Smallholder dairy production and marketing in eastern and southern Africa: Regional synthesis

H.G. Muriuki¹ and W. Thorpe²

1. Ministry of Agriculture and Rural Development
P.O. Box 30028, Nairobi, Kenya
2. International Livestock Research Institute
P.O. Box 30709, Nairobi, Kenya

Introduction

The production of milk for the market (dairy production) in eastern and southern Africa (E&SA) has some common characteristics. In all countries except South Africa and Zimbabwe, it is dominated by smallholders. With the exception of Sudan and Somalia, where camels make a major contribution, cattle are by far the largest source of marketed milk (Table 1). Milk is produced in three systems: by pastoralist herds; by herds kept by agro-pastoralists; and by crop–livestock farmers (Walshe et al. 1991). The demand for milk by the producer household and its neighbours, the potential to produce a surplus over these requirements and the accessibility of ‘external’ markets, particularly urban centres, determine the level of milk off-take and therefore the importance of marketed milk production in each of these systems (de Leeuw et al. 1999). Consequently, in E&SA the importance or otherwise of smallholder dairy systems reflects the proximity to major markets, the dietary and cattle (or camel) keeping habits of the local population and the milk production potential of the farming system.

Table 1. Total milk availability for sub-Saharan Africa (SSA) and in selected countries of eastern and southern Africa, 1985 and 1998.

Source: Tambi et al. (2001).

While in the arid, semi-arid and highland zones of E&SA there is a long tradition of cattle (or camel) keeping and milk consumption, cattle production in the more humid zones has been constrained by tick-borne diseases and trypanosomosis, resulting in the predominance of sheep and goats and minimal milk production and consumption (de Leeuw et al. 1999; Kurwijila 2002a). As a result, smallholder dairy systems in E&SA tend to be concentrated in the (subhumid) highlands, the wetter semi-arid and drier subhumid areas, and near to or within urban consumption centres (Walshe et al. 1991). Less proximate production occurs only in those regions with concentrations of traditional consumers and/or with an efficient market infrastructure (Kurwijila 2002a; Muriuki 2002; Omiti 2002; Tsehay 2002).

The majority of the rural systems producing marketed milk in E&SA are integrated crop–dairy systems, which benefit from the positive synergies between the dairy enterprise, staple food crops (generally maize) and any cash cropping (Kurwijila 2002a; Muriuki 2002; Staal 2002). On the other hand, the potential for marketed milk production from pastoralist and agro-pastoralist systems generally depends on the cost of collection and transport of milk that is surplus to producer household needs, which is largely determined by the distance to a consumption centre (Omiti and Staal 1996; Omiti 2002). If there is demand for an easily transported dairy product such as butter, which dominates the Ethiopian dairy market (Tsehay 2002), these costs will be lower than for liquid milk. Conversely, where demand for fresh liquid milk and production is close to a concentration of consumers, pastoralists and agro-pastoralists are able to compete with peri- and intra-urban dairy producers, as shown by Maasai herders in Tanzania supplying the major market of Dar es Salaam (Kurwijila 2002a).

It is these characteristics of smallholder dairying which explain the large variation in dairy consumption, production and marketing amongst and within the countries of E&SA, and which determine the technical, policy and institutional challenges affecting smallholder dairy development.

Milk availability and consumption

Tambi et al. (2001) have reported a comprehensive analysis of the patterns of change from 1985 to 1998 in dairy consumption and production in developing countries, including results for individual countries in E&SA. The measure of milk consumption used in their study is a simplification of the system used by the Food and Agriculture Organization of the United Nations (FAO) in its ‘food balance sheets’. It uses ‘domestic supply’ without adjustments for changes in stocks and is referred to as ‘milk availability’, calculated as domestic production plus net imports. In the absence of documented studies on actual consumption in the majority of countries in E&SA, the estimates of Tambi et al. (2001) give a reasonable comparative picture of the regional variation in dairy consumption.

