Financial analysis of livestock production systems around Lake Mburo National Park, in South Western Uganda

M Ocaido, R T Muwazi* and J Opuda-Asibo**

Department of Wildlife and Animal Resource Management, Faculty of Veterinary Medicine, Makerere University, P.O. Box 7062, Kampala, Uganda
mocaido@vetmed.mak.ac.ug
*Department of Veterinary Anatomy, Faculty of Veterinary Medicine, Makerere University, P.O. Box 7062, Kampala, Uganda
rmuwazi@vetmed.mak.ac.ug,
**Department of Epidemiology, Veterinary Public Health and Preventive Medicine, Faculty of Veterinary Medicine, Makerere University, P.O. Box 7062, Kampala, Uganda
opuda-asibo@vetmed.mak.ac.ug

Abstract

Financial analysis was done for pastoral and ranch livestock production systems around Lake Mburo National Park, Uganda. The agricultural income was from cattle, goat and crop enterprises. The major source of agricultural income was from cattle, accounting for 92.1±4.2% in ranch and 78.4±7.4% pastoral households. Milk and cattle live sales were major outputs from the cattle enterprises, with live sales contributing 47.5±0.2% (ranches) and 54.9±2.2% (pastoral); and milk contributing 47.8±2.5% in ranch and 45.4±2.4% in pastoral systems. Manure contributed 4.4±0.8% and 2.7±0.6% to total output in ranch and pastoral cattle enterprises respectively. The major inputs in cattle enterprises were disease control contributing 44±11% (ranches) and 44.2±5.4% (pastoral); and labour costs contributing 40.7±6.9% (ranches) and 44.9±7.5% (pastoral). The contribution of goats to agricultural income in ranch and pastoral households was 5.5±2.9% and 18±8.3% respectively. Meanwhile, crop contribution to agricultural income was minor, being 2.4±1.3% (ranches) and 3.6±1.8% (pastoral). The major proportion of agricultural inputs were used in cattle enterprises, being 97.7±1.5% in ranches and 92.2%±3.5% in pastoral. The remainder of the inputs were in crop enterprises. The mean annual gross margins from agricultural enterprises for pastoral and ranch households was Ug Shs 6,246,000 ±2,511,000 (1 USD = Ug Shs 1,420) and Ug. Shs 12,371,000±1,680,000 respectively, giving higher annual returns per hectare for pastoral systems (Ug. Shs 108,800±38,000) than the ranching systems (Ug. Shs 16,300±4,400).

Keywords: Agricultural income, Ankole ranching scheme, gross margin analysis, pastoral systems

Introduction

This study was carried out in Uganda in Kiruhura district, Nyabushozi County in Nshara Dairy Cross Breeding Ranch (DCBR), Ankole Ranching Scheme (ARS) and smallholder pastoral communities settled in the ranches and communal grazing areas surrounding Lake Mburo National Park (LMNP). Lake Mburo is a small national park (26,000 hectares (ha)). Livestock ranching production system consisted of 50 cattle ranches of Ankole Ranching Scheme  (777 ha) and Nshara Dairy DCBR (6,475 ha). Nshara DCBR was set up by government of Uganda to provide farmers with F₁, dairy breed heifer crosses for farmers so as to improve the national dairy herd (Nsubuga 1977).  The main objective of the setting up Ankole ranching scheme was for commercial beef production. About 856 pastoral households were settled around LMNP, each occupying on average 39.8 ha of land.

Currently, the Lake Mburo ecosystem looks very fragile due to increasing pressure on natural resources due to increasing human populations and livestock numbers, leading to increased land fragmentation, encroachment of the National Park and rangeland deterioration (Ocaido et al 1996; Schwartz et al 1996; Kisamba–Mugerwa et al 2006). Elsewhere, the fragility of integrity of semi-arid rangelands have been observed among the Masai in Kenya and Tanzania (Musimba and Nyariki 2003; ILRI 2006; Mworia and Kiyamario 2008; Baldwin et al 2008; Curtin and Western 2008).   To maintain the integrity of Lake Mburo ecosystem it is therefore necessary to identify the most profitable enterprises or forms of multiple land use that can take care of human needs, more specifically food security and increased household incomes as well as conserving the natural resources. No work had been done to assess financial performance of different livestock enterprises in this geographical location. It was against the above background that this study was initiated.

Materials and methods 

Study area description

The study was done in Kiruhura district, Nyabushozi County comprising of smallholder sedentary Bahima pastoral livestock keeping communities and ranches around of Lake Mburo National Park (LMNP). The area lies in a low rainfall belt (700-800 mm per annum) with bimodal pattern (Monday 1991).  The long rains fall between March and June, while short rains are experienced from October to December (Monday 1993; Ocaido 1995).  However this area is characterised by severe dry seasons and rains tend to be unreliable.  The average temperature recorded for the area is 27.5⁰C with daily variations ranging from 21.5⁰C to 34.0⁰C (Monday 1993).

This area has a long history of livestock grazing for over 500 years by Hamitic tribes  (Randall 1944; Rollinson 1962) that led a semi- or fully nomadic life in search of pasture and water. At one time, this area became infested with tsetse flies and became deserted. But in 1966, after eradication of tsetse flies, the government decided to utilise the 72,000 hectares (ha) of the reclaimed land for cattle ranching with the help of United States Agency for International Development (USAID) funding (Anon 1988). Ankole Ranching Scheme (ARS) with fifty ranches  and Nshara Dairy Cross Breeding Ranch (DCBR) were set up.  Nshara DCBR was to provide farmers with F₁ dairy breed heifer crosses  for farmers.  The main objective of the ARS was to obtain a large block of commercially viable and economically productive beef ranches.  By 1978, this objective was achieved with most ranches fully developed with peak stock numbers (Anon 1988).  After 1979, there was a breakdown of veterinary services and general economic malaise due to war hence this situation was reversed leading to decline of most ranches.

