The Borana Plateau of southern Ethiopia: Synthesis of pastoral research, development and change, 1980-91 summarises results from work conducted in the southern Ethiopian rangelands between 1980 and 1991. The global objectives of this inter-disciplinary project were to describe the evolving production system of the Borana pastoralists and prescribe best-bet component interventions and policies that might promote growth in the livestock sector, alleviate poverty among pastoral producers and encourage ecologically sustainable patterns of resource use. A large effort was also devoted to contrasting our research results with other findings in the pastoral literature, largely from eastern and southern Africa. This comprehensive system study is intended to serve two main audiences: (1) professionals within Ethiopia who deal with range research and development, who need detailed interpretation of local data but who also have poor access to the international literature and (2) an international audience concerned with more general implications of the work for pastoral research and development in sub-Saharan Africa.
Borana society is in crisis today, mostly due to human over-population. Scholars of pastoral development will recognise many aspects of system change that have been observed elsewhere in Africa. Despite daunting challenges, we believe that a combination of policies, procedures and technical options could help manage the system to reverse the downward trend in human welfare. Such efforts, however, will require a high degree of creativity and commitment on the part of the Borana people, the Ethiopian Government and development agencies. Problems need to be addressed in new ways if major impact is to be achieved.
These highlights are structured as responses to 13 major questions that decision makers are likely to have.
1. Why focus on the Borana pastoral system?
The original reason for focusing on the Borana pastoral systems was that the semi-arid southern rangelands are valuable to Ethiopia as a source of livestock for use by smallholders in the highlands and for export to generate foreign exchange. The region also had the highest ecological potential among major range-development areas and the semi-sedentary Boran were regarded as relatively easy to study and work with. Development of infrastructure in the south started in the 1960s and it was thought that this would facilitate the impact of research results on development.
2. Does Ethiopia still have a stake in the rangelands?
Yes. Although the highlands are justifiably the major focus of agricultural development efforts in Ethiopia, the rangelands cannot be ignored in a comprehensive national strategy. High rates of population growth throughout the country dictate that commerce should be allowed to flow freely and thus permit comparative production advantages of different agro-ecological zones to be expressed. The rangelands will increasingly serve as an important source of animals for highland smallholders and for export. The Boran are now in dire need of grain from the highlands, at favourable terms of trade, to reduce risks of famine and lessen the need to expand cereal cultivation onto fragile upland range soils in the rangelands. A loop of mutual assistance can now be completed, achieving the vision set for Ethiopia by planners over 20 years ago. What is required are integrated policies and improved access to inputs that allow producers and traders to create mutually beneficial networks themselves.
3. The Boran have persisted for a long time. Why bother to develop the southern rangelands now?
Unless efforts are now made to improve human welfare among the Boran, future commercial linkages with the rest of the nation and the social welfare of urban centres in the rangelands will be in jeopardy. Increased risk of famine, increasing poverty and the undermining of traditional cultural values may all erode the traditional social order of the Borana production system. If the traditional social order is not maintained, there are acute dangers of increased regional insecurity and less efficient operation of the deep wells. Requiring large amounts of coordinated labour, the deep wells are virtually the only supply of water in dry seasons and their efficient operation underpins the viability of the entire livestock production system. Ecological sustainability, on the other hand, is threatened by the spectre of increased cereal cultivation on upland soils and by woody encroachment and soil erosion, which can be attributed to heavy cattle grazing.
4. What is the outcome of development in the southern rangelands thus far?
Starting mostly in the 1970s, there have been considerable efforts to develop infrastructure and provide veterinary services to the Boran. The intent was to stimulate livestock commercialization. This involved the typical, but often erroneous, assumptions of pastoral behaviour that have characterized African pastoral development in general.
It was prominently assumed, for example, that if the Boran were given access to markets they would readily sell cattle and improve their lives by increasing their cash income. Sales of immature cattle were to form the basis of a stratified livestock industry. Today, however, instead of happy, prosperous pastoralists we find a situation in which 200000 people are on food relief and half of the region's cattle died during the 1990-91 drought. Rates of cattle offtake remain low. Without economic development, food relief is likely to become a permanent fixture, regardless of drought. Woody vegetation has encroached on 40% of the land area and 19% of the area has suffered significant soil erosion. What went wrong?
First, we contend that interventions to boost livestock production actually worked. New ponds increased access to underutilized productive lands and veterinary campaigns lessened risks to animal production. Herd size likely grew, but then so did the human population. Improvements in animal production were probably largely absorbed by a growing subsistence population, rather than being marketed. The traditional Boran probably had little desire to routinely enter the market-place because they did not need to. They produced most of their own food as milk, traded for a bit of grain and used cash sparingly to buy discretionary items such as coffee, shoes or sugar. Their cattle had greater perceived utility as accumulated assets than as a cash crop. People with larger herds have fewer economic risks, are socially influential in the community and even have landmarks and encampments named after them. This is what people aspire to here.
It is also likely that a traditional Boran household waited until it had an acute need for cash before selling an animal. The main season for selling cattle was thus the warm dry season, when milk supply is lowest and the people needed to buy food. The terms of trade would often be unfavourable to the Boran during this season, but this was of only marginal relevance because the people commonly hoped to avoid the sale all together. They thus behaved more like optimistic gamblers, hoping that good rains or some other favourable circumstance would help them obtain the food they needed to carry them over to the next rainy period and allow them to avoid the sale. The people could endure considerable misery in waiting out a dry season.
