposes. During a recent visit to the Eastern Cape Province (South Africa), a close look at an "ox-team" of six ploughing animals revealed that there were three bulls, one was a cow and only two were, in fact, oxen. This practice used to be common among smallholders in parts of Europe. In parts of India, bulls are castrated only after they have spent one season in the yoke (Bayer, 1994).

Another way of organizing reproduction, especially if herds are too small to make it worthwhile to keep a bull for breeding purposes only, is to practise communal grazing. If some of the herds with bulls graze together with herds without bulls, the bulls will not miss a cow in heat. Sending out bulls to communal grazing can be an act of good will towards poorer neighbours. Another way is the use of artificial insemination (AI), but this is restricted to dairy animals, as the semen are from bulls selected for progeny with high milk yields. Interviews with farmers in communal areas in South Africa indicated that they had only just begun to keep dairy cows or that this was an intention rather than a reality. Although AI is currently not used to any great extent, the discussions with farmers showed that they were well aware of these possibilities.

In western and central Europe, smallholder animal husbandry was important until the 1950s and it is still important in eastern Europe. Here, people developed a range of possibilities to manage their livestock and breeding. Where people kept livestock as a sideline, e.g. in mining areas, a village herd was formed and the village herder - a respected person - was obliged to keep a bull or buck or ram, depending on the species being kept, and had the right to charge for successful mating. Now, the village herds and herders in western Europe have a value only as keepers of tradition and can contribute to the tourism value of the area. In eastern Europe, however, for example in Bulgaria, this practice was revived after the collapse of the centrally planned economies. Here, sheep and goats are the prime species for village-level herding and breeding.

In some countries, a few farmers in a village keep a bull and charge for its services. In the foothills of the Himalayas in northern India, where buffaloes are kept as dairy animals (Bayer, 1994), the keepers of buffalo bulls are landless people. In some villages, in return for supplying the services of their bulls at a moderate rate, they gain free access to feed and pastures for their animals. In other villages, they charge somewhat more but have no privileged access to feed and pastures. The venture is economical only if the bulls can depreciate over a long time, say six to eight years. If it is desired to use a bull for a shorter period, e.g. to avoid inbreeding, there is a strong argument for subsidies. In Germany, a township or village council was required by law to ensure that a town or village bull or buck was available for the farmers. When AI became more popular, many township councils opted for subsidies of AI.

In some densely populated areas, important functions of large stock are to provide manure and draught power (and milk, as a source of regular income or revenue from cattle keeping). In many cases in Europe, the small-scale farmers were so short of resources that they usually did not raise the offspring of their cattle. In parts of Germany, this was the case until the early 1960s. Calves - male and female - were sold at about six weeks of age as young vealers. Replacement stock was bought on markets in better-endowed areas. Smallholders in northern India are using the same system today. In southern Africa, communal farmers frequently buy in stock from commercial farms.

What kind of breeding strategies can be developed under such circumstances? None. Farmers have to be opportunistic and cannot be choosy. Cattle that are bought in should be reasonably priced, they should not be too demanding with respect to feed and veterinary care, but they must also fulfil other functions, including providing prestige. Many farmers are well aware of the numerous functions of their animals and the need to bring in animals that are adapted to the local conditions. However, role models still have an important influence on their decision-making about selection of breeds. If the role model is that a "modern" farmer keeps a good dairy cow, farmers in some areas still try to keep "modern" animals (i.e. of potentially high-yielding breeds), although they are aware of the difficulties of keeping such cows.

Selective breeding of livestock for commercial purposes

Smallholders also appreciate their livestock for its commercial value. In some countries markets are more informal, and farmers sell to other farmers or primarily to consumers. Other markets are highly regulated, commercial farms dominate formal markets, but smallholders in Namibia, South Africa or Botswana can participate, provided that the products meet market requirements. Increasing the participation of smallholders in formal markets is an expressed policy, for example, in South Africa.

Examples for involvement of smallholders in commercial markets are cattle and small-stock auctions for export, the much-supported small-scale poultry enterprises in South Africa, and Karakul pelt production and wool.