Meanwhile, the pastoralists continued to occupy part of the reclaimed land, which was a game reserve until 1982, when this area was gazetted as a national park.  In 1986, the size of the park was reduced to resettle pastoralists who were forcefully evicted from it at the time of its creation.  Despite the eviction, some pastoralists remained in the park and many others were squatters in the ranches.  This prompted the government in 1990, to set up a Ranch Restructuring Board (RRB) to readdress this problem.  The RRB then decided to reduce the size of most ranches by 40% (from 1,295 ha to 777 ha) to cater for the pastoralists. Currently, these people have been resettled on these pieces of land (Schwartz et al 1996). Follow up studies to determine the financial performance of these livestock enterprises had not been done.

Data collection

Data was collected by carrying out both cross sectional and longitudinal studies. Initially, a reconnaissance survey was done. Here, focus group discussions were held with sedentary pastoralists and ranchers using rapid appraisal methods with aid of a checklist of questions. Later a structured questionnaire survey and longitudinal study was done in Nshara DCBR, 39 ranches of Ankole Ranching Scheme (ARS) and 89 sedentary pastoral households (out of 856 registered households). The sample size was determined at 95% level of confidence using the equation (Dohoo et al 2003; Thrusfield 2007):
 

Where:

Q= 1 – P

Z =  1.96

e = Confidence level  = 0.05;
P = Prevalence of TBDs abut 7%

A same approach was used in the ranch sample size determination. Due to the limited number of ranches (50), a finite sample size adjustment for new ranch sample size n₁ was determined using the formula given below:

Where

n= calculated sample size,
N= Total population size. 

Thirty-nine ranches were randomly selected from 50 ranches of ARS. From each of the ranches, 2-3 of sedentary pastoral households were randomly selected.

Pastoral cattle herds were categorised as small when herd size was below 40 heads, medium (40 –70) and large (>70 heads of cattle), while ranches were divided into two groups: small (below 300 heads) and large (³300).  The number of the different herd sizes to be used in the study was proportionately allocated to the required sample size according to their proportion in the population. Of pastoral households selected 33 were small, 25 medium and 31 large. For ranches, 21 were small while 18 were large sized. All goats herds were pool monitored.

Later, a longitudinal study was done for 18 months. Stock owners were taught on how to keep health and production records under supervision of extension workers. Health and production data collection forms for livestock enterprises (cattle and goats) were designed for the farmers. Data collected included: livestock production parameters (livestock herd structure, calving intervals, calving rates, first age of calving;  and calving / kidding intervals); livestock enterprise yields parameters (milk yield, traction, annual herd growth;  livestock eaten, battered, given out as gifts or stolen); crop enterprise yields like crop harvests; output in terms of sales of crops, milk, beef and livestock (cattle, goats); prices of outputs (prices per litre of milk,  meat, skins / hides, livestock age specific group sold and traction per hectare); prevalence of livestock diseases of common diseases; age specific morbidities and mortalities due to common livestock diseases; losses due to common diseases like: losses caused by mortalities, chemo-prophylaxis costs, weight loss, milk loss and traction loss and disease control costs (vaccination, tick control (dipping, spraying and labour), antihelmintic treatment and labour in controlling diseases); other inputs incurred like labour, bush clearing, interest paid on loans acquired; prices of inputs; sources of inputs and how they are accessed and gender labour inputs in control of diseases.

Data analysis

For financial analysis gross margin analysis was performed. Gross margin (GM) was taken as the gross income of an enterprise less variable costs (Putt et al 1987). It can be represented by the equation as:

GM =Σ (yi*py_i) - Σ (x_i*px_i)

Where:

Σ (yi*py_i) = gross income

y_I  = quantities of the products.

py_I  = respective prices of the products.

Σ (x_i*px_i) = total variable costs

x_i  = quantities of costs

px_i= respective prices.

GMA was performed using spreadsheet models developed using Microsoft Excel version 5.0 (Microsoft Corporation, USA). It was used for comparing the profitability of pastoral farms, ranches and enterprises (crop and livestock) under the present management systems at different farms or ranches. When performing GMA for crop enterprises, the gross income was taken as quantities consumed by household, sold for income and donated out valued at market prices. Meanwhile, variable costs were taken as quantities of inputs including family and hired casual labour to perform activities like herding, bush clearing, crop planting and weeding valued at their market prices.

Similarly, GMA for livestock enterprise, the gross income was taken as quantities of milk and livestock sales, manure, annual herd growth and quantities of products and livestock donated or consumed by the family and workers valued at market prices. Variable costs for livestock enterprise were taken as disease control costs and other inputs like feeding and labour. The value of family labour was determined by its opportunity cost, which was taken to equal the most enumerative alternative that the family labour could earn. In valuing family labour, the total family labour on livestock and crops from children and adults was converted into man-hour equivalents, which were the multiplied by the prevailing hourly market wage rate. For quantification of manure output, it was taken that an African Livestock Standard Unit (LSU) cattle was estimated to produce 394 Kg of manure annually valued at 4USD per tonne (McLeod 1998).

Total farm agricultural income / output was taken as a sum of income from crop and livestock enterprises. Economic indicators mainly GM (return) per unit resource and resource productivity were determined to give economic efficiency for each resource being utilised. Resource productivity was calculated by dividing total farm agricultural income by the units of each resource like land (hectares), capital and labour used for production during the study period.

Results and discussion 

Livestock production system description

An average ranch was 813.2 ±51.8 ha. The ranches had a mean number of cattle of 288.6±46.2 and goats 9.3±3.48. The small herd ranches had an average cattle number of 250±28.9 and large ranches had 558±14.8. The mean pastoral household size was 12.9±1.1 located on an average of 39.8±4.04 ha of land. The pastoralists had a mean number of cattle of 66.9±8.5 per household at a stocking density of one cattle per 0.56 ha of land. The mean herd size of cattle in pastoral small herds was 23.8±3.2 in 30±4.9 ha of land; 63.4±7 of cattle in 44.8 ha of land in medium sized herds and 87±3.7 cattle in 46.7 ±7.8 ha of land in large sized herds. There was increasing cattle stocking density with herd size implying that despite settling pastoralists on pieces of land, the concept of “tragedy of commons” that is open access grazing (Hardin 1968 and 1988)  was still being practised. This agrees with what was observed by Nge’the (1993) and Mworia and Kinyamario (2008) among the Masai in Kajiado district in Kenya where the pastoral households with large cattle herds refused to honour the recommended stocking rates.  