If a household were forced to sell cattle, it would tend to sell an older, mature male rather than younger animals. This is because the higher gross proceeds from the sale of a large animal would be enough to buy both the commodities needed and replacement calves. It would thus meet both the immediate need for cash and promote herd growth. Net proceeds are not very relevant where costs of production are almost nil and concern over how fast the animal grew would not have been prominent compared with efforts just to keep it alive. A poor household would typically have to sell more animals than would a wealthy household to buy food throughout the year. This is because poor households have more people per milk cow than do wealthy households and hence have too little milk to support them. Fewer cattle also implies less diversity in age and sex classes. Thus, a poor household would tend to sell an immature animal more commonly than would a wealthier household because of lack of choice. Selling an immature animal to buy food also offers less likelihood of money being left over to buy replacement stock and thus less opportunity for herd-building and a greater likelihood of continued poverty.
Some views in this scenario above may be controversial but together they help explain (1) the traditional economic rationale; (2) why range development projects have commonly failed to meet expectations; (3) why droughts can decimate cattle herds and (4) why pastoralists are commonly victimised by poorer seasonal terms of trade of livestock for grain.
In sum, the traditional Borana household did not need much money but aspired to having a large cattle herd. It preferred to sell older animals rather than immatures and probably was concerned only with low-input means to keep animals alive. In addition to the problem that the goals of livestock commercialization conflicted with the traditional production rationale, livestock prices within Ethiopia were regulated and kept low until 1991 and there were chronic difficulties in coordinating external market linkages from the southern rangelands. Little wonder, then, that development expectations were not fulfilled. This is not a society that would respond to widespread marketing initiatives or want to sell immatures as part of a stratified cattle production industry. This also suggests that efforts to provide early warning of droughts to encourage destocking may not work very well. What appears to be commonly misunderstood is the role of cattle as a primary asset, not as a cash crop, and what this implies for human economic behaviour.
Lack of development impact has not been due to scarcity of technology or inappropriate behaviour on the part of the Boran. Expectations were unrealistic because Western-trained planners had an inadequate understanding of social values and a production rationale that differed from their own.
This is not to say that development efforts have been fruitless but that impact has been more indirect than direct. Roads and markets, less used in the past, will soon become the critical lifeline for the Boran because they can no longer feed themselves using traditional methods. The small towns that have grown up as a result of improvements in infrastructure provide crucial market outlets and have led to widespread dissemination of new ideas among the pastoral population. We thus postulate that the response of the pastoral population to infrastructure is highly dependent on timing. The willingness of the Boran to receive innovations is dictated by population pressure; change occurs when the people have little choice. The stage is now set to accommodate a major economic shift in Borana society.
5. Is the traditional system changing?
Yes, and very quickly. If cattle prices are competitive with those offered on the Kenyan black market and trade links within Ethiopia are made more efficient, the good news for planners is that rates of cattle offtake for Ethiopian markets should increase dramatically. More immature cattle will be sold. The original vision for increased marketing by pastoralists can be fulfilled. The bad news, however, is that this increased involvement in the market economy will largely result from the increasing poverty and risk of famine that is undermining Borana society. A larger segment of the society is becoming poor and acting in accordance with the schema of behaviour of the poor as reported above. Today, 51% of the households may be considered poor; these households control about 10% of the regional cattle herd. Around 18% may be wealthy and control 65% of the regional cattle herd. Traditionally, the poor would petition the wealthy to redistribute cattle at dozens of annual clan meetings. Such meetings are still held but needs are beginning to dwarf possibilities for redistribution.
The long-term trend that drives increased commercialization among the Boran is the declining ratio of cattle to people. The human population is increasing at a net rate of 2.5% per year, with a 50% increase in population possible within 14 years. Apparently, relatively few people are killed by drought; even births are reported during drought. In contrast, land availability imposes a ceiling on cattle numbers and large numbers of cattle die during droughts. The net result is a downward "ratchet" effect: the steady decline in the ratio of cattle to people is periodically exacerbated by drought. More people are thus becoming poorer in cattle assets.
A decline in the number of cows per person reduces per capita milk production and forces people to buy grain or grow crops. A decline in the number of male cattle per person limits, among other things, the ability of the people to rebuild their herds after drought by exchanging males for cows. More households are likely to be squeezed out of the system during drought because of the intensified competition of more people for a finite base of resources. Furthermore, this base is becoming even smaller due to ecological degradation and loss of traditional grazing reserves to encroachment by the growing population.
People squeezed out of the system will increasingly become destitute farmers or pert-urban sellers of milk, chickens and firewood. Peri-urban dairy marketing results, in part, from these processes. For example, there are women living near towns who must sell a cup of milk from their single cow every day to buy a survival ration of grain for the family; this would not be possible except for the favourable terms of trade (both on a cash and per unit energy-yield basis).