Breeding strategies for such enterprises differ very little from those on commercial farms. In the case of poultry, chicks are bought from commercial suppliers; in the case of cattle, sheep and goats for meat, the bucks, rams or bulls come from studs or nucleus herds and the types of animals have to meet market demands. Generally, large-framed animals are favoured and those large-framed animals usually have higher nutrient requirements and are more susceptible to disease and stress than smaller-framed indigenous breeds. As long as the animal classification system favours the larger-framed animals, it is rational that smallholders try to keep these animals if they have commercial ambitions.

The usefulness of some commercial ventures should be questioned. The whims of the world market are unpredictable. In the 1970s, Namibian commercial and smallholder farmers together could sell some 4.5 million Karakul pelts annually. The sales are now down to about 200 000 pelts (Karakul Breeders’ Society of Namibia, 2000).

In the case of poultry, there are economies of scale and commercial farms, producing 100 000 broilers a year, can buy feed and animals more cheaply. An enterprise producing 300 broilers in one round will not be able to compete with these large enterprises on the urban market. For a village market, 300 birds within a week or so are far too many. It is therefore doubtful whether such an enterprise would be economically viable. Here, niches have to be sought where smallholders have a comparative advantage, such as a market for special products on the urban market, or smallholders have to be content with the local village markets.

Some important questions in this context are:

Not all breeding strategies are equally suited for outside support. Which criteria are necessary to decide on suitable outside support for breeding? What kind of support may have higher priority than support for breeding operations for which scenarios?

Indigenous knowledge is the backbone of the present animal-breeding practices. However, especially in South Africa and Namibia, "modern" commercial animal breeding has a strong influence on breeding practices among smallholders. How can the two knowledge systems be combined to the advantage of smallholders?

With respect to commercial ventures there are economies of scale and ventures such as broiler production in large-scale enterprises near large markets (cities) have an advantage. Which are possible criteria to determine a comparative advantage for smallholders? Which support is necessary?

Conclusions for development activities related to AnGR

As the situations of smallholders are very diverse, meaningful interventions will vary accordingly. The guiding principle should be that we should always try to think from the viewpoint of the customers. This approach will probably contribute more to food security than trying to follow national policy targets, such as more production for export. The world market for animal products is very volatile and risky. The generally low-risk horizon of smallholders must be taken into account when designing interventions in community-based management of farm animal resources.

The current breeding strategies and breeding objectives of the smallholders should be clear before support is given to any specific type of breeding operation or suggestions are made for improvement. Extensionists and farmers should understand the consequences of changes in breeding operations or breeding objectives. With due caution, the following types of intervention for different settings and strategies among smallholder livestock keepers are suggested:

The introduction of new animal species is generally not a major thrust in improving animal husbandry. The main direction in this intervention should be to identify suppliers of appropriate breeding stock. If the keeping of a particular species is rare among smallholders, they need access to a commercial supplier of a landrace type of animal. If the keeping of a species becomes more widespread, an interchange of breeding stock between villages and even regions becomes a viable option. However, advice on livestock husbandry, such as disease prevention, better feeding or appropriate animal housing, may be more effective in improving the economic outcome of the operation than breeding interventions.

In the case of extensive small-stock keeping, the potential benefits of breeding interventions are minimal. Before improving breeding practices, animal health measures (such as vaccination against Newcastle disease in the case of chickens) and some improved and strategic feeding to increase the survival rate of young stock are better options. Once these improvements are made, the management of hardy breeds may face a new challenge: that of intellectual property rights (IPRs), as breeding firms are eager to identify adaptive traits of animals and may go to the extent of not only using but also patenting them. Here, development organizations should be very concerned and defend the IPRs of local farmers.

With respect to commercially oriented small stock, the market risks need to be realistically assessed. Development organizations could give some support by helping to open up new markets.