Forty one point two percent of the ranchers and 77.5% of pastoral households kept goats. Pastoralists with small cattle herds had an average of 7.5±1 goats, medium had 16±5.1 and large had 10.6±0.4.  Pastoral communities keep goats because they want to diversify animals that they keep as a form of security against drought (Scoones 1995; Njeru 1996; Mworia and Kinyamario 2008).

Sheep and chicken kept were insignificant, averaging 0.5±0.3 and 0.06±0.057 per pastoral household respectively and 3.4±2.4 and 0.07±0.03 respectively per ranch. This was because their consumption was unpopular and a taboo among the Bahima people (Ntozi 1995). This was similar to Zimbabwe, where sheep keeping was an unimportant activity among African communities because consumption of mutton was unpopular (Ndlovu 1993). This differed with pastoral and agro-pastoral communities in Eastern Uganda, where chicken keeping was popular, with 22±2.3 birds per household in Teso, Soroti district (Opolot 1998) and 11±1.2 birds per household in Karamoja, Moroto district (Lokii 2000).

Thirty seven percent and 88.6% of ranchers and pastoralists respectively practised some crop farming. Average land for ranchers and pastoralists under crops was 1.01±0.7 ha and 0.6 ±0.08 ha respectively. The crops grown were bananas, beans, maize, fruits and cassava. Crops were grown purely for subsistence. These observations were also true for pastoral systems elsewhere (ILRI 1995; Leonala et al 1995; Njeru 1996; McDermott et al 1999) and most mixed crop-livestock systems (Sansoucy et al 1995).

Economics of cattle production

The cattle herd structure in ranch and pastoral herds were as shown in Tables 1 and 2 respectively.  The cattle herd structure reflected that both ranchers and pastoral households were geared towards increasing herd sizes as shown by high numbers of cows and low numbers of steers and bulls. The herd sex ratio of bulls to adult female cows of 1:26 for pastoralists and 1:35 for the ranches was typical of the one seen in Masai pastoral herds in Kajiado, Kenya (Nge’the 1993). This cattle herd composition was normal among pastoralists but not among ranchers. Usually among the pastoral households bull calves and steers are sold or eaten leaving behind the females for breeding and production of milk. This practise was unusual with the ranchers who would have retained steers for fattening for sale for meat. The cattle herd composition among ranchers suggested an attempt to increase their herd sizes near the optimum sustainable carrying capacities of their land, especially among those with small and medium sized herds. Meanwhile, in agro-pastoral areas of Soroti district in Uganda, cattle herd composition was usually geared towards supporting traction with bulls and steers constituting over 36.4% of the herds (Ocaido et al 2005).

Inputs in cattle enterprises

The variable costs in cattle keeping enterprises were incurred in disease control, feeding, labour, repair of farm implements and structures, and purchase of fuel. Fuel costs were only incurred in Nshara DCBR.

The mean disease control cost per head of cattle per annum was Ug. Shs 8,000± 1,200  in the ranches and Ug. Shs 6,960 ±870 in pastoral herds. Details were as shown in Table 3.

It was shown that the mean percentage contribution of tick control costs to total disease costs were significantly higher (p<0.05) in pastoral herds (74.8±5.8%) than in the ranches (61.7±3.4%). ECF and tick related diseases constituted 53.1% of the total costs of  clinical cases treated in Nshara DCBR.  Meanwhile, in other pastoral and ranch cattle herds,  ECF was the major clinical problem constituting to about  98.1% and 95.8% of total treatment costs of clinical cases respectively. The treatment of scouring in calves constituted the remaining 1.9% and 4.2% of the total treatment costs in other ranch and pastoral herds respectively. Helminthic control costs constituted 12.4±1.8% and 9.6±3.3% of total disease costs in the ranch and pastoral herds respectively. Vaccination costs constituted 11.2±1% and 6.1±2% of total disease costs in the ranch and pastoral herds respectively.

The labour costs were incurred in bush clearing and herding of animals. Herding was a major labour cost. With Lake Mburo rangelands being invaded by weed plants inter alia  Acacia hokii and Cymbopogon nardus, bush and weed clearing had become an important routine activity. Family members contributed labour in dipping and spraying of cattle in both ranch and pastoral cattle production systems. There was a very highly significant positive correlation of the number herdsmen employed (r=0.85, p> 0.001) with the number of cattle kept. The number of herdsmen employed could be predicted from the regression equation: Y = 0.03X – 1.91 ( p=0.0000). Where Y = Number of herdsmen,  X = Number of cattle in the ranch. This means that no herdsman was employed if the number of cattle were less than 96. If the number of cattle were below 96, cattle keepers depended entirely on the family labour.

Managers were only employed by the ranchers.  The number of managers employed had a highly significant correlation   (r=0.56, p<0.01) with the number of cattle kept; and could be predicted using the regression equation: Y = 0.00147X -0.255 (p<0.01). Where: Y = Number of managers employed, X = Number of cattle kept. This means that no manager was employed if the number of cattle was less than 506. If the number of cattle was below 506 the management of the ranch depended on the family entrepreneurship. The details of the costs of each labour activity were as shown in Table 4.

Herding was a major labour cost contributing 93.1±2.3% and 80.5%±5.8% of total labour costs in pastoral and ranch cattle enterprises respectively. Meanwhile, bush clearing constituted 6.9±2.3% and 18.3%±3.5 of the total labour costs in the pastoral and ranch cattle production enterprises respectively.  The mean labour cost per head of cattle per annum in the ranches was Ug. Shs 4,300±600 and Ug. Shs 8,200±3600 in pastoral herds. Nshara DCBR, besides bush clearing and herding, incurred 30.4% of the total labour costs as payment for salaries of other employees.

The feeding costs were mainly incurred in provision of mineral lick supplements.  However, in Nshara DCBR, the breeding bulls were additionally given concentrate feed supplement (maize bran) during the dry season. Each cow in the ranches or pastoral herds was consuming mineral licks equivalent to Ug. Shs 1,200 per annum.