The analyses show that in 1998, developing countries had over 228 million tonnes of milk available for consumption and other uses. Of this total, India alone had 31% and sub-Saharan Africa (SSA) only 8%. Whilst camels’ milk was not important in west, central and southern Africa, it was in eastern Africa (Table 1). Nevertheless, cows’ milk dominates marketed milk production throughout the region. In SSA, net imports accounted for only 10% in 1998, down from 16% in 1985. These imports were mainly serving consumers in West and Central Africa. In E&SA, net imports formed a very small proportion of available milk (Table 1) indicating the high degree of self-sufficiency in production of milk and dairy products in the region, which is an oft-stated national policy (e.g. in Ethiopia, Tsehay 2002; and Tanzania, Kurwijila 2002b).

Between 1985 and 1998, of all developing country regions except West Asia and North Africa (WANA), SSA had the smallest percentage increase in total milk availability (27%). Kenya and Sudan accounted for over 70% of the change (Table 1). The insecurity and political instability in many eastern and southern African countries, e.g. Angola and Somalia, contributed to the region’s relatively poor performance, which in turn affected availability per capita.

What is more, from 1985 to 1998, milk availability in SSA grew less rapidly than the human population, such that by 1998 SSA experienced a decrease of approximately 4 kg in milk availability per capita (Table 2). While cows’ milk production per capita decreased in several countries in E&SA, it increased for the region as a whole because Sudan and Kenya—the region’s largest milk producers—reported significant increases in cows’ milk production. Although net imports and production from other species declined in Kenya, the increase in cows’ milk production more than compensated for the decline. As a result, there was an overall increase in total milk availability per capita, which was reported as 85 kg in 1998 (Table 2). This value was approximately four times higher than the availability in Ethiopia and Tanzania, and approximately double that of Zimbabwe (countries including large areas with endemic trypanosomosis and tick-borne diseases), but much lower than for Sudan and Somalia, countries with many pastoral people.

Table 2. Per capita milk availability (kg) for sub-Saharan Africa (SSA) and in selected countries of eastern and southern Africa, 1985 and 1998.

Source: Tambi et al. (2001).

Neighbouring countries, such as Ethiopia, Somalia, Tanzania and Uganda, exhibited different patterns of change of milk availability (Table 2). In Ethiopia (including Eritrea), per capita availability decreased between 1985 and 1998 by 21% from the already low level of 26 kg. This was due to decreases per capita in net imports and in milk production from ‘other species’. Somalia, Tanzania and Uganda each experienced a decline in domestic production and net imports. In southern Africa, Zimbabwe exhibited a similar pattern of change to those of the eastern African countries. Between 1985 and 1998, availability per capita decreased by 14 kg, mainly because of declining production of cows’ milk and a reduction in net imports (Table 2). These changes reflect the rapid growth of the country’s human population, the decline of the dairy herd in the large-scale commercial sector and the slow development of smallholder dairying (Smith et al. 1998).

Kurwijila (2002a) reports that dairy consumption has risen faster in urban and peri-urban areas of Tanzania than in rural areas because of the growth in peri-urban and intra-urban dairy herds and the increased availability of milk and dairy products for urban consumers. In Dar es Salaam, for example, the dairy cattle herd increased from about 2 thousand in 1984 to over 20 thousand head by 1995. Consequently, the per capita consumption is much higher in urban centres (40 litres/annum) than in rural areas (15–20 litres/annum) (Kurwijila 2002a). On the other hand, Muriuki (2002) cites recent results for the central region of Kenya, where smallholder dairy production is a major part of the farming system. These data show that dairy consumption is higher in rural producer areas than in the major urban centres of Nakuru and Nairobi where, unlike in Dar es Salaam, urban dairy production is not common.

These results suggest the need in the E&SA region for more detailed studies of dairy consumption levels and their patterns, and for a better understanding of the factors affecting these levels and their likely trends over the next 10–20 years. Only with use of reliable estimates of current and projected demand will producers, processors, market agents, and the technicians and policy makers who serve them, be able to support the development of smallholder dairy through efficiently meeting consumers’ needs.

Dairy production systems

As for milk availability, Tambi et al. (2001) have estimated the changes in milk production from 1985 to 1998 in developing countries. For E&SA, the results relate to cattle, the dominant dairy species in most countries in the region (Tables 3 and 4). In addition, the authors ‘decomposed’ the changes in milk production due to changes in total herd size (herd effect), the proportion of animals milked (milking effect) and the productivity per animal (productivity effect) (Table 5). The analyses showed that SSA had the lowest milk production per cow of all regions of the developing world. Similarly, between 1985 and 1998, SSA had the smallest increases in the proportion of animals milked and their productivity.

