SEPTEMBER 2007
 Production Effi ciency and Soil Nutrients 
under Different Farming Systems
 Key points
 ? Current policies and some develop-
 ment projects have promoted dairy 
intensifi cation systems, such as 
zero-grazing, in a variety of settings. 
However, the results show that less 
intensive production systems can be 
equally appropriate in most areas.
 ? Smallholder dairy production is 
shown to be profi table and compet-
 itive in a variety of settings, where 
level of intensifi cation suits different 
local circumstances.
 ? There is need for targeted inter-
 ventions of dairy intensifi cation to 
meet area-specifi c conditions and 
the farmers? specifi c circumstances, 
particularly those that place addi-
 tional demands on the farmers? la-
 bour. 
? Nutrient balances of nitrogen, phos-
 phorus and potassium are positive 
(though small) at the farm level, but 
strongly negative at the land level. 
? Organic fertilizer (manure) from the 
livestock enterprise is rarely used on 
crops. 
? Farmers could improve soil produc-
 tivity by using available manure, al-
 though labour constraints may be an 
impediment.
 ? Development efforts should consider 
the technical and economic feasibil-
 ity, environmental sustainability and 
social benefi ts of the project.
 Research Theme on Improving Market Opportunities
 Researchbrief
 Markets Theme Research Brief 1
 Dairy Farming in Uganda
DA I RY  FA R M I N G  I N  U G A N DA
 2
 Notable policy and institutional changes in 
the Ugandan dairy sector
 1980s:  Promotion of zero-grazing by 
NGOs such as Heifer Project 
International.
 1986: Introduction of the National 
Rehabilitation and Develop-
 ment Program led, inter alia, 
to the introduction of the zero-
 grazing system.
 1992:  The Government launched a 
Milk Master Plan to improve 
dairy production and process-
 ing.
 1993:  Liberalization of the dairy sub-
 sector and emergence of pri-
 vate milk processors.
 1998:  The Dairy Industry Act estab-
 lished the Dairy Development 
Authority to realize objectives 
of Milk Master Plan.
 Introduction
 A series of policy changes, coupled with the pro-
 motion of intensive production systems, have re-
 sulted in signifi cant changes in the dairy sector 
in Uganda over the last two decades. Prior to the 
1980s smallholders made little contribution to 
the nation?s marketed milk production; now both 
overall production and the proportion of milk pro-
 duced and marketed by smallholders have signifi -
 cantly increased. 
Study rationale and objectives
 To answer these questions, an in-depth study was 
carried out between 2001 and 2005 by a team 
of researchers from the Ugandan National Ag-
 ricultural Research Institute (NARO), Makerere 
University, the International Livestock Research 
Institute (ILRI) and the Danish Institute of Agri-
 cultural Sciences (DIAS) with funds provided by 
the Danish International Development Agency 
(DANIDA). 
Improved understanding of dairy production 
systems in Uganda was expected to enable bet-
 ter targeting of evidence-based extension mes-
 sages and development of policies that would 
improve the contribution of dairying to sustain-
 able livelihoods of resource-poor farmers. 
The study had three components:
 ? Characterization of dairy farming systems in 
Uganda
 ? An economic component that examined the 
factors and incentives infl uencing farmers? 
adoption of dairy intensifi cation and crop?
 livestock interactions, such as feeding crop 
residues and using manure to improve soil 
fertility; determined the degree of risk aver-
 sion; and assessed the enterprise choices 
consistent with farmers? resources and ob-
 jectives.  
? A nutrient cycling component which evalu-
 ated key nutrients essential for successful 
crop production (nitrogen, phosphorus and 
potassium).
 Despite these changes, there were major knowl-
 edge gaps concerning the dairy sector in Ugan-
 da. Important unanswered questions included:
 ? What factors and incentives infl uenced 
adoption of dairy intensifi cation measures 
by smallholder farmers?
 ? Is intensifi cation always the best option?
 ? What are the trends in nutrient balances of 
soils on smallholder dairy farms?
3
 DA I RY  FA R M I N G  I N  U G A N DA
 Data sources and methods
 Dairy systems at various levels of intensifi ca-
 tion were studied in Jinja, Masaka and Mbara-
 ra districts of Uganda (study sites shown in 
Figure 1). These ranged from intensive systems 
in which exotic dairy cattle were maintained 
in zero-grazing units through to extensive 
systems in which predominantly local breeds 
were grazed on natural pastures. 
Two surveys were conducted: a cross-sec-
 tional survey (one visit per household) in 2001 
covering 303 farmers and an in-depth longi-
 tudinal survey comprising repeat visits to 24 
farmers over a one-year period between Au-
 gust 2003 and August 2004.
 Data analysis focused on two main aspects: 
economic effi ciency and nutrient manage-
 ment. 
Economic effi ciency and factors driving inten-
 sifi cation were assessed by ?net farm benefi t 
and regression analyses?. Net farm benefi ts 
were evaluated as revenues from all farm ac-
 tivities, including consumption of farm outputs 
(crops and milk) by the household and relatives, 
minus all expenses relative to farm activities. 
