-
FINAL EVALUATION OF THE FY 13 FOOD for PROGRESS 
FEED ENHANCEMENT FOR ETHIOPIAN DEVELOPMENT 
PROJECT PHASE II (FEED II) (Final Report)
By: ICOS CONSULTING PLC
M A Y 2 0 1 9
A D D I S A B A B A , E T H I O P I A
Program: Food for Progress
Agreement Number: FCC-663-2013/039-00
Funding Year: Fiscal Year 2013
DISCLAIMER: This publication was produced at the request of the United States Department of Agriculture. It was 
prepared by an independent third-party evaluation firm. The author’s views expressed in this publication do not 
necessarily reflect the views of the United States Department of Agriculture or the United States Government.
Accessibility Note: An accessible version of this document can be made available by contacting 
fas.monitoring.evaluation@usda.gov 
ACRONYMSAND ABBREVIATIONS
AIBP Agro-Industrial By-Products
CP Crude Protein
DA Development Agent
DM Dry Matter
EC Ethiopian Calendar
ETB Ethiopian Birr
FAO Food and Agriculture Organization
FCM Formulated Concentrate Mixture
FEED Feed Enhancement for Ethiopian Development Project
FGDs Focus Group Discussions
FTC Farmers Training Center
GO Government
GM 
GTP
Gross Margin
Growth and Transformation Plan
HH Household
ILRI International Livestock Research Institute
KGs Kilograms
KII Key Informant Interview
LMP Livestock Master Plan
M&E Monitoring and Evaluation
ME Metabolisable Energy
NGO Non-Governmental Organization
PCs Primary Cooperatives
ROI 
SHF
Return on Investment
Smallholder Farmers
SNNPR Southern Nations, Nationalities, and Peoples' Region
SNNPR Southern Nations, Nationalities and Peoples Region
ToT Training of Trainers
USD United States Dollar
USDA United States Department of Agriculture
VDFACA Veterinary Drug and Animal Feed Administration and Control Authority of Ethiopia.
% Percent
& And
x̄ Mean/Average
Σ Sum
Definitions and Exchange Rates:
Woreda is a district level administrative unit in Ethiopia. 
Kebele is the lowest administrative unit in Ethiopia.
Ethiopian Birr (ETB) 28.5196 = US$1 (Weighted Average Exchange Rate on April 5, 2019 as per National Bank of Ethiopia)
TABLE OF CONTENTS
Executive Summary........................................................................................................................ 7
1. BACKGROUND ....................................................................................................................... 1
1.1. Introduction .................................................................................................................... 1
1.2. Evaluation Approach and Methodology ....................................................................... 2
1.2.1. Design and Methodology ................................................................................. 2
1.2.2. The Study Population........................................................................................ 2
1.2.3. Sampling............................................................................................................. 3
1.2.4. Methods and Data Collection Instruments...................................................... 5
1.2.5. Recruitment and Training of the Evaluation Team........................................ 6
1.2.6. Data Collection/Field Work.............................................................................. 6
1.2.7. Data Management and Analysis ....................................................................... 7
1.2.8. Limitations and Challenges of the Evaluation Study...................................... 7
2. RESULTS AND DISCUSSIONS................................................................................................ 8
2.1. Effectiveness ................................................................................................................... 8
2.1.1. Demographic Characteristics of Respondents...................................................... 8
2.1.2. Livestock Production and Productivity ................................................................. 8
2.1.3. Poultry Management and Productivity ............................................................... 19
2.1.3.1. Poultry size and breed kept by households .................................................... 19
2.1.3.2. Poultry management......................................................................................... 21
2.1.4. Sustainable Forage Production Systems ............................................................. 23
2.1.5. On-Farm Feeding Practices and Adoption of Other Improved Practices ......... 32
2.1.6. Manufactured Feed Development........................................................................ 39
2.1.7. Income from Livestock and Livestock Products................................................. 63
2.2. Relevance ...................................................................................................................... 68
2.2.1. Alignment of project initiatives with government and target groups priorities
68
2.2.2. Stakeholders’ participation in project implementation and satisfaction with
the its results ....................................................................................................................... 72
2.3. Sustainability ................................................................................................................ 73
2.4. Cross-Cutting Issues ..................................................................................................... 77
3. LESSONS LEARNED, CONCLUSIONS AND RECOMMENDATIONS.................................. 77
ANNEXES...................................................................................................................................... 81
Annex 1) FEED II Project’s Achievement on Outcome Indicators and Targets ............ 82