Table 3. Cattle numbers and milking cows in sub-Saharan Africa (SSA) and in selected countries in eastern and southern Africa, 1985 and 1998.

Source: Tambi et al. (2001).

Table 4. Cow’s milk production and yield per cow in sub-Saharan (SSA) Africa and in selected countries in eastern and southern Africa, 1985 and 1998.

Source: Tambi et al. (2001).

Table 5. Sources of change in cows’ milk production in sub-Saharan Africa (SSA) and in selected countries in eastern and southern Africa, 1985–98.

Source: Tambi et al. (2001).

Most cows’ milk production in SSA was concentrated in eastern Africa; Sudan, Kenya, Ethiopia, Somalia and Tanzania were the top five countries producing about two-thirds of the total cows’ milk in the continent (Table 4). For the period 1985 to 1998, these eastern African countries demonstrated different patterns of change in cows’ milk production, in part because unlike the cattle population in Kenya, the large cattle populations of Sudan, Ethiopia and Eritrea had relatively limited numbers of exotic dairy cattle and their crosses (Table 6; Tsehay 2002). Partially as a consequence, milk yields increased by only 17 kg/animal per year in Ethiopia and Eritrea (Table 5); this small increase accounted for one-quarter of the total increase in milk production. An increase in the size of the cattle herd by 6.5 million head also contributed significantly to the increase in milk production. In Sudan, milk yield per milking cow stagnated, whilst the high growth in the cattle herd accounted for >80% of the increase in milk production (Table 5). A similar pattern was noted in Uganda, where milk yield was static with the increase in milk production coming mostly from expansion of the herd and the proportion of animals milked.

Table 6. Dairying in eastern and southern Africa: Cattle, milk production and per capita milk availability.

* LME = liquid milk equivalent.
Sources: Omiti and Staal (1996); Omore and Staal (1998); Mpofu (1999); Tambi et al. (2001); Tsehay (2002).

In Kenya, where 60% of milk comes from dairy cattle (Muriuki 2002), milk yields/cow were more than double those in Ethiopia (Table 4), but they grew only marginally faster (about 44 kg/animal) than in Tanzania. An increase in the proportion of milking animals from 25 to 33% of the herd contributed >60% of Kenya’s 800 thousand tonnes increase in milk production. The increase in productivity contributed about one-fifth of the increase in milk production (Table 5). By contrast, in Somalia increases in cattle numbers were responsible for the entire increase in milk production, while in Zimbabwe a change in the herd structure towards increased milking was responsible for the entire increase in milk production between 1985 and 1998 (Table 5).

Relative to all other countries in E&SA, Kenya has a very large herd of exotic breed dairy cattle and their crosses. The Kenyan dairy herd probably accounts for over 75% of all specialised dairy cattle in E&SA (Table 6). These dairy cattle, the descendents of the cattle from European settler farms established almost a century ago (Conelly 1998), are predominantly owned by smallholders (Muriuki 2002). Several major factors have contributed to the widespread adoption of dairying by smallholders in Kenya (Muriuki 2002; Omiti 2002) including: the importance of milk in the diets of most Kenyan communities; a favourable production environment (mid to high altitudes with bimodal rainfall); the presence of the original settler dairy cattle population; and policy and institutional environments (through to the early 1990s) conducive to large- and small-scale dairying. As for Kenya, in Tanzania the exotic breed-based dairy cattle population produces the majority of the country’s marketed milk (Table 6; Kurwijila 2002a). It is significant that, in common with the systems in Kenya, approximately 65% of Tanzania’s dairy cattle are located in the bimodal rainfall northern highlands in smallholder crop–dairy systems that support high human population densities (Omore and Staal 1998). It is these areas of good potential for biomass production, combined with ready markets for milk, which have significant potential for smallholder dairy development in E&SA.