The keeping of large stock under smallholder management will not decrease in importance in the near future. It is not a particular breed that should be promoted, but rather hardy and adapted animals, which may be of different breeds in different areas. Because animals adapted to the local environmental conditions can be kept without or with minimal tick control and can normally cope with intestinal parasites and some diseases, acaricides, anthelmintics and other chemicals are not necessary. However, the wider implications of giving up dipping need to be considered and this may require a change in national or provincial policy. Where small herds predominate, a bull-sharing scheme could be developed; here, European experience could offer some ideas. However, care should be taken not to disturb already existing breeding arrangements developed by smallholders.

In summary, it is necessary first to gain an understanding of the breeding objectives and practices of smallholder livestock keepers. Interventions that have a bearing on community-based management of farm animal genetic resources cannot be confined to issues of breeding and have to fit into the wider livelihood systems of smallholders.

As we have seen, inappropriate role models can misguide farmers into choosing breeds that are not appropriate for the local conditions - conditions that they cannot influence to any great extent. Interventions are conceivable on different levels. On the level of policy, the carcass grading system needs to be critically examined, as the present system favours large, fast-growing animals. Furthermore, animals adapted to a harsh environment can normally manage without the use of acaricides or anthelmintics. These animals would be classified as "organic" in industrialized countries. Encouraging smallholders to keep breeds adapted to local conditions thus offers a chance for entry into an attractive niche market, if - at the same time - good marketing support is provided. These possibilities should be further explored. On the community level, better cooperation between farmers could be sought with respect to the keeping of sires to provide breeding services. Above all, before suggesting changes, the present systems of managing farm animal genetic resources should be better understood. The understanding of the indigenous livestock systems in southern Africa is still very much at the beginning, particularly in countries with a large commercial sector.

References

Arthus-Bertrand, Y. & Raveneau, A. 1994. Viechereien. Cham, Switzerland, Müller Rüschlikon Verlag. 159 pp.

Bayer, W. 1986. Traditional small-ruminant production in the subhumid zone of Nigeria. In von Kaufmann, S. Chater & R. Blench, eds. Livestock systems research in Nigeria’s subhumid zone, pp. 141-166. Addis Ababa, International Livestock Centre for Africa.

Bayer, W. 1994. Report on consultancy to Indo-German Changar Eco-Development Project, Palampur, India. 7 November - 7 December 1994. Eschborn, GTZ. 42 pp.

Bayer, W. 2000. Report on two consultancies to BASED in the Northern Province of the Republic of South Africa. Eschborn, GTZ. 39 pp.

Bayer, W. & Zemmelink, G. 1998. Management of livestock and forage resources. In W. Bayer & A. Waters-Bayer, eds. Forage husbandry, pp. 41-65. London, MacMillan.

ILCA. 1990. Livestock systems research manual. Working Paper Module 5. Addis Ababa, International Livestock Centre for Africa.

Jones, R.J. & Sandland, R.L. 1974. The relation between animal gain and stocking rate: Derivation of the relation from the results of grazing trials. J. Agricultural Sci. (Cambridge), 83: 355-342.

Karakul Breeders’ Society of Namibia. 2000. Year Book 2000. Windhoek. 69 pp.

de Ridder, N. & Wagenaar, K.T. 1986. Energy and protein balances in traditional livestock systems and ranching in Botswana. Agricultural Systems 20: 1-16.

Rischkowsky, B. 1996. Untersuchungen zur Milcherzeugung mit Saanenkreuzungsziegen in kleinbäuerlichen Betrieben Malawis. Justus-Liebig Universität Giessen. 300 pp. (Dissertation)

Workneh Ayalew, K. 2000. Do smallholder farmers benefit more from crossbred (Somali x Anglo-Nubian) than from indigenous goats? Göttingen, Cuvillier Verlag. 155 pp.

Zemmelink, G. 1980. Effect of selective consumption on voluntary intake and digestibility of tropical forages. Wageningen, Pudoc Scientific. 100 pp.

Zemmelink, G., Brouwer, B.O. & Subagiyo, I. 1992. Feed utilization and the role of ruminants in farming systems. In M.N.M. Ibrahim, R. de Jong, J. van Bruchem & H. Purnomo, eds. Livestock and feed development in the tropics, pp. 444-451. Malang, Indonesia, Brawijaya University.