Annex 2) References .......................................................................................................... 84
Annex 3) Sample Beneficiary Households’ Sex, Age, Education Status, Family Size and
Membership in CooperativesDisaggregated by Region ...................................................... 86
Annex 4) Percentage of Households and their Ownership of Mature Cows by Size of
Holdings, Region and Breed Type at Baseline and Endline ................................................. 87
Annex 5) Percentage of Households by Size of Bull/Ox Holding and Region at Baseline 
and Endline 88
Annex 6) Percentage of respondents who have access the different feed items in
terms of how far is the nearest distribution point ............................................................... 89
Annex 7) Return on Investment and Gross Margin of Feed Manufacturing Enterprises
90
Annex 8) Return on Investment of Livestock/Poultry/Forage Enterprises................. 91
Annex 9) List of Key Informants Consulted ..................................................................... 93
Annex 10) List of Farmers that Participated in Group Discussions............................... 103
Annex 11) List of beneficiaries on Whom Case Stories were Built................................ 113
Annex 12) List of Project Woredas and Unions Visited for the Final Evaluation ......... 114
Annex 13) List of FEED I Unions Visited/Consulted ....................................................... 115
LIST OF TABLES
Table 1: Planned and Actual Sample Size for Direct Beneficiary Households Survey.................. 3
Table 2: Stakeholders Consulted for the Final Evaluation ............................................................. 6
Table 3: Percentage and Average Livestock and Poultry Ownership by Animal Category at
Baseline and Endline....................................................................................................................... 9
Table 4: Percentage of Households that Own Dairy Cows by Region at Baseline and Endline .. 11
Table 5: Average Maximum and Minimum Quantities of Milk Reported from Lactating Cows by
Breed Type.................................................................................................................................... 11
Table 6: Percentage of Dairy Cow Owners by Types of Feed Provided to Dairy Cows and
Region at Baseline and Endline .................................................................................................... 12
Table 7: Milk Productivity and Type of Feed Fed to Dairy Cows ............................................... 14
Table 8: Milk Productivity and Feeding Practices........................................................................ 15
Table 9: Percentage of Respondents involved in Fattening by Types of Feed they Provided to
Fattened Animals at Baseline and Endline ................................................................................... 16
Table 10: Percentage of Households by Region and Number of Cattle Fattened at Baseline and
Endline .......................................................................................................................................... 17
Table 11: Percentage of Households by Region and Number of Sheep Fattened at Baseline and
Endline .......................................................................................................................................... 19
Table 12: Households Ownership of Local and Improved Chickens during the 12 Months
Preceding the Survey by Types and Region at Baseline and Endline .......................................... 19
Table 13: Total and Average Number of Improved Chickens by Region at Baseline and Endline
....................................................................................................................................................... 20
Table 14: Total and Average Egg Production by Breed Type and Region at Baseline and Endline
....................................................................................................................................................... 21
Table 15: Percentage of Chicken Holder Households by the Type of Housing System and Region 
at Baseline and Endline................................................................................................................. 21
Table 16: Percentage of Chicken Holder Households by Types of Poultry Feeding Systems and
Region at Baseline and Endline .................................................................................................... 22
Table 17: Average Egg Production during the Past 12 Months Preceding the Survey, and Poultry
Feeding Practices by Region......................................................................................................... 23
Table 18: Percentage of Households that Cultivated Improved Forage 12 Months Preceding the
Survey by Region at Baseline and Endline................................................................................... 24
Table 19: Percentage of Households that Cultivated Improved Forage 12 Months Preceding the
Survey by Region at Baseline and Endline................................................................................... 26
Table 20: Use of improved forage and milk productivity............................................................. 31
Table 21: Percentage of Households that Adopted Improved Practices at Baseline and Endline 33
Table 22: Percentage of follower farmers that received transfer of knowledge from model
farmers over the past four years.................................................................................................... 36
Table 23: Return on Investment in Feed Manufacturing Enterprises........................................... 47