By contrast, in Malawi, South Africa and Zimbabwe, the mid altitude, mono-modal rainfall agro-ecologies of southern Africa (with their lower potential for biomass production), the risk for cattle diseases and, until recent years, the policy and institutional environments have inhibited adoption of dairy production by smallholders. Consequently, their dairy cattle populations are small relative to their total cattle populations (Table 6; Smith et al. 1998; Mpofu 1999). Furthermore, as yet, market mechanisms are not in place to extract milk from traditional systems, which, in any case, are largely in agro-ecologies adverse to producing milk in excess of the needs of the producer households and their neighbours.

Most countries in E&SA have, therefore, not benefited in the way that Kenya has and to lesser extents Tanzania and Uganda have from smallholder dairy development. As Muriuki (2002) reports, dairying in Kenya has become a very significant source of income to the estimated 625 thousand smallholder producer households and to those employed in the marketing of milk, in total some 25% of all households. In addition, dairying plays a crucial role in sustaining smallholder crop–dairy systems through nutrient cycling within the systems (Kurwijila 2002a; Muriuki 2002; Staal 2002) which, along with the adoption of planted forage and agro-forestry technologies, has played a crucial role in the development of crop–dairy systems that sustain increasingly high human population densities, even in some semi-arid areas (Tiffen et al. 1994).

In the face of sub-division of family farms as land passes from generation to generation, adopting dairying and owning a dairy cow (most households own only one or two) is, therefore, a means of survival for many smallholder families in Kenya (Muriuki 2002). It is also a potential means of accumulating some capital. Incremental daily inputs of labour, land, feed and other inputs over time are accumulated and compounded in the form of an additional cow or heifer, or a saleable male (Staal 2002). These animals may be sold when needed to meet lump-sum expenditures, such as school fees or medical bills, or to invest in upgrading farm facilities. As such in E&SA and particularly in Kenya, smallholder dairy represents an important tool in reducing poverty in rural and peri-urban areas. Whereas, at times it is argued that those already owning dairy cattle are not poor, most smallholders practising dairying were poor and struggled to acquire their first cow. Dairying was a means to escape poverty and to sustain their families, with particular benefits accruing to women and children (Kurwijila 2002a and b; Muriuki 2002; Omiti 2002; Tsehay 2002; Staal 2002; Tangka et al. 1999).

Throughout E&SA, returns to dairying vary considerably because producer prices for milk vary between surplus and deficit and between urban and rural areas, and because costs of production vary depending on the production system and whether outputs other than milk are valued products (Kurwijila 2002a). For example, in northern Tanzania, manure and heifers are particularly important outputs in the coffee/banana × zero-grazed dairy-based economies where manure is ranked second in value after milk (Rugambwa et al. 1995). However, Kurwijila (2002a) concludes that the cost of producing milk in Tanzania is seemingly higher than in neighbouring Kenya and Uganda, where farmers accept lower prices for both milk and heifers.

In Kenya, the costs of and returns to production have been estimated recently for two contrasting sites (Staal 2002). The first was an area of relatively extensive crop–dairy production in Nakuru, Rift Valley, where farmers keep three to five crossbred dairy cattle and rely mostly on grazing. The second was in Kiambu in the intensive central highlands, where land sizes are smaller (an average of two acres), so farmers keep two to three high-grade dairy cattle. Because the land is not able to provide adequate animal nutrition, farmers purchase some fodder and concentrate feeds. Profits to dairy production were US$ 0.02 and 0.04/litre for the extensive and intensive sites, respectively. These were returns after normal wage costs (currently, rural wages are approximately US$ 25/month) had been deducted for the family labour. They suggest that smallholder dairying can compete well against alternative enterprises available to the farmers. Staal (2002) suggests that these two sites are representative of the important dairy production areas of highland eastern Africa and that the results imply good opportunities for smallholder dairy producers in the region.