Most Boran will be unable to fully compensate for increasing poverty by becoming agropastoralists. This is not because of lack of rainfall but because less than 12% of the land (i.e. valley bottoms and swales) may be cultivated sustainably. It is postulated that the people seek to grow grain and diversify herds to include more small ruminants to avoid selling cattle at a rate that would deplete their assets. Small ruminants are more of a cash crop than are cattle and would be a substitute sale item.
The declining economic status of the Boran may result in famine if markets do not offer grain at favourable terms of trade. It may also result in expansion of cultivation onto shallow upland soils, increased felling of trees to sell as firewood and charcoal and possible dilution of the Boran cattle breed. The last results from the Boran trading male Boran cattle for reportedly inferior cows from the southern highlands in an attempt to replace the large number of Boran cows killed during drought.
These negative trends may have emerged because the pastoral sector has been too successful in terms of human reproduction. Now the Boran need links to the rest of Ethiopia to enable them to work their way out of crisis. The free flow of commerce is no longer a luxury for the Boran: it is now a necessity.
6. What can be done to alleviate this situation?
Overview: Despite the daunting nature of the task, the major premise of this research is that the entire system can be managed to (1) promote growth in the livestock sector; (2) alleviate poverty and (3) reduce risks to the environment. This requires an integrated set of policy measures, procedures, participatory development tactics and selective use of technology and management innovations. A focus on technological impact alone is not very useful. Even if there were, for example, a "magic forage," planting it would do little change the fundamental causes of instability and poverty in the system. Development impact is also increasingly dependent on linkages from the rangelands to the outside world, including markets, fuel, veterinary inputs, school teachers and accessible banking for pastoralists. If such dependence is viewed as inappropriate, we must dramatically lower our expectations for impact and let the Boran fend for themselves.
Attacking the problems requires a systems approach that recognises that livestock development has social, economic, biological and ecological dimensions. The time when one agency or a few technologies could improve the lot of the Boran is over. Agencies and government ministries must achieve a common vision of the problem at hand and collaborate on policy refinements. Technical innovations should first be extended with a priority for the pert-urban sector, which is compatible with the modest logistical capabilities of extension and would better meet the needs of a very poor, and growing, segment of pastoral society.
The rate of human population growth is not excessive and is consistent with that of other semi-sealed pastoral groups. We suggest, however, that the root cause of system problems is that too few Boran are able or willing to emigrate from the system to balance the net reproduction rate. This suggests that more must be done to educate the Boran in order to make inroads on their cultural isolation and give them a choice of life-style. If attempts to develop human potential over the past 20 years had been similar to those directed towards stimulating cattle offtake, the need for crisis management today might have been lessened. Lack of emigration also suggests that economic underdevelopment of the urban sector is an important constraint with ramifications for the stability and sustainability of rural production systems in general.
Over the short to medium term, efforts need to focus sequentially on (1) improving food security; (2) reducing risks to animal production and asset accumulation; (3) enhancing livestock production and herd turnover and (4) reducing risks to the environment. Assuming that the first goal can be achieved, attainment of the second goal is the key to everything else. Attempts to reduce risks for the environment will not be fruitful unless human welfare has been improved.
Policies, procedures and technology for food security: Food security is ultimately tied to the human population density. For example, one way to improve the situation in the study area would be to reduce the human population by 50% (i.e. by 39000) and then provide jobs outside the pastoral sector for the 2000 people reaching working age each year. Since the low level of economic development will permit this approach, food security must be dealt with in a step-wise manner. Over the short to medium term, relief organizations must be prepared to stay in Borana. Inter-regional commerce must be allowed to open up to facilitate trade of range livestock for surplus maize from the southern highlands. This involves complex national issues such as increasing cereal production in the highlands and solving regional transport and security problems. Markets for small ruminants should be expanded, since the Boran view sheep as more of a cash crop than are cattle and hence sheep offer better opportunities for increasing incomes rapidly. In the longer term the Government should promote the economic growth of urban centres in the rangelands to provide more local market demand for animal products and future employment opportunities for pastoralists. The future of the Boran is closely linked to the future of such towns as Yabelo, Negele, Moyale and Mega.
Using information provided in the Borana System Study, land-use planners can designate agro-ecological zones and sites within zones suitable for sustainable cereal cultivation in the rangelands. Extension could promote technical and management measures to increase maize yields in these locations and reduce the need of pastoralists to cultivate on more fragile sites elsewhere. Suitable locations include valley bottoms in higher rainfall areas but these may comprise a relatively small percentage of the region.
This is not an endorsement of widespread cultivation in the rangelands, since cultivation may cause soil erosion on upland soils under a variable rainfall regime. Rather, it is a call to increase cereal production selectively to reduce some local risks of famine, particularly for people with poor market access. The only way to control the spread of cultivation over the short term is to forge a regulatory partnership between development agencies and the Boran. Increased maize production on appropriate sites is recommended only as an emergency food-production measure over the short to medium term. We believe that widespread emergence of agropastoralism would be detrimental to the long-term ecological sustainability of the system. Cultivation can ultimately be discouraged through favourable marketing interventions, increasing options for investment other than in cattle and spurring human emigration in an ecosystem management approach.