Table 24: Percentage of Households by Distances to the Nearest Animal Feed Distribution
Center and Region at Baseline and Endline.................................................................................. 49
Table 25: Impact of feeding/not feeding concentrates on dairy cow productivity ....................... 50
Table 26: Distance from nearest distribution point for factory formulated and milk productivity
....................................................................................................................................................... 51
Table 27: Return on investment (RoI) of livestock and poultry enterprises supported by the
project ........................................................................................................................................... 55
Table 28: Beneficiary households’ access to poultry market ....................................................... 60
Table 29: Distribution of Households by Number of Fattened Oxen/Bulls Sold and the Value of 
those Sales and Region at Baseline and Endline .......................................................................... 63
Table 30: Distribution of Households by Number of Fattened and Sold Cows and the
Corresponding Sales and Region at Baseline and Endline ........................................................... 64
Table 31: Distribution of Households by Quantity of Milk Sold and Corresponding Value of
Sales from Milk at Baseline and Endline...................................................................................... 65
Table 32: Distribution of Households by Quantity of Butter Sold and the Value of Sales and
Region at Baseline and Endline .................................................................................................... 65
Table 33: Distribution of Households by Number of Pullets Sold and the Value of Sales and
Region at Baseline and Endline .................................................................................................... 66
Table 34: Distribution of Households by Number of Broilers Sold and Value of Sales and Region 
at Baseline and Endline................................................................................................................. 67
Table 35: Distribution of Households by Number of Spent Hens Sold and Value of Sales and
Region at Baseline and Endline .................................................................................................... 67
Table 36: Number of Eggs Sold and Value of Sales from Eggs in the Last Year by Region at
Baseline and Endline..................................................................................................................... 68
LIST OF FIGURES
Figure 1: Percentage of households that received training on dairy animal nutrition and 
management .................................................................................................................................. 13
Figure 2: Percent of Households that Fattened Cattle, Sheep and/or Goats during the Past 12
Months at Baseline and Endline ................................................................................................... 15
Figure 3: Percentage of respondents that bought forage planting materials (seeds, seedlings, 
cuttings, splits, etc) or got them for free ....................................................................................... 25
Figure 4: Number of Improved Practices and Percentage of Households that Adopted the
Practices at Endline....................................................................................................................... 33
Figure 5: Percentage of households that did/did not transfer knowledge they gained through the 
project ........................................................................................................................................... 34
Figure 6: Percentage of households by type of knowledge/practices transferred and adoption of 
these practices............................................................................................................................... 35
Figure 7: Percentage of follower farmers and their use of different improved practices ............. 36
Figure 8: Formulated Feed Value Chain....................................................................................... 39
Figure 9: RoI and Gross Margin Growth Trends in Feed Manufacturing Enterprises................. 47
Figure 10: Buyers of milk produced by smallholder farmers....................................................... 59
Figure 11: Level of satisfaction of direct beneficiary farmers with the project support............... 72
Executive Summary
The Feed Enhancement for Ethiopian Development Project – Phase II (FEED II) was designed 
and implemented during the past five years. Funded by USDA, the project was initially 
designed to be implemented for three-years but later extended for additional two years. The 
ultimate goal of FEED II is to increase the incomes of Ethiopian smallholder livestock 
producers by improving access to, and use of consistent, affordable, high quality animal feed 
that can support greater livestock productivity and efficiency. To sustainably increase the 
incomes of Ethiopian smallholder farmers in project target areas, FEED II aims to increase 
the productivity of livestock and poultry by 15% by developing the animal feed sector and 
expand trade of agricultural products in the livestock and poultry sector by 15%. A terminal 
evaluation was conducted to determine the achievements of the project against its targets 
and objectives. The evaluation used mixed methods and surveyed 1,894 direct beneficiary 
households, 73 follower farmers, 12 union-based feed manufacturing enterprises and 
hundreds of other stakeholders.