The repair of farm / ranch infrastructure included: repair of fencing (wire, live plants or wood) for paddocks, perimeter enclosures and homesteads. In addition, infrastructure repairs were done on buildings and crushes. Repairs of farm / ranch implements  included bicycles, dip tanks  and  spray pumps. Total costs incurred in repair of farm / ranch implements and infrastructure were as shown in Table. 6.  About a quarter of the fencing of the Nshara DCBR was destroyed by wild animals, annually. In pastoral households, most of the costs in repair were on buildings, homestead fences (wood poles / pegs) and bicycles.

Fuel and vehicle maintenance costs  were only incurred in Nshara DCBR as other cattle keepers used bicycles for transport. Fuel and vehicle maintenance costs constituted to  11.9% of the total variable costs in Nshara DCBR

Cattle enterprise outputs

Outputs from cattle enterprise were from sale of live cattle and milk; and herd growth. No income was generated from sale of meat. Cattle were rarely slaughtered for home consumption. Only one household slaughtered one steer for home consumption which was valued at Ug. Shs 60,000.  The steers were mostly sold to local butchers in the area along the high way to the city. They sold beef  at Ug. Shs 1,300 per Kg to travellers.  The adult cattle were sold directly to cattle traders in Kampala, who transported them on lorries to Kampala.

There was no significant difference (p>0.05) in off-take rates of cattle in the pastoral herds and the ranches. The mean off-take rate were 20.7±5.3% in the ranches and 28±3.5% in pastoral herds. This was contrary to what was estimated by the cross sectional survey to be 10.4±1.43 %. The high off-take especially observed with pastoral households was due to drought which occurred during the study period so cattle were hurriedly sold. The mean percentage age specific composition of cattle live sales and contribution to live sale output; and live sale prices  in the ranches and pastoral herds were as  shown in Table 5.

There was a significant difference (p<0.05) between the percentage composition of steers sold than any other age group category of cattle. Just like any pastoral system (ILRI 1995), the steers were the major category of cattle sold. Once again, this practise shows that the ranches were not geared towards maximising meat off-take.

During the study period, there was a short spell of drought for 4 months, where cattle were sold at salvage prices. Bulls were sold for Ug. Shs 65,000 (0.4 of its original peak market value); adult cows for Ug. Shs 25,000 (0.1 of its peak market original value), heifers for Ug. Shs  30,000 (0.32 of its peak market original value) and steers sold for Ug. Shs 20,000 (0.32 of its original peak market value). This was true with pastoral areas elsewhere (Toulmin 1995; Morton 2006).  About 55.6±7.8% and 5.5% of cattle in pastoral and ranch herds, respectively, were sold during this period. The percentage of cattle sold during drought period was significantly higher in pastoral herds than in the ranches (p<0.05).

On average 0.88±0.05 litres and 2.06±0.035 litres was produced by each lactating cow daily during the dry and wet seasons respectively. The mean farm gate price of milk per litre was Ug. Shs 212.8±6.8 during wet season and Ug. Shs 286±4.56 during the dry season. Most milk was bought by vendors directly from the households. These vendors then sold their milk to milk collecting and cooling plants in along the city highway who in-turn sold the milk to Uganda Dairy Co-operation or retail to local urban consumers travelling along the highway. Daily per capita milk consumption for  households was found to be 0.68±0.1 litres, an equivalent of Ug. Shs 74,460 annually.

Cattle first calved at an average age of 2.5±0.2 years under ranch management and 3 years under communal pastoral grazing management The mean calving interval was 14.8±0.3 months. The mean overall calving rate for cattle herds around LMNP was 61.6±3.1%. Specifically, the mean calving rate in pastoral and ranch herds was 60.5±4.5% and 63.9±0.6% respectively. Nshara DCBR had a calving rate of 62.9%. 

Herd growth and herd growth value depicted the change in number and value of the cattle in a herd respectively, from the beginning and at the end of the study. The average percentage herd growth by the end of herd monitoring exercise was found to be 7.7±05% in pastoral cattle herds. Meanwhile, the mean percentage herd growth value with exception of Nshara DCBR was  –17.7±6.4% and –11.4±4% in pastoral and ranch herds respectively.  The observed decrease in herd value was due to drought which killed high value adult cows. Meanwhile in Nshara DCBR, there was an increase in both the physical herd growth and value. There was 123.9% increase of the value of the herd in Nshara DCBR. This was because Nshara DCBR was not severely affected by drought since they had relatively more pasture and developed water catchments (valley tanks / dams) that had water throughout the drought period.

There was no organised use of manure. Manure was left to naturally re-cycle in the rangeland thereby fertilising the soil. Cattle were not being used for traction

Gross margin analysis (GMA) of the cattle enterprise

The details of gross margin analysis were as shown in Table 6. In both ranch and pastoral systems, the major inputs in cattle keeping enterprises were disease control and labour costs. There was no significance difference (p>0.05) in their average overall percentage contribution to total variable costs in the ranch and pastoral herds. In the ranch and pastoral cattle enterprises, the mean overall percentage contribution of  disease control to total variable costs was  44±11% and 44.2±5.4% respectively.  While the labour costs, contributed 40.7±6.9% and 44.9±7.5% of total variable costs in ranch and pastoral cattle enterprises respectively. The labour costs were significantly higher in pastoral households with small cattle herds, constituting 60% of the total variable costs. There was a negative correlation between percentage proportion of labour costs to total inputs in livestock production with the cattle herd size holdings (r= -0.90, p<0.0). Always, labour costs are quite high for smallholder cattle keepers in sub-Saharan Africa, constituting about 40% of total inputs (Sansoucy et al 1995; McDermott et al 1999). In the agro-pastoral areas of Soroti, Uganda, labour costs contributed 65% of the total cost of agricultural inputs (Opolot 1998). The cattle pastoral system around LMNP was labour intensive, involving use of family labour for herding, milking, watering and spraying of animals. The major labour cost was herding constituting to 80.5% and 93.1% of the total labour costs in ranch and pastoral enterprises respectively. Bush clearing was the second most important labour activity because the rangeland had been invaded with weed plants: Acacia hokii and Cymbopogon nardus (Schwartz et al 1996). Ranchers were investing more in bush clearing than with pastoral households. The average cost of labour per head of cattle was quite high for pastoralists (Ug. Shs 8,200) than the ranchers (Ug. Shs 4,300). These findings agreed with those of ILRI (1995), Sansoucy et al (1995) and McDermott et al (1999). This difference in labour cost per animal was brought about by the economies of scale enjoyed by ranchers with larger herds. This explains why the labour cost per head of cattle decreased with increasing herd size among pastoralists (see Table 6).