Interventions to increase milk production from cows should focus on health measures, in particular the use of acaricides to reduce tick damage to cow udders. The problem, however, is that acaricides must be imported. Lack of acaricide use points to the fact that the absence of sustained extension of existing technology is a greater problem than generation of new technology. Despite the rangelands having been viewed as an important source of foreign exchange for the nation, range development programmes are commonly unable to extend technologies because of a lack of access to foreign exchange, because extension is poorly funded or because international procurement is excessively bureaucratic. As with other interventions, the Boran should pay for technological inputs and extension using local currency generated from livestock sales. This will provide a good test of the people's priorities.
Primary attention to topics such as terms of trade of livestock for grain and selected improvement of cultivation is in recognition that the biggest problem in the system over the short term is securing more food energy for people. It is clear that plants extended merely as forages will not be high on the Boran's list of priorities. The best way to introduce new forages is through sustainable cropping systems using dual-purpose legumes, such as cowpea, which produce both food for people and feed for livestock.
Policies, procedures and technology for risk management: There are about six drought-grazing reserves that have been traditionally used as fall-back areas for cattle herds during the early stages of drought. These are now reportedly being routinely encroached upon by people and animals. This is occurring because of over-population and probably contributes markedly to the apparently increased instability of the cattle population in response to drought. These fall-back areas need to be re-established by relocating any residents and their carrying capacity increased through water distribution and forage improvements.
Management plans specific to grazing territories (madda) under resource stress should also be implemented. Madda are highly variable in resource endowments, which requires site-specific resource-use strategies. Reclamation might include bush control, prescribed burning and site restoration using local methods and native forage species. Practices such as regulated charcoal making can generate funds from site reclamation: a profit of about US$ 3200.00 per hectare could be realised from making charcoal from dense stands of otherwise useless Acacia drepanolobium. The difficulty in implementing such projects lies, however, in regulation and the fear that charcoal production would spread uncontrolled. While this is a valid concern, it has been found elsewhere in Ethiopia that pastoralists in fixed territories can effectively regulate harvest of wood products in some situations. It might thus be worthwhile conducting pilot projects to see if the Boran can regulate some profitable aspects of resource use themselves. Development agencies might serve as marketing conduits in this process. The Boran have ample knowledge of which plants are useful for the production system and which are not.
Specific grazing territories that would benefit from improvements in carrying capacity should be identified according to their importance to the resource diversity and stability of the community at large. This could be assessed in consultation with local leaders.
Given that the system is over-populated, we believe that the crucial risk management intervention is one that allows the Boran to hold some of their cattle assets in a non-livestock form. The ability to manipulate the asset function of cattle is vital to efforts to adjust stocking rates and relocate households from sensitive areas. This activity embodies "efficient opportunism" and involves storing a portion of the value of male cattle as simple savings accounts in local banks. There is evidence that some Boran may be doing this already.
During the high-density phase of the cattle population, when about 300000 head of cattle occupy the study area, roughly 25% of the herd is comprised of mature males. About 67% and 25% of these males may be held by wealthy and middle-class Boran, respectively. Mature males serve a valuable traditional economic role but at high stocking rates they also compete with milk cows and other productive stock for limited forage and water. This leads to social stress in the community. The males probably also contribute to sudden system crashes during drought in which many households lose all their cattle Relieving pressure on the system by banking part of the male component each year as cash achieves many system-management goals simultaneously. The tendency, regardless, will be for the Boran to gradually increase the percentage of milk cows in their herds to cope with declining per capita milk production. This will come at the expense of traditional investment potential. Banking livestock capital is offered as a means to help them achieve traditional goals, possibly at a lower risk, despite population pressure.
Constraints to implementing the banking innovation are numerous but are probably less formidable than those involved in extending technical interventions described thus far. Constraints include possible distrust and lack of knowledge of banking among the Boran, cultural mores which work against selling cattle of illiterate people gaining access to the banking system and aspects of national currency management such as inflation which might erode wealth stored in non-livestock forms. Local constraints of banking access could be overcome but this would require creativity and incentives. We envision a portfolio-management approach that recognises the maximum benefit from a mix of assets held as livestock or cash in sequences of years in which production risks vary according to stocking rate and rainfall. Banking livestock capital is proposed as the keystone intervention for managing the system out of famine, poverty and increasing risk of environmental degradation.
Policies, procedures and technology for improved animal production: The notion that animal production is uniformly poor in rangeland systems is not supposed here. Indeed, under near-average rainfall and low stocking rates, cattle productivity can be extremely high. At high stocking rates, however, risk of forage competition becomes paramount and animals are more likely to be in poor condition, give lower milk yields and die.
Because the stocking rate of cattle changes somewhat cyclically in response to drought (see below), cattle productivity also varies in a cyclic fashion in an inter-drought sequence of five to 10 years. This may be relatively predictable. Using various risk-mitigation measures described above, particularly banking livestock capital, would provide a major stimulus to cattle production per head as a result of destocking during years of the high-density phase of the cattle population when stocking rates exceed on the order of 20 head/km2. This is because livestock production and mortality are density dependent, with stocking rate mediating effects of annual rainfall on the population. Any measures that increase animal sales to improve human welfare are also valuable; this includes sales of animals to fund water-development activities and construction of grain stores. These activities also build on traditional values of the community.