The findings of the final evaluation showed that a significant proportion of target 
smallholder farmers adopted the different improved livestock and agricultural technologies 
and practices promoted by the project, which ranged from improved forage cultivation to 
feeding concentrates to applying feeding troughs. Development of animal feed sector and 
adoption of the different improved practices promoted by the project contributed to 
increased livestock productivity. Unit productivity of fattened animals measured by number 
of fattened animals per 100 households increased by 32% from baseline, exceeding the 15% 
target set for livestock productivity. Unit productivity of poultry increased by 12% following 
the project intervention. Average ownership of chickens increased to 9.98 chicken (both 
local and improved) at endline compared to 8.9 at baseline, a 12% growth from baseline. 
Milk productivity of beneficiary households also more than doubled. A significant proportion 
of households also moved away from local to improved breeds following the project 
intervention. The project also expanded trade of agricultural products in the livestock and 
poultry sector by 210%, exceeding its 15% target significantly.
1
1. BACKGRO UND
1.1. Introduction
Ethiopia has the largest livestock population in Africa and the fifth largest in the world. The livestock 
subsector contributes nearly 20 percent of total GDP and foreign exchange earnings of the country, 
and some 35 to 40 percent of agricultural GDP (World Bank, 2017). Despite high livestock population 
and the contribution of livestock to the economy and small holders’ livelihood, the production system 
is subsistence oriented, is not adequately market-oriented and livestock productivity remains very 
low (Negassa et al., 2011). The key challenges that continue to undermine the performance and 
potential of the livestock sub-sectors include reduced availability and access to communal grazing 
and natural pasture; insufficient access to forage, forage seeds and feed supply; poor animal health 
due to disease prevalence; low livestock genetic make‐up; limited adoption of improved livestock 
practices and poor provision of livestock support services. In addition, the sector has a low 
commercial market off‐take due to inadequate processing and marketing infrastructure.
It was against this background that the Feed Enhancement for Ethiopian Development Project –
Phase II (FEED II) was designed and implemented during the past five years. The Ethiopian 
government in its Growth and Transformation Plan (GTP) has also identified the transformation of 
the livestock sector as a source of economic growth in the second GTP and beyond through 
intensification and commercialization of production. Funded by United States Department of 
Agriculture (USDA) Food for Progress Program, FEED II has been implemented by ACDI/VOCA since 
September 2013 in Amhara, Oromia, SNNP and Tigray regional states of Ethiopia. The project was 
initially designed to be implemented for three-years but later extended for additional two years.
The ultimate goal of FEED II is to increase the incomes of Ethiopian smallholder livestock producers 
by improving access to, and use of consistent, affordable, high quality animal feed that can support 
greater livestock productivity and efficiency. To sustainably increase the incomes of Ethiopian 
smallholder farmers in project target areas, FEED II aims to increase the productivity of livestock and 
poultry by 15% by developing the animal feed sector and expand trade of agricultural products in 
the livestock and poultry sector by 15%. These objectives are to be achieved through four 
Intermediate Results (IRs), which address availability, access, knowledge about and utilization of 
feed resources. FEED II project implementation covers four regions – Amhara, Tigray, Oromia and 
SNNPR – where over 80% of the country’s livestock population resides. Within these regions, the 
project has worked in 39 woredas and over 186 kebeles. The project has been working with a total 
of 24 cooperative unions (12 from FEED I and 12 FEED II), which have been the key entry and 
scalability points through which the project is implemented.
As FEED II is now ending, ACDI/VOCA commissioned ICOS Consulting PLC to conduct a final 
evaluation of the project. The purpose of the final evaluation is to assess whether the project has 
achieved the expected results outlined in the results framework. The final evaluation assessed areas 
of project design, implementation, management, lessons learned and replicability. It also sought to 
provide lessons learned and recommendations for USDA, program participants and other key
2
stakeholders for future food assistance and capacity building programs. In general, the evaluation 
assessed: a) relevance, b) effectiveness, c) efficiency, d) impact, e) sustainability, and f) other areas 
of interest to USDA and stakeholders. In addition, the evaluation has identified what the project 
should continue, cease and change to improve outcomes to ensure sustainability in FEED III. More 
specifically, the final evaluation intends to:
i. Collect household socio-economic data pertaining to livestock farmers in the project zone of
influence;
ii. Determine the endline values for contextual outcomes and impact level performance
indicators as outlined in the project performance monitoring plan;
iii. Identify gaps and weaknesses in feed manufacturing supply chains in each of the four project
areas including feed regulatory framework;
iv. Identify and document innovative feed and forage production systems that promote
smallholder and market resilience;
v. Identify feed and forage production capacity gaps, sector policy constraints and
organizational weaknesses related to feed manufacturing and feed distribution supply
chains;
vi. Identify activities that can act as catalysts in developing sustainable forage production
systems that can facilitate on-farm feeding practices;
vii. Identify opportunities for collaborative learning and action along the feed value chain that
can foster livestock production systems change;
viii. Identify constraints in maximizing beneficial use of available feed resources among livestock
and poultry smallholder farmers;
ix. Assess the impact of the project on upstream and downstream economic activities of
buyers/suppliers of sub-grantees; and
x. Identify recommendations for FEED III based on findings.