However, in Nshara DCBR, the labour costs were even higher (53.1% of the total variable costs and cost per head is 25,300 per annum). This was because in addition to costs of herding and bush clearing, there were costs incurred for payment of salaries of the ranch manager, veterinarian surgeons, drivers and watchmen.  To cut labour costs per head of cattle, it is therefore necessary to increase the herd size of Nshara DCBR to a maximum optimum sustainable level so that it can also enjoy economies of scale.

In ranch and pastoral systems, disease control costs were 44% and 44.2% of the total variable costs respectively. The average annual cost for controlling diseases for both pastoral and ranch herds was Ug. Shs 6,900. Tick  control was the major disease control  activity followed by chemotherapy and vaccinations (see Table 3). Tick control constituted 61.7% and 74.8% of total disease control costs in pastoral and ranch herds respectively. ECF was the major disease treated in the ranch and pastoral cattle herds. ECF and tick related problems constituted 53.1% of total disease control costs in Nshara DCBR.  Meanwhile, in the ranches of ARS and pastoral herds, the costs of treating ECF and tick related problems constituted 98.1% and 95.8% of the total disease chemotherapeutic costs in the pastoral and ranch herds respectively. The lower percentage of disease control costs attributed to treatment of ECF and tick-related diseases in Nshara DCBR was because more costs were incurred in treatment of scouring and dystocia. Besides, in Nshara DCBR, there was efficient health service provided by resident veterinarians. Whereas, the ranches of ARS and pastoralists depended on a few and rarely available private para veterinarians and veterinarians. Dystocia was common in Nshara DCBR, because the small Ankole cows / heifers were bred with larger Friesian bulls resulting in bigger calves that were difficult to deliver.

With exception of Nshara DCBR, there was no significant difference (p>0.05) in the mean percentage overall contribution of milk and cattle sales to total output from cattle enterprises in the ranch and pastoral herds. It was found that the overall contribution of cattle sales to total output of cattle enterprises was 47.5±0.2% and 54.9±2.2% in ranch and pastoral herds respectively. Whereas milk was contributing 47.8±2.5% and 45.4±2.4% of total output from cattle enterprises in ranch and pastoral herds respectively. This showed that the current livestock production systems for both pastoralists and ranchers were not solely for beef production, but had a dual strategy of producing both milk and beef. This strategy was true for pastoral systems elsewhere (Kerven 1987; Perrier 1995). This strategy does not fall within the goals of beef ranch production where the aim is increased off-take of beef animals (Jalvis 1993) or the reasons why the Ankole Ranching scheme was established (Nsubuga 1977). This finding disagreed with former donor and governments assumptions that in pastoral areas, increased off-take of live animals could improve the welfare of pastoralists and therefore be adopted by them. In most pastoral areas, where ranching schemes were introduced with aim of increasing live animal off-take, the projects failed. This was because the projects never considered the interests of pastoralists on milk production. A survey carried twenty years later, after the ranching scheme was introduced among the Kenyan Masai, showed that the Masai continued to manage cattle for milk (Kerven 1987; Curtin and Western 2008). Similarly, Niamir (1982) observed that one of the reasons why livestock production development schemes for the Dinka in Sudan aimed at increasing beef production failed, was because they ignored importance of milk. This failure of livestock pastoral development projects is not confined to Africa alone. In Afghanistan among the Sheikhanzai, the World Bank project also failed, because it was introduced with the aim of increasing the off-take of beef animals forgetting the importance of milk to the community (Tavakolian 1984).

Manure was contributing to 4.4±0.8% and 2.7±0.6% of total output from cattle enterprise in ranch and pastoral herds respectively. Elsewhere, in other pastoral and ranch areas there have been no organised studies to quantify the contribution of manure to total output from cattle enterprises. However, in mixed crop and livestock systems, studies have been done to quantify benefits of manure to crop production. For example in Embu district in Kenya manure was found to be the most important output of cattle enterprises (Lekasi et al 1998) and in districts of Kiamba and Murang’a  manure was considered the second most important output to milk (Lekasi et al 2001).

In Nshara DCBR, the only source of agricultural income was from cattle. Outputs from cattle enterprise were from milk, live cattle sales, herd growth and manure (see Table 6). According to Nsubuga (1977) the major aim of establishing Nshara DCBR was to produce female F1 dairy crosses for sale to the farmers but the output from live sales was very low only 8.6%. This was because the ranch was just recovering from vandalism of the past wars. Income from live sales came from in-calf heifers and steers. Meanwhile, the annual returns per ha of land was higher than the returns per hectare in the ranches of Ankole Ranching Scheme. On contrary, returns from pastoral system were four times higher. The returns per ha of land was expected to improve when the ranch becomes fully operational.

Economics of goat production enterprise

The mean herd structure of goats in ARS was as shown in Table 7. The mean annual kidding rate was 194±3.1%. Meanwhile, the mean overall mortality rate was 6.06±3.24%. The overall average goat off-take rate in ARS was 37.7±4% annually, being specifically 39.6±6.9% and 34.7±0.1% for pastoral households and ranches respectively. The goat off-take was in form of live goat sales, goats eaten by the household, goats given out or stolen. The mean percentage age specific compositions of goat off-take were as shown in Table 7.

Little attention was paid to goats which were left to roam in the range.  No treatment, vaccination and tick control was done on goats. The price of adult male and female goats ranged from Ug. Shs 20,000 to 50,000 with a mean price of Ug. Shs 35,900±2,900 per adult male goat, Ug. Shs 35,000±3,100 per female adult goat and Ug. Shs 15,000 for immature male goat. The gross output and gross margin analysis from the goat enterprises were as shown in Table 8. Gross output included value of goats sold, eaten, given out and the value of herd growth.