In summary, there is much variation in the level of adoption of market milk production (dairy production) by smallholders in E&SA. Particularly Kenya, and to some extent Tanzania and Uganda, have significant dairy subsectors based on smallholders, the outputs of which are reflected in high per capita availabilities of milk (Tables 2 and 6). By contrast, the high per capita availability of milk in Zimbabwe (Table 6) and South Africa reflect their continuing dependence on large-scale production units. Factors explaining this marked variation and underpinning the high adoption of smallholder dairying in East Africa include: the importance of milk in the diets of rural communities and urban consumers; a favourable production environment (mid to high altitudes with bimodal rainfall); the availability of the original settler dairy cattle populations; and policy and institutional environments (through to the late 1980s/early 1990s) conducive to smallholder dairying.

Constraints to increased productivity in current systems include: inadequate year-round feeding; losses from cattle diseases, particularly tick-borne diseases; and poor access to input and output market services. In much of E&SA, the lack of adequate feed (particularly in the mono-modal rainfall areas) and disease challenge, interacting with a lack of veterinary services, inhibit the adoption of dairying by smallholders. Because of the resultant low milk production densities (litres/km²), these constraints impose a very high cost on attempts to introduce milk collection schemes and related output market services. Until these interacting inhibitory factors are addressed, increased milk production in the region (and in West and Central Africa) is likely to continue to result from expansion of the indigenous cattle population and some increase in the proportion milked, with relatively little coming from increases in productivity (which are usually associated with the use of exotic dairy breeds and their crosses).

Milk marketing systems

In most of E&SA, milk production and marketing systems are those described by Tsehay (2002) for Ethiopia: an urban system; a peri-urban milk system; and a rural system. While in Ethiopia, butter is the predominant traded product (Tsehay 2002), generally consumers in E&SA demand fresh liquid milk and its marketing is dominated by traditional (the so called ‘informal’) markets, with only small proportions of total production being marketed through a cold-chain, pasteurised process (the so called ‘formal’ market). For example, in Ethiopia the proportion of total marketed milk sold formally is very small (Tsehay 2002); in Tanzania and Uganda it is estimated at <5% (Omiti and Staal 1996; Kurwijila 2002a) and in Kenya it is about 15% (Omore et al. 1999). Approximately 30–35% of production is consumed on farm (by the family and calves), with the balance (generally four to six litres) marketed.

In the dominant traditional (or ‘informal’) markets, the milk may pass straight from the producer to a domestic or institutional consumer, or it may pass through two or more market agents before reaching the consumer (Staal et al. 1997; Omiti 2002). Tsehay (2002) and Kurwijila (2002a and b) have noted the importance of intra-urban dairy production in Ethiopia and Tanzania, which, as for peri-urban systems, shortens the market chain for the fresh milk that is preferred by the majority of consumers. As the majority of those consumers have no access to refrigeration, invariably, the custom is to boil the milk to extend its shelf life (Walshe et al. 1991; Omiti and Staal 1996; Kurwijila 2002a; Muriuki 2002; Redda 2002).

As Omiti (2002) has discussed for Kenya, the macro-economic reforms implemented or being implemented in E&SA, have increased the competition for marketing functions (such as collection, transportation, processing and distribution/retailing) and have resulted in increased income and employment opportunities, especially for small-scale milk traders (Omiti and Muma 2000). Many sell <120 litres of milk per day, but this business activity enables them to earn a daily income equal to approximately twice the national average (Omore et al. 1999; Staal 2002), which represents a significant contribution to poverty reduction. Similar estimates are available for Tanzania (Kurwijila 2002b) and Uganda (Omiti and Staal 1996), and presumably are estimable for southern African countries like Malawi and Zimbabwe, where there are some indications that enforcement of regulations banning the informal marketing of milk is being relaxed. In the face of these strong informal markets, many governments are having to address how best to ensure fair competition between the ‘formal’ and ‘informal’ markets to the benefit of producers and consumers, most of whom are in low-income households (Muriuki 2002; Omiti 2002).

Input services

In the same way that policy and institutional reforms are being implemented to support competitive milk marketing, public institutions and the private sector in many E&SA countries are trying to develop innovative and progressive institutional mechanisms for input services. These include efforts to encourage the formation of rural savings and credit co-operative organisations. As Omiti (2002) points out, where cash is a serious constraint to dairy development for subsistence-oriented farmers, designing appropriate institutional mechanisms of availing credit is a prime concern.