This report presents the findings of the FEED II project final evaluation that was conducted from
December 2018 through February 2019.
1.2. Evaluation Approach and Methodology
1.2.1. Design and Methodology
The final evaluation utilized the FEED II baseline study design and methodology so as to ensure 
consistency and comparability. Hence, like the baseline survey, the final evaluation used a non￾experimental independent sample pre/post-test evaluation design. The study was also a cross￾sectional survey and designed to provide region, livestock, household and cooperative level 
information related to project indicators. The final evaluation also employed mixed (quantitative and 
qualitative) approaches.
1.2.2. The Study Population
The major study populations for the final evaluation were direct beneficiary smallholder livestock 
producer farmers and feed manufacturing enterprises established under cooperative unions. In 
addition to smallholder farmers, follower farmers; primary cooperatives and private enterprises
3
engaged in dairy, poultry, fattening and forage production; commercial feed ingredient suppliers; 
feed equipment manufacturers; development agents, relevant experts of the federal, regional, zonal 
and woreda offices of agriculture and livestock development, and cooperative promotion offices and 
other implementing partners were also targeted by the evaluation.
1.2.3. Sampling
i. Sample Size
The sample size for the household survey was determined in such a way that it enables regional level 
representation. FEED II hypothesizes that income from livestock depends, in part, on use and non￾use of improved new technologies introduced by the project. The project targeted to reach 40,481 
direct and over 242,886 indirect beneficiaries (source: modified project document). It also aimed to 
ensure that 70% of targeted direct beneficiaries adopt improved agricultural technologies such as 
improved forage-growing systems, increased use of manufactured feed, improved feedlot, dairy and 
poultry production practices. Like the baseline survey, t sample size was determined considering a
5% margin of error and 95% confidence level, 1.3 design effect and 10 percent non-response rate
and using Cochran’s general formula (J.E. Bartlett et al., 2001).
n = ([Z2X PQ/e2] X design effect)*NR
Where n is the estimated sample size
Z= 1.96 from a standard distribution table at 95% confidence level
P= proportion of beneficiaries expected to adopt the new agricultural technologies (0.7) 
Q= 1-P (those who will not adopt the new agricultural technologies) (0.3)
e = the level of error term which we are going to accept/tolerable (0.05 or 5%) 
NR= Non-response rate
Feeding these values to the above formula resulted in a sample size of 462 smallholder farmers per 
region. In total, the final evaluation team planned to interview 1848 smallholder farmers in the four 
regions. The field team actually surveyed 1,894 smallholder farmers, exceeding the target sample 
size. The planned and actual sample size for the household survey is presented in Table 1 below.
Table 1: Planned and Actual Sample Size for Direct Beneficiary Households Survey
Region Target Sample Actual Sample
Amhara 462 480
Oromia 462 464
SNNP 462 483
Tigray 462 467
Total 1,848 1894
4
ii. Sampling Procedure
A multi-stage sampling was utilized to reach to sample beneficiary smallholder farmers for the 
survey. The sampling process started with selection of woredas. The woredas selected for the final 
evaluation were the ones that were covered by the baseline study, which considered geographical 
disparity as well as the types of cooperatives in each woreda in selection of the woredas for the 
survey. A total of 32 project woredas (eight woredas per region) were covered by the final evaluation. 