Economics of crop production

The major input in the crop enterprise was labour with both ranchers and pastoral households hiring labour during the months of cultivation seasons. Elsewhere among pastoral communities in sub-Saharan Africa, pastoralists hire labour to farm their crops (Sylla 1995). The major crop output was from bananas. Due to severe drought there was hardly any harvest during the first season of study.  The harvest was used for subsistence consumption at the household level. Details of gross margin analysis of the crop enterprise were as shown in Table 9.

Agricultural income ranch and pastoral household

The agricultural income in  ranch and pastoral household around LMNP comes as outputs from cattle enterprise, goat enterprise and crop enterprises. Details of gross margin analysis at ranch and pastoral household level were as shown in Table 10.   Livestock keeping constituted 96.2% and 96.5% of total agricultural income in ranch and pastoral households respectively. Livestock contribution to household income has been found to be higher in most mixed livestock crop systems in Ethiopian highlands (Gryseels 1988; Sancoucy et al 1995; Omitti 1995). In Mali, in smallholder mixed crop-livestock systems, livestock contributed 78% of total agricultural income (Debrah and Sissoko 1990; Sansoucy et al 1995). These findings agree with Leonala et al (1995) who defined pastoral systems as those systems in which more than 50% of gross household revenue comes from livestock.  Low contribution of livestock to household income is always seen among smallholder dairy systems (Leonala et al 1995; McDermott et al 1999).

The overall mean contribution of cattle to agricultural income in ranch and pastoral households was 92.1±4.2% and 78.4±7.4% respectively. The percentage contribution of cattle to agricultural income in ranches was significantly higher in the ranches (p<0.05) than in pastoral households. The  percentage  contribution of milk towards pastoral household income was positively correlated to the  number of cattle the household owns (r=0.79, p<0.05).   Milk was  important for pastoral livelihood for consumption and sale. In the study area milk sales contributed 35.6% of total household income. This was true elsewhere among pastoral communities. Milk contribution to total household income, was 38% among the Masai (White and Meadows 1981; Kerven 1987), 32% among WoDaabe in Niger (Swift 1984; Kerven 1987) and 38% among the Bella in Mali (Kerven 1987). The mean annual per capita milk consumption among pastoral households in the study area was 248 Kg. Elsewhere, milk contribution to annual Kilocalorie intake among the Maasai in the group ranches was 63% (Kerven 1987), 60% among Tuareg in Mali (Wagenaar-Brouwer 1984; Kerven 1987), 40% among WoDaabe (Kerven 1987) and 49% among the Bella (Kerven 1987).

Meanwhile, the mean contribution of goats to agricultural income in ranch and pastoral households was found to be 5.5±2.9% and 18±8.3% respectively. The mean percentage contribution of goats to overall agricultural income to ranch and pastoral households was significantly different (p<0.05). Like wise, the contribution of goats towards household income was negatively correlated to the number of cattle owned (r=-0.7, p<0.05).  Meanwhile the percentage mean contribution of crops to total agricultural income was 2.4±1.3% and 3.6±1.8% to ranch and pastoral households respectively.

The major ranch and pastoral farm agricultural inputs were invested in cattle enterprises, constituting to about 97.7±1.5% and 92.2%±3.5% respectively of total agricultural inputs. The remainder of the inputs were put in the crop enterprises, constituting to about 2.1±1.3% and 7.8±3.5% of total agricultural inputs in ranch and pastoral households respectively. Hardly was there any input in goat enterprises, only 0.2% of the total agricultural inputs in the ranches. The major inputs in cattle enterprises were labour and disease control costs. Their overall contribution to total variable costs was not significantly different. McDermott et al (1999) highlighted disease control and labour costs as being the major costs in smallholder cattle production systems. Ng’ethe (1993) observed that in the Masai group ranches in Kajiado district, in Kenya, disease control costs were the major costs.

The mean overall gross margins from agricultural enterprises for pastoral and ranch households were Ug. Shs 6,240,000±2,511,000 and Ug. Shs 12,371,000±1,681,000 respectively. The returns per ha per annum was significantly higher (p<0.05) in the pastoral systems (Ug. Shs 108,800±38,000) than the ranching systems (Ug. Shs 16,000±4,400). The returns per ha of land in pastoral systems was 6.8 times higher than the ranching systems. Earlier, reports in Uganda by Behnke (1985) and Ruthenberg (1980) stated that returns per hectare of land for pastoral systems were twice higher than for ranching systems. Jalvis (1993) has reviewed many reasons why pastoral systems have better returns than ranching systems in sub-Saharan Africa.  However, in this area, high returns from pastoral areas was due to high stocking rates synonymous with what happens with the tragedy of the commons. The observed high correlation of returns per hectare of land with increasing herd sizes implied that it was difficult to convince pastoralists with big herd sizes to reduce their cattle numbers because of high economic returns they derive from them. Similar observations have been reported in Kenya (Nge’the 1993; Bekure et al 1992), Zimbabwe (Scoones 1992), Botswana (Abel 1993) and Mozambique (Rocha et al 1991). In Ethiopia, some pastoral systems have even been reported to have higher returns per hectare than the industrialised ranching systems in Australia (Upton 1989).

Conclusions 

• Cattle enterprises were the major source of income for both ranch and pastoral households derived from milk and live cattle sales. Like elsewhere in East Africa, pastoral households derived substantial income from goat enterprises. As usual, crop output was insignificant source of income in both ranch and pastoral households.

  
  

• The major inputs in cattle enterprises were in disease control and labour costs.

  
  

• Returns per hectare was higher in pastoral systems than ranching production systems in this area. The return per hectare for pastoral households were exceptionally higher in this area than the same areas with similar ecosystem. This was due to over stocking. The pastoral households were motivated to overstock due to higher returns. There was a need to optimise stocking rates by de-stocking pastoral herds and increasing stocking of ranches to sustainable levels.

Acknowledgements 

The authors are grateful to International Foundation of Science (IFS), German Academic service (DAAD) and Government of Uganda through Board of Postgraduate Studies and Research, Makerere University, Kampala,Uganda for providing with funds for this research. The staff of management of Nshara Dairy Cross Breeding Ranch and ranchers of Ankole Ranching Scheme and members of pastoral households in which the study was conducted, are all thanked for their co-operation.