Other input markets serving smallholder dairying are also major concerns. The 1980s and 1990s have seen most E&SA countries experience the collapse of some and the decline of the remainder of government input services (veterinary, artificial insemination (AI) and extension advisory services) for smallholders, with an increased reliance for service delivery on the private sector, including community-based organisations (CBOs) and co-operatives (Owango et al. 1998). Clinical and preventive veterinary services have been a prime target for privatisation (Tambi et al. 1997 and 1999).

Nahdy (2002) articulates eloquently these issues as faced by Uganda and describes its government’s radical plans for restructuring research and extension services, and their decentralised funding and management by local communities. Complementary to and underpinning this will have to be a mindset shift by technicians and their managers to participatory approaches for research and extension services for smallholders, such as the steps reported by Mwangi and Wambugu (2002) for improving the availability of feed resources in Central Kenya. Their efforts are building upon the experiences gained in Coastal Kenya, where gender-sensitive, client-oriented research and extension support was developed for smallholder dairy producers and the early adopters of dairy production (Maarse et al. 1998).

As Kurwijila (2002b) states, with the current move to privatise most government services, the challenging question is how to internalise costs such as those for research and extension services, training, the control of epidemic diseases and the delivery of AI services which have elements of public goods; the cost of these services may well be beyond the reach of smallholder farmers, at least in the short term. The Ugandan example (Nahdy 2002) is one approach, which includes the implicit gradual scaling down of donor/government subsidies (Kurwijila 2002b; Omiti 2002). Certainly, the priority for the shift from public to private delivery of input services is a managed transition (Kurwijila 2002b; Muriuki 2002), not the often observed sudden withdrawal of operating funds to government agencies to the detriment of the livelihoods of many rural households.

Policy and institutional issues

Although it has been shown that in E&SA smallholders are competitive in dairy production, it is probable that policy interventions will be required to sustain their viability, especially to support the more widespread adoption of dairying by smallholders within the region (Omiti 2002; Staal 2002). This is because the small scale of milk production and marketed output implicit to smallholder systems can often result in low bargaining power and limited ability to capture economies of scale in marketing. As was described in the section on milk marketing systems, farmers use a variety of strategies to overcome this. The first approach is generally to sell their milk directly through the informal market to consumers, thereby achieving higher prices and reducing transactions costs (Staal 2002). The use of milk traders to bulk milk, and perform distribution and marketing services is also common. Dairy farmer co-operatives and other farmer groups, if efficiently managed, can improve the market position of smallholder farmers through collective actions (Owango et al. 1998; Kurwijila 2002b; Muriuki 2002; Omiti 2002;  Tsehay 2002). Policies that support these activities, and do not interfere with individual market activities of farmers and traders, are likely to sustain competitiveness of producers and contribute effectively to improved rural and peri-urban livelihoods.

Kurwijila (2002b) has also stressed the need for improvements in processing, quality assurance and efficiency if the smallholder dairy subsector in Tanzania and elsewhere is to survive in an increasingly liberalised global market. He argues that this will require self regulation rather than control from the government, which, by implication, means that the industry will have to organise itself to better face the challenges of today and tomorrow. Currently, there is movement in E&SA towards stakeholder-managed national dairy boards with efforts to ensure the effective representation of and voice for the smallholders and their market agents who dominate dairy production and marketing.

Likewise, the continued privatisation of input services will function for those services and regions where smallholders can afford the costs of targeted quality services (Tambi et al. 1999; Kurwijila 2002b), but for other services and in more marginal areas, public support may continue to be needed (Tambi et al. 1997; Omiti 2002; Staal 2002). These input services include research and extension, as discussed by Nahdy (2002) and others. Last, but by no means least, is the urgent need in the E&SA region for more investment in national infrastructure such as rural access roads, water supplies and electricity distribution (Muriuki 2002). Such investment would ensure that these essential public services are available to support the smallholder production and marketing of perishable products like milk.