Dairy benefi ts were evaluated as revenues 
from sale of milk and dairy animals, plus the 
value of milk consumed by the household, mi-
 nus all expenses relative to dairy.
 A linear programming model was developed to 
evaluate the farmers? choices of crop and dairy 
activities. The model provided the combina-
 tion of farm activities that maximizes benefi ts 
and can therefore be used to assess individual 
farmers? effi ciency by looking at the difference 
between current (or observed) farm plans and 
profi t maximization plan. 
Nutrient management was assessed by 
calculating detailed nutrient balances of 
nitrogen (N), phosphorus (P) and potassium 
(K) at whole farm and cultivated land lev-
 els. Farm nutrient balance was calculated as 
the quantity of nutrients entering the farm 
minus the quantity of nutrients leaving the 
farm. Land nutrient balance was calculated 
as the quantity of nutrients applied to the 
land minus the quantity removed from the 
land. Simulation analyses were carried out 
to assess what the nutrient balances would 
be if the farmers were to apply the manure 
available on their farms to the soil. 
Figure 1: Study sites
 4
 DA I RY  FA R M I N G  I N  U G A N DA
 Key fi ndings
 Farm characterization 
The study identifi ed fi ve different dairy pro-
 duction systems, listed below in descending 
order of intensifi cation:
 ? zero-grazing
 ? semi-intensive
 ? fenced
 ? tethered
 ? herded
 Zero-grazing is the most intensive of the fi ve 
production systems. In this system improved 
dairy cattle are permanently kept in stalls 
with no free-grazing and there is a high level 
of external inputs, including bought-in feeds, 
and signifi cant expenditure on livestock and 
veterinary services. 
Semi-intensive is similar to zero-grazing 
except cattle spend some time confi ned to 
stalls and some time grazing in paddocks. 
In fenced systems, cattle ? a large proportion 
of which are improved dairy breeds ? graze 
in paddocks. 
In the tethered system, predominantly local 
breeds of cattle are grazed mostly off-farm, 
freeing up land for cultivation. Density of 
cattle tends to be high. 
Herding is the least intensive of the fi ve pro-
 duction systems and consists of mostly lo-
 cal breeds of cattle being grazed on and off-
 farm.
5
 DA I RY  FA R M I N G  I N  U G A N DA
 Economics of milk production
 The cost of milk production was compared for 
each of the fi ve dairy production systems. Cost of 
production was highest in the tethered and lowest 
in the herded system. Estimates of fi xed costs (like 
stalls and dairy equipment) were not included and 
therefore a true profi t margin could not be calcu-
 lated. However, the contribution margin ? which is 
the difference between sales price and the variable 
costs of dairy production ? was highest for zero-
 grazing and lowest for the tethered system. Break-
 downs of the cost components under the different 
systems are shown in Figure 2.
 While dairy farming is profi table under all fi ve 
farming systems, profi tability in terms of contribu-
 tion margin was highest under the two most in-
 tensive systems, zero-grazing and semi-intensive. 
However, intensifi cation was not necessarily the 
best option in all situations; profi tability was high-
 er under the herded system, considered to be the 
least intensive, than under the fenced or tethered 
systems. 
Figure 2: Cost (Ush per litre) of milk production and its components by dairy production system.
 Extension messages and policies regarding 
intensifi cation in the smallholder dairy sec-
 tor need to be carefully targeted to ensure 
they are relevant and appropriate. 
The study also highlights the value of seeing 
benefi ts for oneself, such as by working on 
neighbour?s farms, and of farmer-to-farmer 
training approaches.
 Economic effi ciency of dairy production
 Economic effi ciency was evaluated from two points 
of view: opportunity cost of labour and the profi t 
maximization plan.
 Analysis of milk productivity of the different farm-
 ing systems revealed that increasing the level of in-
 tensifi cation resulted in an increase in the milk pro-
 ductivity and the percentage of milk sold (Figure 
3). However, despite the variability in milk produc-
 tivity and the market orientation between farming 
systems, the net farm and dairy benefi ts per hour 
remained above the cost of labour (Figure 4) and 
there was no apparent relationship between degree 
of intensifi cation and the dairy benefi ts per hour. 
This indicates that in the current situation, all sys-
 tems are remunerating the farmers? labour above 
what they could get if they were to offer their serv-
 ices outside their farms.
Figure 4: Comparison of the net farm and dairy benefi ts per hour and the local wage rate for different cattle 
farming systems.
 900
 Us
 h
 Herded Fenced Semi-intensive Zero-grazed
 Net farm benefits/hour Dairy benefits/hour Wage rate
 Tethered
 800
 700
 600
 500
 400
 300
 200
 100
 0
 6
 Figure 3: Relationship between milk productivity, market orientation and cattle farming system.
 herding tethering semi 
intensive
 fenced zero grazing
 Litres of milk per
  TLU per
  y
 ea
 r
 Pct of milk sol
 d
 0
 500
 1000
 1500
 2000
 2500 80
 70
 milk per TLU
 % milk sold
 60
 50
 40
 30
 20
 10
 0
 DA I RY  FA R M I N G  I N  U G A N DA
 To measure economic effi ciency in relation to the 
profi t maximization plan, the net benefi ts achieved 
using actual farm production practices were com-
 pared with net benefi ts achievable if farmers used 
profi t-driven optimal production practices as pre-
 dicted by the linear programming model. Results 
showed that farmers are operating near the profi t 
maximization levels for dairy but could improve 
their overall farm benefi ts by discontinuing fallow 
and switching to more profi table crops (Figure 5).