Calf mortality in near-average rainfall years is the main production factor that requires technical attention. High calf mortality is attributable mainly to poor calf nutrition due to milk restriction in poorer households and to high incidence of diseases resulting from lower management inputs per calf in wealthier households. Shortage of water is also a problem, one that could be addressed by building cement cisterns in certain situations. Attempts to reduce calf mortality fit the cultural rationale of the people, are more consistent with effective use of small quantities of local resources and need an intervention that the Boran can extend among themselves.
The main intervention to reduce calf mortality is making hay from native grasses to improve dry-season calf feeding. The Boran do not traditionally make hay but pilot trials suggest they can make large quantities of hay of suitable quality after the long rains. Feeding hay rather than the traditional cut-and-carry grass could increase the crude-protein content of calf diets by 60% and the digestibility of the diet by 45% during the dry seasons, both on a dry-matter basis. Grass hay can be supplemented with high-protein native legumes such as acacia fruits and leaves and with cowpea (Vigna unguiculata) hay where this crop is grown.
Performance of calves and small ruminants might also be improved by providing better veterinary services at the "farm gate." Veterinary services have, however, been very difficult to sustain. New efforts to extend health services and calf feeding packages should initially be focused on pert-urban areas, with the Boran paying for services where possible.
In addition to reducing mortality rates, improved calf feeding could increase milk production by prolonging lactations and might also allow pert-urban households to take more milk from the calf and sell it. Improvements in cattle recruitment would only be sustainable over the longer term if offtake is accelerated through commercial links or banking livestock capital. Policy and technical issues are thus inter-linked.
Implications for poverty alleviation and ecological sustainability: In conjunction with attempts to spur human emigration, the policy and technical interventions described thus far should act in tandem to reduce chronic insecurities that now exist in the system. Banking livestock capital is the key to poverty alleviation and facilitating range management over the short to medium term. Opening markets, increasing milk production, encouraging pockets of sustainable cereal production and facilitating human emigration should all act to reduce the threats of widespread cultivation to the environment, but cumulative impacts of different innovations would be felt over different time frames.
7. Why is banking livestock capital proposed as the keystone intervention?
Banking livestock capital is the only intervention that would have large and simultaneous effects throughout the system. It would improve food security and risk management, alleviate poverty and reduce threats of environmental degradation. It would also help reduce the danger of genetic dilution of the Boran came breed by helping stabilise the cattle population, thus reducing the need for pastoralists to trade for highland cows during drought recovery. Banking livestock capital could be a nutritional intervention for cows, an ecological intervention for the plant community and an economic intervention for people. It is thus a classic "system intervention" because it requires an inter-disciplinary knowledge of system function to know when to implement it and why. While pilot projects might attempt to extend this intervention concept soon, additional social and economic research is required to thoroughly assess its implications.
Banking livestock capital may not have been a viable innovation as recently as 10 years ago. However, we believe that it is now viable because of the increased risks to households of holding all of their assets as livestock. This is because of a declining resource base and implications this has for increased system instability.
8. Could banking livestock capital even assist small towns?
Yes. The Boran and the small towns in the rangelands are increasingly interdependent. To the extent that economic development of small towns is constrained by lending capital, banked livestock wealth may have important implications for development of small-scale industries in the urban sector. Conservative calculations suggest that if every wealthy and middle-class household in the study area banked from one to three male cattle each year this would generate about US$ 1.7 million annually from the sale of 14500 head. This represents only a modest percentage of the total male inventory. Animals would, however, have to be traded out of the system for the initiative to help stabilise cattle population dynamics and contribute to creating a sustainable yield scenario.
9. How should interventions be implemented?
Interventions should be primarily directed to deal with two population phenomena: (1) the long-term trend (see above) and (2) inter-drought cycles, usually lasting 10 years or less, which consist of a drought-recovery phase followed by a phase of high stocking density. Changes in cattle stocking rates affect many ecological, agricultural, social and economic dynamics in the system. The high-density phase, with more than 20 head of cattle per square kilometre, is essentially a different system, with different constraints, than that in the drought-recovery phase.
The long-term trend: In 1990, the long-term trend may have been an average system state in which the ratio of cattle to people was about 4.5:1. In 1959 the average ratio is thought to have been near 7.4:1. In 2006, barring a large change in net human population growth, the ratio should be 3.3:1.
The negative effects of this long-term trend in cattle-to-people ratio could be lessened using policies and procedures that increase commerce and allow more Boran to emigrate. For an example involving technical perspectives, the long-term trend militates against such interventions as improved dairy technology to process milk surpluses and aspects of system extensification. The long-term trend is better confronted by focusing on the likes of facilitation of dairy marketing and intensification of some aspects of range management.
The inter-drought cycle, drought-recovery phase and high-density phase: In 1993 we are in the second year of the drought-recovery phase following the 1991 drought. There is a 75% probability, based on rainfall records, that the high-density phase will be reached by 1997, allowing for the effects of one dry year or less on the growth of the cattle herd. Once the high-density phase is reached, however, there is a 50% chance of a population crash during the first three years due to the combined effects of the high stocking rate and the risk of one or more years of below-average rainfall. It is thus suggested that the impact of drought on the cattle population is now as much a function of high stocking rates as it is of below-average rainfall. Without adequate drought-grazing reserves, a modest dip in rainfall will now kill far more came than it would have 30 years ago.