Conclusions and challenges

Delgado et al. (1999) presented a convincing case for continued demand-driven dairy development to 2020 through which smallholders, including those in E&SA, if given adequate technological, policy and institutional support, will benefit by meeting the projected large increases in milk consumption. The projected estimates of increased demand for milk are based upon the expected growth of the human population, its urbanisation and its increased purchasing power (Delgado et al. 1999). In E&SA, as elsewhere, the ‘white revolution’ as some have called it, is therefore linked to overall economic performance (especially the creation of urban jobs). Current political and economic uncertainties in E&SA, the importance to dairy development of government reform processes, including the need for urgent public investment in infrastructure (especially roads and water supplies), allied to the unfolding tragedy of AIDS and its effects on communities and their productivity, suggest that some re-examination of the projections for E&SA may be required. A re-examination would serve to guide and stimulate more-focused national and regional programme support for smallholder dairy development.

An important part of that re-evaluation should be ex-post impact assessment studies of dairy development in the region, including evaluation of the returns to the many major project investments during the last 40 years (e.g. in Tanzania; Kurwijila 2002b). Many useful lessons can be learnt from the successes and the more numerous failures of past efforts.

These analyses should be guided by the broad conclusion from this synthesis (and the earlier study by Walshe et al. 1991) that the factors explaining the marked variation in adoption of smallholder dairying in E&SA are: whether or not milk is important in the diets of rural communities and for urban consumers; whether there is a favourable agro-ecological environment for producing milk in excess of the needs of the producers’ households and their neighbours; whether urban consumer centres are accessible by smallholder producers; whether there are sources of dairy genotypes; and whether policy and institutional environments are conducive to smallholder dairying. Without good access to efficient output and input markets (including research and extension services), favourable production environments will not be enough to stimulate smallholder production in the areas of E&SA where, as yet, dairy development has been slow.

Therefore, the major challenges to be faced in E&SA are the need for:

• Better information on the demand for milk and dairy products and the determinants of that demand
• An understanding of the lessons learnt from the wide variation in dairy development in the region and identification of the determinants of that variation
• Communication of those lessons to policy makers and development planners backed by estimates of the aggregate value to society of dairy development, especially smallholder dairy development and
• Stimulation of a change of mindset amongst many advisers and technicians from a supply-driven to a demand-driven development process.

References

Conelly W.T. 1998. Colonial era livestock development policy: Introduction of improved dairy cattle in high-potential farming areas of Kenya. World Development 26:1733–1748.

Delgado C., Rosegrant M., Steinfeld H., Ehui S. and Courbois C. 1999. Livestock to 2020: The next food revolution. Food, Agriculture and the Environment. Discussion Paper 28. IFPRI (International Food Policy Research Institute), Washington, DC, USA; FAO (Food and Agriculture Organization of the United Nations), Rome, Italy; and ILRI (International Livestock Research Institute), Nairobi, Kenya. 72 pp.

Kurwijila L.R. 2002a. Dairy development in Tanzania: Country paper. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Kurwijila L.R. 2002b. An overview of dairy policies in Tanzania. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

de Leeuw P.N., Omore A., Staal S. and Thorpe W. 1999. Dairy production systems in the tropics. In: Falvey L. and Chantalakhana C. (eds), Smallholder dairying in the tropics. ILRI (International Livestock Research Institute), Nairobi, Kenya. pp. 19–37.

Maarse L., Wenholt W. and Chibudu A. 1998. Making change strategies work: Gender sensitive, client oriented livestock extension in Coast Province, Kenya. Royal Tropical Institute, Amsterdam, Netherlands. 114 pp.

Mpofu N. 1999. Sourcing of dairy cattle by smallholder farmers in southern Africa. Market-oriented Smallholder Dairy Project Working Document 6. ILRI (International Livestock Research Institute), Nairobi, Kenya. 15 pp.

Muriuki H.G. 2002. Smallholder dairy production and marketing in Kenya. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Mwangi D.M. and Wambugu C. 2002. Forage legume technologies: Participatory experiences with farmers and community based organisations in central Kenya. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South– South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Nahdy S.M. 2002. Decentralisation of services in Uganda: The formation of national agricultural advisory services. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Omiti J.M. 2002. Impacts of liberalisation in Kenya’s dairy sector. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Omiti J. and Muma M. 2000. Policy and institutional strategies to commercialise the dairy sector in Kenya. Occasional Paper 6/2000. IPAR (Institute of Policy Analysis and Research), Nairobi, Kenya.