References 

Abel N O J 1993 Reducing animal numbers on southern African communal range: is it worth it?. In: Range ecology at disequilibrium: New models of natural variability and pastoral adaptation in African savannahs (R H Behnke, I Scoones and C Kerven editors. Overseas Development Institute London.

Anon 1988 Report of Commission of Inquiry into Government Ranching Schemes. Government Printer Entebbe

Baldwin K L, DeVeau V, Foster K and Marshall M 2008 Traits affecting household livestock marketing decisions in rural Kenya. Selected paper presented at the American Agricultural Economics Association Annual Meeting, Orlando, Florida, July 27-28 2008.

Behnke R H 1985 Measuring the benefits of subsistence versus commercial livestock production in Africa. Agricultural systems 16: 109-135

Bekure R H, de Leeuw P N, Grandin B E and Neate P J H 1992 Masai herding : An analysis of the livestock system of Masia pastoralists in Eastern Kajiado District, Kenya. International Livestock Centre for Africa (ILCA) Addis Ababa.

Curtin C and Western D 2008 Grasslands, people and conservation: Over the horizon learning exchanges between African and American pastoralists. Conservation Biology 22(4): 870-877

Debrah S and Sissoko K 1990 Sources and transfers of cash income in rural economy: The case of smallholder mixed farmers in the semi-arid Mali. African Policy Analysis Network paper No. 25. International Livestock Centre for Africa, Addis Ababa, Ethiopia.

Dohoo I, Martin W and Stryhn H 2003 Veterinary Epidemiologic Research. Charlottetown, Prince Edward Island, Canada: AVC Inc.

Gryseels G 1988 The role of livestock in the generation of smallholder farm income in two vertisol areas of the central Ethiopian highland. In: Management of vertisols in sub-Saharan Africa. International Livestock Centre, Addis Ababa.

Hardin G 1968 The tragedy of the commons. Science 162:1243-1248.

Hardin G 1988 Commons failing. New Scientist 22 October.

ILRI 1995 Livestock policy analysis. ILRI training manual 2, ILRI, Nairobi, Kenya, 264pp. http://www.ilri.org/html/trainingMat/policy_X5547e/opening_X5547E00.HTM

ILRI 2006 Pastoralism surest way out of poverty in East Africa. ILRI, Nairobi, Kenya 2pp. http://www.ilri.org/ILRIPubAware/Uploaded%20Files/2006711123340.NR_EV_060629_002_Pastoralism%20counters%20Poverty.pdf

Jalvis L S 1993 Sustainable animal agriculture: The role of economics in recent experience and future challenge. In: Strategies for sustainable animal agriculture in developing countries. Proceedings of the FAO expert consultation held in Rome, Italy, 10^th-14^th December90 (S Mack ed.) FAO Animal Production and Health Paper 107: 23-36 http://www.fao.org/DOCREP/004/T0582E/T0582E06.htm

Kerven C 1987 Some research and development implications for pastoral dairy production in Africa. ILCA Bulletin, 26: 29-35.

Kisamba-Mugerwa W, Pender J and Kato E 2006 Impacts of individualisation of tenure on livestock and rangeland management in South-western Uganda. A paper presented at the 11^th Biennial Conference of International Association for Study of Common Property Held in June 19-23, Bali, Indonesia.

 Lekasi J K, Tanner J C, Kimani S K and  Harris  P J C 1998 Manure Management in the Kenya Highlands: Practices and Potential. HDRA, Ryton Organic Gardens, UK.

Lekasi J K,  Tanner J C, Kimani S K  and Harris P J C 2001 Managing Manure to Sustain Smallholder Livelihoods  in the East African Highlands . DFID - Natural Resources Systems Programme HDRA, Coventry, United Kingdom

Lenaola I, Jenner H H and Wichert T 1995 Land tenure in pastoral lands. In: Land we trust: Environment, private property and constitutional change (C. Juma and J.B., Ojwang eds.). Initiatives Publishers, Nairobi, London 231-257.

Lokii J B 2000 Constraints to optimal goat production and management in Matheniko County, Moroto District, Uganda. A special project submitted for partial fulfilment of the requirements leading to Award of Bachelor of Veterinary Medicine of Makerere University Kampala. Faculty of Veterinary Medicine, Kampala, 79pp.

McDermott J J, Randolph T F and Staal J 1999 The economics of optimal health and productivity in smallholder livestock production systems Epizootics 18(2): 399-424

McLeod R 1998 Impact of ticks and associated diseases on cattle in Asia and Australia. Australian Centre for International Agricultural Research, International Livestock Research Institute (ILRI), Nairobi.

Monday G 1991 Interactions between domestic animals and wild ungulate species in and around Lake Mburo National Park. Thesis for Master of Science, Makerere University, Kampala.

Monday G 1993  Large mammals in Lake Mburo National Park.  Their population status and possible seasonal adjustments.  Uganda National Parks, Lake Mburo National Park Planning Project Report.

Morton J 2006 Pastoralist coping strategies and emergency livestock market intervention. In: Pastoral livestock marketing in Eastern African research and policy changes (J McPeak and P Little editors). London: Practical Action.

Musimba N K R and Nyariki D M 2003 Development of and policy on the range and pastoral industry with special reference to Kenya. Anthropologist 5(2): 261-267

Mworia J K and Kinyamario J I 2008 Traditional strategies used by pastoralists to cope with la Nina induced drought in Kajiado, Kenya. African Journal of Environmental Science and Technology 2(1): 010-014 http://academicjournals.org/AJEST/PDF/pdf%202008/Jan/Mworia%20and%20Kinyamario.pdf

Ndlovu L R 1993 Research on small ruminant production systems in Zimbabwe. In: Future of livestock industries in East and Southern Africa (J A Kategile and S Mubi editors). Proceedings of a Workshop Held at Kadoma Ranch Hotel, Zimbabwe, 20-23 July 1992. ILCA, Addis Ababa, 97-103 http://www.fao.org/wairdocs/ILRI/x5485E/x5485e0g.htm

Nge’the J C 1993 Group ranch concept and practice in Kenya with a special emphasis on Kajiado district. In: Future of livestock industries in East and Southern Africa (J A Kategile and S Mubi editors). Proceedings of a Workshop Held at Kadoma Ranch Hotel, Zimbabwe, 20-23 July 1992. ILCA, Addis Ababa 187-200 http://www.fao.org/Wairdocs/ILRI/x5485E/x5485e0t.htm

Niamir M 1982 Report of animal husbandry among the Ngok Dinka of the Sudan, Harvard Institute of International Development and United States Agency for International Development, Cambridge, USA.