All but one baseline woreda were covered by the final evaluation. This project woreda, in Oromia 
region, had to be replaced by another woreda in the same region owing to security concerns. In terms 
of project woreda representation, out of the 39 project woredas, the final evaluation covered 32 
(82%) of them.
In the second stage of sampling, kebeles were selected from each sample woreda to administer the 
household and local official surveys. Hence, a total of 64 kebeles from the 32 woredas were covered 
by the final evaluation.
Some of the kebeles that were covered by the baseline study were also included in the final 
evaluation. The selection of kebeles for the baseline survey gave major emphasis to high livestock 
potential kebeles. Kebeles that were not covered by the baseline were also surveyed for the final 
evaluation due to additional considerations. One of the considerations was the need to ensure 
representation of smallholder farmers by their year of entry into the project (from 2014 to 2017). 
The evaluation team had to include new kebeles for the survey so as to find adequate number of 
smallholder farmers that were reached by the project in 2017.
The third stage in the sampling process was the selection of smallholder farmers for the survey, 
which was undertaken using systematic sampling technique based on the list of beneficiary farmers 
maintained by the project and/or its local government counterparts. On average, about 29 
beneficiary smallholder farmers participated per kebele in the endline survey.
In addition to households, the evaluation team interviewed cooperative unions supported by the 
project, nursery sites, livestock enterprises, buyers and providers of these sub-grantees and others. 
A total of 12 unions (8 FEED II and 4 FEED I Unions) were interviewed. The selection of the 
cooperative unions in each category (FEED I and II) was undertaken using purposive sampling 
technique. The key criterion for inclusion was representation of FEED I and II unions in the sample. 
Another criterion was the woreda or location in which the union was located. FEED II unions that 
reside in woredas selected for the smallholder farmer survey were selected for consultation. The 
selection of other sub-grantees such as dairy, poultry and fattening farms followed a two-stage 
process. The first stage was stratifying these sub-grantees based on their area of engagement (such 
as poultry, dairy, fattening, etc.) under each of the four regional states. Thereafter, sample sub￾grantees were selected randomly. The overall field work performance in terms of the number of areas 
covered and people consulted is presented in Table 2.
5
1.2.4. Methods and Data Collection Instruments
The final evaluation study employed both quantitative and qualitative data collection techniques. 
Quantitative data was largely collected through a structured questionnaire. Qualitative data was 
gathered through key informant interviews, FGDs, and observations.
i. Questionnaire: Different sets of semi-structured questionnaires were designed and 
administered to gather quantitative data from direct and indirect project beneficiaries. To 
ensure comparability and consistency, direct beneficiary smallholder farmers were surveyed 
by administering the structured questionnaire used during the baseline after making some 
adjustment on exposure variables. The questionnaire was originally developed in English and 
later translated into Amharic and Oromiffa languages for administration. The local language 
versions of the questionnaire were administered using tablets. Other short surveys were also 
administered on to a sample of follower farmers, as well as input suppliers and output buyers 
of sub-grantees of the project. A total of 73 follower farmers were surveyed (nearly one 
follower farmer per sample project kebele) with the aim to capture how they received the 
knowledge transfer, whether they adopted the promoted improved practices, and so on. 
These farmers were identified for interview by model farmers. One follower farmer per 
sample kebele was selected and interviewed by field supervisors. Short semi-structured
questionnaires were also administered on input suppliers and output buyers of sub-grantees 
of the project. The questionnaires contained close and open-ended questions. These 
questionnaires were designed in English and were administered through translation.
ii. Focus Group Discussion (FGD): A total of 32 focus group discussions involving 269 direct 
beneficiary smallholder farmers (193 male and 76 female) were conducted. The focus group 
discussions were facilitated by field coordinators guided by semi-structured questions. On 
average, about 8 people participated in the FGDs and attempts were also made to ensure 
gender representation in the groups. The participants were selected in consultation with 
development agents and local officials. See Table 2 for details.