The drought-recovery phase should be characterised by more maize cultivation, pert-urban sales of milk and small ruminants, opportunistic production values of the Boran, high rates of cattle production and the honouring of reciprocal rights of grazing among communities. The high-density phase will be more a time of risk management, negative density-dependent effects on livestock production, grazing-induced bush establishment, increased rates of cattle and conservative production values of the Boran.
In the drought-recovery phase of 1992-96, innovations should be employed that complement immediate food-procurement strategies of the Boran (e.g. obtaining maize or selling sheep and dairy products) or that are dependent on low stocking rates for their success (e.g. site reclamation). During the high-density phase, with its high stocking rates and higher risks of asset losses, efforts should encourage banking of livestock capital, sales of cattle to fund water development and promotion of improved calf feeding and grazing management.
In sum, the development strategy has to be opportunistic. There is, however, a reasonable degree of predictability of system dynamics that could guide planning for system management. Another important implication of this schema is that windows of development opportunity are not static; they may be gradually opening, gradually closing or opening and closing cyclically. At one extreme, some innovations would be adopted, dropped and re-adopted in a cyclic pattern.
10. What are the constraints on system transformation?
In past decades it has been common to blame pastoralists, or harsh rangeland environments, in Africa for the apparent failure of range development projects. However, this research indicates that the main constraints actually lie outside the pastoral sector. Planners and researchers commonly do not understand how the pastoral way of life differs from Western concepts such as ranching. They have also lacked appreciation of the complexity of pastoral strategies and how constraints change over time in response to internal and external pressures. Bottom-up approaches to development have been discounted in favour of exotic technologies and Western ideas. Extension is under-funded and personnel is limited and poorly trained. Access to existing technology is limited. The ability of agencies to monitor and regulate resource use is poor and this leads to blanket prohibition of certain practices that might be useful in some situations. True partnerships among pastoralists and development agencies are not traditional and may be difficult to create. Underdevelopment of the nation contributes to major uncertainties in urban employment, commerce and extension. Bureaucratic decision-making is inimical to the opportunistic nature of range ecosystem management. Lack of coordination among government and development agencies obstructs coordinated policy and technical implementation.
The Boran are open minded and can produce animals very well; they just need some stronger links to the outside world. In contrast to many other African producers, the Boran can create large amounts of capital quickly if it can be better harnessed in a rapidly changing world. In one sense, this is all good news. This is because it discounts the notion that this pastoral system is resistant to constructive change from within.
11. What is the source of this systems approach? Does it have wider applicability?
The dynamic view of system interactions requires a different way of thinking. It has been inspired by knowledge of (1) population ecology and (2) predictable relationships among people, animals and the land. A large dose of participatory Farming Systems Research was also mixed in. Clues from producers and traders gave important insights as to how the system changes from year to year and decade to decade. The approach has no strong roots in traditional agricultural investigation.
This philosophy recognises that the Borana system is at a particular point of change along a general continuum. This point has been passed by other African pastoral groups decades or even generations ago. The Borana system may be unusual in that the linkages between the pastoral and non-pastoral sectors are still restricted. This must change if the system is to avoid internal collapse. There are probably other pastoral groups in remote areas of Africa that have yet to reach this stage of change. Systems at different points along a continuum of change require different intervention concepts. The systems perspective is thus applicable beyond the southern rangelands.
When commodity and systems perspectives are run in parallel, the combined answers help tell you what to do, when to do it and why. Neither perspective alone can do as much. Although it has its own scientific merit in terms of inter-disciplinary integration, the systems perspective has an added advantage in that it assists formulation of local strategies that will have an impact on people. We contend that this framework can be easily modified to suit a variety of animal-production circumstances, including smallholders.
12. This all looks expensive. How can this approach be employed elsewhere at low cost?
The systems approach is expensive but systems studies do not need to be repeated often. The key is that researchers and development people learn to think in terms of interacting system components while designing projects. This may require more inter-disciplinary training. Ideally, research should focus on extracting more systems principles from existing work and testing hypotheses concerning specific interactions. Using a systems approach to assess intervention strategies to be used by development people requires more attention to conceptualisation of the system of interest and perhaps examination of important system interactions using producer-participatory approaches along the lines of Rapid Rural Appraisal.
In sum, it is important to know where a system is at any point in time and what the future trends might be, based on probabilities. The practical outcome concerns how population pressures or outside influences might help or hinder uptake and impact of innovations.
13. What are some other research and development implications?
· Future research priorities involve sociology, economics and ecology in the context of risk management. Routine system monitoring is needed to test hypotheses embedded in the theory of local system dynamics, survey felt needs of the pastoralists, observe shifts in resource use and performance of markets and examine the fate of emigrants. Monitoring could be quantitative or qualitative, depending on research budgets. The theory of local system dynamics also has large implications for monitoring range trends. Given near-average rainfall, herbaceous dynamics may tend to be cyclic in response to cattle stocking rates rather than linear. Establishment of woody seedlings may also be more episodic than continuous, with eruptions of woody seedlings occurring more during high rainfall years in the high-density phase of the cattle population.