Njeru E H N 1996 The application of Indigenous knowledge in Pastoral production systems.  The African pastoral Forum, working paper series No.6. Pastoral information network programme, Nairobi, Kenya.

Nsubuga H S K 1977 Livestock development in Uganda with particular reference to Ankole-Masaka cattle ranching project. Ph.D thesis, Mkaerere University.

Ntozi J P M 1995 High fertility in rural Uganda: The role of socio-economic and biological factors. Fountain Publishers, Kampala 217pp.

Ocaido M 1995 Effects of diseases on developing mixed game and livestock ranching around Lake Mburo National Park in Uganda. MSc thesis, Makerere University, Kampala, 179pp.

Ocaido M, Siefert L and Baranga J 1996 Disease surveillance in mixed livestock and game areas around Lake Mburo National Park in Uganda. South African Journal of Wildlife Research 26(4): 133-135

Ocaido M, Otim C P, Okuna N M, Erume J, Ssekitto C, Wafula R Z O, Kakaire D, Walubengo J and  Monrad  J 2005 Socio-economic and livestock disease survey of agro-pastoral communities in Serere County, Soroti District, Uganda. Livestock Research for Rural Development 17(8) Art. #93.  Accessed August 2008 http://www.lrrd.org/lrrd17/8/ocai17093.htm

Omitti J 1995 Economic analysis of crop-livestock interaction: the case of the Ethiopian highlands. Ph.D thesis. Department of Agricultural and Resource Economics, University of New England: Armidale, Australia.

Opolot J 1998 Assessment of the contribution of livestock to household income in Soroti County, Uganda. A special research project report for partial fulfilment of the requirements fore Award of Degree of Bachelor of Veterinary Medicine of Makerere University. Faculty of Veterinary Medicine, Kampala, Uganda, 51pp.

Perrier G 1995 New directions in range management planning in Africa. In: Living with uncertainty : New directions in pastoral development in Africa (I. Scoones ed.). Intermediate Technical Publications, International Institute of Environment and development, London, 47-57.

Putt S N H, Shaw A P M, Woods A J, Tyler L and James A D 1987 Veterinary epidemiology and economics in Africa. A manual for use in design and appraisal of livestock health policy. ILCA manual No.3, ILCA Addis Ababa. http://www.fao.org/Wairdocs/ILRI/x5436E/x5436E00.htm

Randall J 1944 Report on livestock production in Uganda.  Government Printer, Entebbe.

Rocha A, Starkey P and Dionisio A 1991 Cattle production and utilisation in smallholder farming systems in southern Mozambique. Agricultural Systems 37:55-75

Rollinson D H L 1962 Climate and cattle breeding in Uganda. Bulletin of Epizootics of Diseases of Africa 10: 523-535

Ruthenburg H 1980 Farming systems in the tropics. Clarendon Press, Oxford.

Sansoucy R, Jabbar M A, Ehui S and Fitzhurgh H 1995 The Keynote paper: The contribution of livestock to food security and sustainable development. In: Livestock development strategies for low income countries, Proceedings of the joint FAO/ILRI Roundtable on Livestock Development Strategies for Low Income Countries (R T Wilson, S Ehui and S Mark editors),  ILRI, Addis Ababa, Ethiopia, 27 February - 02 March 1995. Food and Agriculture Organisation, International Livestock Research Institute, 9-21 http://www.fao.org/wairdocs/ILRI/x5462E/x5462e04.htm

Schwartz H J, Walsh M and Hary I 1996 Land use potential in the Ankole Ranching Scheme, Nyabushozi, Uganda. A report for GTZ GMBH. Berlin and college station  45pp.

Scoones I 1992 The economic value of livestock in communal areas in southern Zimbabwe. Agricultural systems 39: 339-359

Scoones I 1995 New directions in pastoral development in Africa. In: Living with uncertainty: New directions in pastoral development in Africa (I. Scoones ed.). Intermediate Technical Publications, International Institute of Environment and development, London, 1-37.

Swift J 1984 Pastoral development in central Niger: Report of the Niger Range and Livestock Project. Ministère du Development Rural Republique du Niger and United States Agency for International Development, Niamey, Niger.

Sylla D 1995  Pastoral organisation for uncertainties. In: Living with uncertainty: New directions in pastoral development in Africa (I. Scoones editor). Intermediate Technical Publications, International Institute of Environment and development, London, 134-152.

Talvakolian B 1984 Women and socio-economic change among Sheikhanzai nomads of western Afghanistan. The Middle East Journal 38(2): 433-453.

Thrusfield M 2007 Veterinary Epidemiology, 3rd Edition. London: Blackwell Science.

Toulmin C 1995 Tracking drought: options for destocking and restocking. In: Living with uncertainty: New directions in pastoral development in Africa (I Scoones editor). Intermediate Technical Publications, International Institute of Environment and development, London, 95-115.

Upton M 1989 Livestock productivity assessment and herd growth models. Agricultural Systems 29: 149-164

Wagenaar-Brouwer M 1984 The contribution of livestock products in consumption and nutritional status: The case of pastoral Touareg and agro-pastoral Fulani in inner delta, Mali. Report submitted to International Livestock Centre for Africa, Addis Ababa, Ethiopia.

White J M and Meadows S J 1981 Evaluation of the contribution of group and individual ranches in Kajiado district to economic development and pastoral production strategies. Ministry of Livestock Development, Nairobi, Kenya.

Received 13 June 2008; Accepted 21 October 2008; Published 1 May 2009

Go to top