iii. Key informant interviews: In-depth interviews were conducted with 157 key informants 
from regional and woreda agriculture/livestock office, and development agents; managers of 
feed manufacturing units at Unions and Nursery sites; representatives of Ethiopia Meat and 
Dairy Industry Development Institute, Ethiopian Animal Feed Industry Association, and 
Veterinary Drug and Animal Feed Administration and Control Authority; feed mill equipment 
manufacturers. Different sets of semi-structured interview guides with predominantly open￾ended questions were developed for each category of respondents. Informed oral consents 
were secured from data sources before the interviews were conducted. Some interviews 
involving key informants such as development agents and woreda agriculture offices were 
also recorded with the consent of participants.
iv. Observation: Each field team conducted field visits to selected project activities/sites such 
as Forage Nursery Sites (7 of them were visited), feed manufacturing plants, livestock 
enterprises and smallholder farmers’ backyard forage production activities in order to have 
a visual observation of some of the projects’ activities so as to observe their existence and 
functioning.
6
v. Case stories: The field team developed for 20 (15 male and 5 female) people through 
interview and observation with the aim to document significant success stories and 
impact of the project on direct and indirect beneficiaries. Cases were selected with 
local project counterparts such as development agents. Each case was profiled with 
the help of a checklist and guiding questions.
Table 2: Stakeholders Consulted for the Final Evaluation
Field Performance Amhara Oromia SNNPR Tigray Total
Woredas covered 8 8 8 8 32
Kebeles visited 16 16 16 16 64
Direct beneficiary smallholder farmers 
surveyed
480 464 483 467 1894
FGDs conducted 8 8 8 8 32
FGD participants 62 69 65 73 269
Cooperative unions/Feed manufacturing 
enterprises consulted 4 2 3 3 12
Key informants consulted 49 37 45 26 157
Case stories developed 6 3 2 5 19
Follower farmers interviewed/surveyed 17 15 21 20 73
1.2.5. Recruitment and Training of the Evaluation Team
Field researchers and enumerators were recruited considering their experience in similar 
agriculture research, communication skills and local language proficiency, familiarity with the study 
areas, ability to work as a member of a team, and having relevant technical skills and training in data 
collection. Field supervisors and enumerators were recruited from the respective regional states and 
brought in to Addis Ababa for training. The overall evaluation team including field team members 
were trained for two days in Addis Ababa before they were deployed to the field. The training was 
given in collaboration with ACDI/VOCA FEED II team. In total, three key consultants, eight regional 
coordinators and 31 field supervisors and enumerators, one computer programmer (43 in total) 
participated in the final evaluation study. Each regional team was composed of a regional field 
coordinator, assistant regional field coordinator, field supervisor and enumerators. All the data 
collectors had a minimum of bachelor’s degree with some experience in similar surveys. The field 
team was provided with a data collection manual to read and refer to at times of difficulty during 
data collection in the field.
1.2.6. Data Collection/Field Work
Data collection began simultaneously across the four project regions. Each field team started their
field work by obtaining letter of cooperation from relevant regional government bodies. The field 
team held daily review meetings to review their progress, data gathered using tablets and transfer 
the survey data to online server, revised their next day plan and capture lessons. Each field team also 
held call conference with the team leader to provide progress update and share lessons. Field
7
coordinators were tasked with qualitative data collection through focus group discussions, 
observation and in-depth interviews with different stakeholders.
1.2.7. Data Management and Analysis
Immediately after completion of data collection, a half-day field work review workshop was held to 
review the field proceedings, capture lessons and observations. The preparation of data for analysis 
also started immediately after the field work. Much of the quantitative data was gathered using 
tablets. Hence, data collection and entry happened concurrently. The use of tablets ensured more 
control over the data quality by automatic skip-pattern and validation of the response, minimizing 
data-entry errors. Some of the paper-based short surveys conducted with follower farmers, 
ingredient suppliers and output buyers were entered into SPSS. All survey data was then cleaned 
before analysis by running different tests involving all entries to spot missing values, outliers and 
errors in data collection. Qualitative data was translated and transcribed by field coordinators and 
supervisors that gathered them. The field researchers then typed in English into Microsoft word 
format for analysis.
Different statistics were computed depending on the data type and project indicators that ranged 
from simple descriptive to inferential statistics. Whenever necessary and possible, differences 
between or among survey groups were also tested for their significance at p<0.05. Qualitative data 
analysis and interpretation was conducted by using qualitative content analysis methodology. The 
evaluation team also brought the baseline and endline survey data together to compute some 
statistics on certain project indicators and outcomes. Qualitative data analysis was used to validate 
or supplement quantitative findings and clarify responses from the quantitative survey.