Drivers of intensifi cation
 Four drivers of intensifi cation were identifi ed. 
These included two human capital drivers, 
years of formal education of the household 
head and proportion of adults working off-
 farm, and two demand drivers, human popu-
 lation density and access to local markets. 
Soil nutrient levels and balances
 A worrying situation is revealed with regards 
to nutrient management on mixed dairy-crop 
farms in Uganda. Irrespective of degree of in-
 tensifi cation, levels of some key plant nutrients 
in the soil, especially nitrogen, are already be-
 low the level considered critical for successful 
crop production (Figure 6). Nutrients removed 
in crops and milk are not being replaced at a 
commensurate rate. This has serious implica-
 tions for the long-term sustainability of these 
Figure 6: Comparison of soil nitrogen (N), phosphorus (P) and potassium (K) levels with critical 
minimum levels under different cattle farming systems (land-level based on soil samples).
 Herded
 0
 20
 40
 60
 80
 100
 120
 Tethered Semi-intensive Non cattle
 N (%) x 100
 P (ppm)
 Critical level K
 Critical level P
 Critical level N
 K (meq/100g) x 100
 Zero-grazingFenced
 7
 Figure 5: Comparison of observed and profi t maximization net farm and dairy benefi ts.
 8000 1200
 Ush/hou
 r
 Herded Tethered Fenced Semi-
 intensive
 Zero-
 grazing
 Us
 h
 0
 200
 400
 600
 800
 1000
 0
 1000
 2000
 3000
 4000
 5000
 6000
 7000
 Observed net farm benefits
 Observed dairy benefits/hour
 Profit max net farm benefits
 Observed net farm benefits/hour
 Profit max net farm benefits/hour
 Profit max dairy benefits/hour
 DA I RY  FA R M I N G  I N  U G A N DA
 farming systems. In most systems adequate 
amounts of manure are available on the farm 
to reverse this situation but currently most of 
this potentially valuable manure is not uti-
 lized. 
Shortage of labour may be one reason that 
manure is underused; in cases where manure 
is applied, this tends to be done during the 
school holidays when family labour is most 
plentiful. Also, in some cases the cultivated 
plots may be distant from the zero-grazing 
units, exacerbating transport problems. But in 
feedback provided by the farmers at the end of 
the study, other factors were also mentioned 
such as lack of awareness of the benefi ts of us-
 ing manure, lack of basic equipment for trans-
 porting manure and fear of introducing pests 
and diseases to their plots. 
CONTACTS:
 Sarah Mubiru 
National Agricultural 
Research Organization 
(NARO)
 National Livestock 
Resources Research 
Institute (NALIRRI)
 c/o National Crop 
Resources Research 
Institute (NACRRI)
 P.O. Box 7084
 Kampala
 Uganda
 Tel: 006772418678 
smubiru@naro-ug.org 
sarah2mubiru@yahoo.com
 Isabelle Baltenweck
 International Livestock 
Research Institute (ILRI)
 P.O. Box 30709 
Nairobi 00100
 Kenya
 Tel: (254) 20-4223402
 i.baltenweck@cgiar.org
 www.ugandadairy.dk
 Disclaimer: This publication is an output from a project funded by the Danish International Development 
Agency (DANIDA) of the Kingdom of Demark for the benefi t of developing countries. However, the views 
expressed here are not necessarily those of DANIDA.
 Conclusions and policy 
recommendations
 ? Dairy farming in Uganda is a profi t-
 able activity under all fi ve farming 
systems. 
? Profi tability (as measured by the con-
 tribution margin) was highest under 
the two most intensive systems, zero-
 grazing and semi-intensive.
 ? Intensifi cation was not necessarily 
the best option in all situations; prof-
 itability was higher under the herded 
system, considered to be the least 
intensive, than under the fenced or 
tethered systems. 
? Factors associated with a tendency 
to intensifi cation are: years of formal 
education of the household head and 
proportion of adults working off-farm 
as well as factors associated with in-
 creased demand for milk. 
? Irrespective of degree of intensifi ca-
 tion, levels of key plant nutrients in 
the soil, especially nitrogen, are al-
 ready below the level considered 
More work needs to be done to 
investigate the economics and 
practicalities of manure applica-
 tion and how best to integrate 
this activity with others that com-
 pete for available labour.
 critical for successful crop produc-
 tion. Nutrients removed in crops and 
milk are not being replaced at com-
 mensurate rate.
 ? In all systems except fenced and 
semi-intensive, adequate amounts 
of manure are available on the farm 
to reverse this situation but currently 
most of this potentially valuable ma-
 nure is not utilized.
 ? The fi ve farming systems present 
unique opportunities and constraints. 
Therefore, research and development 
agenda should take cognizance of 
their uniqueness.