· The cattle population tends to exhibit equilibrial characteristics. This is not to say that the system does not change from year to year: it is very dynamic within certain limits. It tends towards equilibrial features because cattle stocking rates reach levels that negatively affect the population. Relatively high rainfall and a dominance of perennial vegetation increase the likelihood of periodically intense plant-herbivore interactions. Stocking rate is a crucial filter that affects the response of the population to variation in rainfall. Both equilibrial and non-equilibrial systems may exist in East Africa, with non-equilibrial ones occurring more when systems are less bounded by resource limitations and/or are subject to very low and erratic levels of rainfall. The equilibrial tendencies of the Borana system periodically generate pressure in the system and motivates producers to consider new ways of doing things. Equilibrial dynamics have probably been promoted in the southern rangelands by infrastructure development and population growth.
· The cattle population appears to have substantially modified the environment, causing grazing-induced woody encroachment and soil erosion. However, the extent of impact in relation to a pristine original condition of the study area has not been quantified. Many impacts observable today may have occurred in previous generations. The relatively high rainfall suggests that the region could naturally support a high density of woody plants and thus may have been densely wooded prior to the arrival of man hundreds of years ago. We may therefore now be witnessing the disappearance of a mixed savannah that has been traditionally maintained by people. Woody encroachment has probably been exacerbated recently by human population growth, which encourages sedentarisation, and by a Government policy that prohibited range burning from 1974 to 1991. Although the ultimate effects of encroachment by woody plants are probably detrimental to cattle grazing, intermediate stages may have a variety of positive, neutral and negative effects. We believe that, under proper management, many areas encroached by woody plants could be reclaimed. Cereal cultivation on upland soils may, however, pose a greater future threat to the environment than cattle grazing.
· Transfer of nutrients by cattle from grazing areas to encampments has been implicated in some aspects of environmental change here. It has also been hypothesised that leaf fall from woody vegetation is an important contributor to the replenishment of soil nutrients on overgrazed sites. These questions require fur her investigation.
· Preliminary information on native plants and wildlife gathered by during the System Study provides a baseline from which to begin to address biodiversity issues. Biodiversity research is, however, a low priority until the human crisis can be alleviated.
· Borana leaders are aware of problems associated with the high human population and high cattle rates and of links between heavy grazing and environmental degradation. They make political proclamations intended to protect their natural resources. The extent to which the traditional leadership can effectively control resource use today is, however, unclear.
· The critical measure of system sustainability is per capita production of milk and male animals as assets. Trends in both now appear to be in precipitous decline because the human population continues to grow while the cattle population is increasingly limited by land availability. Until this situation is dealt with, other aspects of improving social or ecological sustainability must be a lower priority for development.
· Human population dynamics and their effects on the system are poorly understood. The observed increase in the net population growth rate may be due to increased availability of relief grain and health inputs. It may be due also to less adherence to traditional social norms that have regulated reproductive behaviour in the past and/or to a temporary decline in the effectiveness of reproduction rules that are dictated in the Gada generation system. For example, fewer people may be affected by certain Gada rules today than in the past because of recent shifts in the age composition of the society.
· As with other pastoral systems, the Borana system is more efficient than commercial ranching in terms of food-energy yield per person and per unit area. This is largely be cause of the high stocking rates and inclusion of milk as an output. The high stocking rates of pastoral systems can, however, pose greater risks for the environment and for system stability than does ranching. As elsewhere, the ranching concept has failed in Borana Ranch lands are in the process of being transferred back to the Boran. There are several reasons for this failure here, but the key factor is that commercial ranching is fundamentally inimical to the Borana production rationale.
· It is hypothesised that rapid growth in the regional cattle herd (i.e. from less than 10 to over 25 head/km2 during just a few years in the drought-recovery phase) leads to a sequence of nutritional constraints on cattle in successive years (i.e. from minerals to crude protein to energy). This would undermine the general rule of thumb that shortage of crude protein in dry periods is the most common nutritional constraint in the semi-arid zone at all times. Mineral and protein supplements may be more effective at lower stocking rates than at high stocking rates. At higher stocking rates, when competition for forage among cattle limits energy intake, the best intervention is destocking that is compensated along the lines of banking livestock capital.
· Boran cattle show considerable compensatory growth in recovering from early restriction of milk intake as calves. Under experimental conditions, reducing the amount of milk consumed by calves by 170 litres had no long-term effect of the productivity of the animals. Under ranch conditions, adult cattle also are able to compensate for weight lost because of restricted water intake during dry seasons. However, high stocking rates probably periodically hinder compensatory growth in calves and adults. Feeding calves forages to compensate for milk deprivation and to accelerate growth over the long term is not useful because weight advantages can be easily lost and are thus risky. Feeding to reduce calf mortality is much more viable. Ideally, small-scale water development should accompany improved feeding systems for calves, because seasonal lack of water limits dry-matter intake. Studies of the allocation of extra water available from the use of cement cisterns suggest that the major impact of increased availability of water for households is on the calves, not people.
· Besides a primary focus on dual-purpose legumes to provide food for people, forage improvements should focus on the best native grasses and woody species. Some of East Africa's best native forages are found in the southern rangelands. The pre-occupation of past development efforts with exotic forages is unwarranted and resulted from ignorance of the possibilities of using native plants. Trials with herbaceous exotics also have been disappointing; it appears that low rainfall and/or cool temperatures are major constraints to establishment and growth of Stylosanthes hamata cv Verano in particular. Despite their often low and variable productivity, native species are valuable because of their proven persistence in an often difficult environment and because valuable species are already recognised as such by the Boran.