Omiti J. and Staal S. (eds). 1996. The Ugandan dairy subsector: A rapid appraisal. ILRI/NARO/MAAIF Research Report. ILRI (International Livestock Research Institute)/NARO (National Agricultural Research Organization of Uganda)/MAAIF (Ministry of Agriculture, Animal Industry and Fisheries of Uganda), Kampala, Uganda. 137 pp.

Omore A. and Staal S. (eds). 1998. The Tanzania dairy subsector: A rapid appraisal. Volume 3—Main Report. ILRI (International Livestock Research Institute), Nairobi, Kenya.

Omore A., Muriuki H., Kenyanjui M., Owango M. and Staal S. 1999. The Kenyan dairy subsector: A rapid appraisal. Smallholder Dairy (Research & Development) Project Report. Ministry of Agriculture/Kenya Agricultural Research Institute/International Livestock Research Institute, Nairobi, Kenya. 51 pp.

Owango M.O., Staal S.J., Kenyanjui M., Lukuyu B., Njubi D. and Thorpe W. 1998. Dairy co-operatives and policy reform in Kenya: Effects of livestock service and milk market liberalisation. Food Policy 23:173–185.

Rugambwa V.K., de Wolf J., Stanslaus D. and Lutalala E.M. 1995. Credit facilities and heifer supply within the Kagera Livestock Development Programme, Tanzania. In: Proceedings of the FAO (Food and Agriculture Organization of the United Nations) workshop on market orientation of small-scale milk producers and their organisations held at Morogoro, Tanzania, 20–24 March 1995. pp. 66–76

Smith T., Titterton M. and Dube D. 1998. Dairy production systems in Zimbabwe. Paper presented at the BSAS (British Society of Animal Science) conference on food, lands and livelihoods held at Nairobi, Kenya, 27–30 January 1998.

Staal S.J. 2002. Smallholder dairy competitiveness. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Staal S., Delgado C. and Nicholson C. 1997. Smallholder dairying under transactions costs in East Africa. World Development 25:779–794.

Tambi E., Mukhebi A.W., Maina O.W. and Solomon H.M. 1997. Privatisation of animal health services in Kenya: An evaluation of the Kenya Veterinary Association Privatisation Scheme (KVAPS). OAU (Organization of African Unity)/IBAR (Inter-African Bureau for Animal Resources)/PARC (Pan-African Rinderpest Campaign)/ILRI (International Livestock Research Institute) Report. ILRI, Nairobi, Kenya. 84 pp.

Tambi N.E., Mukhebi W.A., Maina W.O. and Solomon H.M. 1999. Probit analysis of livestock producers’ demand for private veterinary services in the high potential agricultural areas of Kenya. Agricultural Systems 59:163–176.

Tambi E., Nicholson C., Staal S. and Thorpe W. 2001. Patterns of change in dairy production and consumption in developing countries from 1985 to 1998. Market-oriented Smallholder Dairy Research Working Document 7. ILRI (International Livestock Research Institute), Nairobi, Kenya. 65 pp. (In press).

Tangka F., Ouma E.A., Staal S.J. and Shapiro B. 1999. Women and the sustainable development of market-oriented dairying: Evidence from the highlands of East Africa. Paper presented at the international sustainable development research conference held at the University of Leeds, UK, 25–26 March 1999.

Tiffen M., Mortimore M. and Gichuki F. 1994. More people, less erosion: Environmental recovery in Kenya. ACTS Press, Nairobi, Kenya.

Tsehay Redda. 2002. Small-scale milk marketing and processing in Ethiopia. In: Rangnekar D. and Thorpe W. (eds), Smallholder dairy production and marketing—Opportunities and constraints. Proceedings of a South–South workshop held at NDDB, Anand, India, 13–16 March 2001. NDDB (National Dairy Development Board), Anand, India, and ILRI (International Livestock Research Institute), Nairobi, Kenya.

Walshe M.J., Grindle J., Nell A. and Bachmann M. 199. Dairy development in sub-Saharan Africa: A study of issues and options. World Bank Technical Paper 135. Africa Technical Department Series. The World Bank, Washington, DC, USA. 97 pp.