· The dynamics of the Borana system can be most comprehensively described in the context of three states: (1) drought, (2) the drought-recovery phase and (3) the high-density phase. From 1980 to 1991 there were roughly four years of drought, four years of drought recovery and three years of the high-density phase.
· The 1983-84 drought contributed to the death of 45% of the milk cows, 90% of the calves and 22% of the mature male cattle in five encampments. This testifies to the risk-mitigation role of hardy, mobile males during drought. More-productive cows may also be more vulnerable to starvation than are poorly productive cows. Drought may thus undermine attempts to upgrade the genetic base of the regional herd. The main value of camels to pert-urban households during drought appeared to be their ability to maintain long lactations. This allowed households to sell milk to buy grain at favourable terms of trade. Cow milk was also sold but in smaller quantities. Dairy products were in very limited supply during the drought and were the only livestock product for which terms of trade for grain did not decline precipitously.
· In the near future, management of drought effects will still rely largely on provision of relief grain. It is envisioned that proper system management would reduce the negative effects of drought that result from high stocking rates and human over-population. Important measures over the short to medium-term could, however, include strategic restoration of drought-grazing reserves, provision of public-works jobs during drought, opening of market linkages between the highlands and lowlands and creation of grain stores both on a household and a regional basis. Banking livestock capital would also be expected to improve the commonly poor terms of trade between livestock and grain during drought. Considering recent patterns of prices and annual terms of trade, if the value of animals were banked during the inter-drought cycle and cash were withdrawn to buy grain during drought, households would on average liquidate only one-third of the animal assets otherwise needed to purchase grain to endure a two- or three-year drought.
· Male camels are important for hauling grain, construction materials and other goods from the market to distant encampments. Given their browsing habits, camels and goats are ecologically more compatible with cattle than are sheep, since both cattle and sheep are grazing animals and would compete for the same feed supplies. There is no evidence to link woody encroachment with camels or goats. Diversification of herds to include more camels and small ruminants is probably most dependent on improving access to camel markets and on veterinary extension; these interventions could be pursued.
· These observations suggest that agropastoralism, herd diversification and peri-urban dairy marketing have evolved in East Africa because of extreme pressure on the traditional livestock system as a result of human population growth. They do not, therefore, necessarily represent improvements in human welfare or an enhanced system state. Rather, they represent indigenous mechanisms for avoiding starvation and/or asset depletion. The development goal for the Borana system should be sustainable, extensive pastoralism, not agropastoralism.
· Borana society is economically diverse and clearly stratified into wealthy, middle-class and poor components. There is also a rural sector and a pert-urban sector (i.e. within 30 km of towns) which are believed to behave differently from each other in social and economic terms. The concept of an overall average household thus has little meaning and does not offer much utility as a research tool.
· Two independent surveys showed that 20 to 25% of household heads were women, particularly in the poor and middle-class strata. It is hypothesised that a higher percentage of households heads are women in the pert-urban sector. There is no information on whether a larger percentage of households are now headed by women than in the past or if the roles of women in the society are changing.
· Women reported that they worked long hours all year but direct observation suggested that about 30% of their time in the warm dry season consists of activities associated with "leisure." An extension trial indicated that women apparently had enough "free" time at the end of the long rains to make up to 300 kg of grass hay. Whether or not production innovations are adopted probably depends largely on social values and whether the women are willing to give up some of their leisure time.
· The perverse supply hypothesis might be valid here for poorer households who are forced to sell cattle That is to say that the Boran will seek to sell livestock in markets with higher prices but that the primary reason they do so is to minimise the number of animals sold. This hypothesis needs to be tested at the household level.
· Although all households will be forced to sell more cattle, informants report that true livestock commercialization in the southern rangelands will emerge from a new class of individuals who have some education, wealth and ties to the traditional sector. Some of these people may have learned the livestock trade while working as government agents. Increased competition and possible conflict between commercial and subsistence herds may occur. One irony is that, in the attempt to gradually guide the traditional pastoral system towards enhanced productivity, sustainability and social welfare, the Government may have to intervene to protect the subsistence population from intense competition with commercial producers.
· The Boran are innovative and have pioneered some of their own resource management concepts such as planned grazing allocations and fodder banks (kalo). Persistence of traditional leadership structures should facilitate introduction of appropriate interventions in this system.
· In the southern rangelands, bottom-up approaches to development of innovations have been more effective than top-down approaches. This is because the producer's behaviour, values and daily life are more complex than, or counter-intuitive to, that imagined by researchers. Top-down concepts such as the pond scoop, the improved butter churn, pasture improvements using exotic forages, speeding-up calf growth through supplementary feeding or drought fodder banks based on Atriplex and Opuntia spp will not have an impact here for various reasons. Bottom-up concepts such as hay-making, using native legumes, intensified aspects of range management, cement cisterns, banking livestock capital and reducing calf mortality are far more likely to have an impact.
· In this project, the interaction between researchers and development agents proved very fruitful in helping design research that had an impact on development. Development agents provided grass-roots links with the community that provided a better bridge to researchers. This permitted a true systems- research process to evolve.