1.2.8. Limitations and Challenges of the Evaluation Study
The survey team faced some challenges that had some impact on data quality. Difficulties faced by 
some beneficiary farmers to remember the support they received from the project might have 
understated some reported figures. The timing of evaluation was not good as there is large 
government official reshuffling owing to current political reforms in the country. The field team tried 
to deal with the challenge by tracking and consulting ex-officials or experts that worked with the 
project or know the project better. But it is likely that some of their views may not represent the 
intention or views of current government officials. The fact that the field work was conducted during 
harvest times in some areas also hampered data collection. There was an error in coding the survey 
questionnaire as related to local cows. As a result of the error, maximum and minimum milk 
production was captured on only 302 households that owned 500 cows. Many sub-grantees including 
feed manufacturing unions were not found to have separate financial records for their operations. 
They tend to include financial records of the project-supported business activities with their other 
operations.
8
2. RESULTS AND DISC USSION S 
In this section of the report, the findings of the final evaluation are presented based on the standard 
evaluation criteria of effectiveness, relevance, sustainability and cross-cutting issues. The 
description of the results and discussion of the results henceforth will portray of the project’s
achievement as the expected results outlined in the results framework.
2.1. Effectiveness
Effectiveness is a measure of the extent to which an intervention’s intended results (outputs or 
outcomes) have been achieved or the extent to which progress toward outputs or outcomes has been 
achieved. In this section, the findings are presented focusing on household socio-economic data 
pertaining to livestock farmers, and contextual outcomes and impact level performance measures by 
comparing final evaluation/endline survey results with that of baseline.
2.1.1. Demographic Characteristics of Respondents
The vast majority of beneficiary households (85%) that participated in the survey were men headed 
while the remaining 15% of households are women-headed. Close to two-third (65%) of the 
respondents are between 40 and 50 years old. Household heads aged below 30 years represented 
only 5% of the respondents. Close to a third of the household heads are over 50 years old. About 28% 
of respondents are illiterate while the majority (59%) of them can read and write or have primary 
education. Respondents with secondary and above level education represented only 13% of the total. 
Seventy-one per cent of the respondents are members of cooperatives and the vast majority of them 
(78%) belong to a multipurpose cooperative. Only 18% of households are members of livestock 
cooperatives. Households in Tigray and Oromia are more likely to be members of cooperatives than 
those in Amhara and SNNP. On average, a household has 6.5 members, which is closer to the baseline 
results of 6.2 but much larger than the national average of 4.6 (3.5 persons in urban areas and 4.9 
persons in rural areas). Average household size did not differ significantly among regions. The 
demographic characteristics of respondents was more or less comparable with those that 
participated in the baseline study (See Annex 3).
2.1.2. Livestock Production and Productivity
2.1.2.1. Livestock Ownership and Composition
As presented in Table 3 below, about 92% of the respondents owned one or more mature cows 
during the 12 months preceding the endline survey. The proportion of target beneficiary households 
that owned crossbreed mature cows increased from 23% at baseline to 38% at endline. In contrast,
9
the percentage of households that owned local cows declined from 79% at baseline to 72% at endline. 
Analysis of survey data also shows that households owned more livestock in the 12 months preceding 
the endline survey than at the time of the baseline survey. On average, a household owned 1.68 and
0.75 local and crossbreed mature cows at endline, respectively, compared to 1.5 and 0.4 at baseline. 
The results show that direct beneficiary households owned a larger number of mature cows at 
endline than baseline. This was largely the case due to increased ownership of crossbreed cows 
during the intervention period. The average number of crossbreed cows owned by sample 
households at endline survey was almost twice as much as that of the baseline, and the difference 
was statistically significant. Compared to baseline, the average number of local cows owned by 
households at endline also increased significantly. However, the proportion of households that 
owned local cows declined by seven percentage points from baseline. The results suggest that direct 
beneficiary farmers are increasingly acquiring crossbreed cows to improve their livestock 
productivity. The training provided by the project on modern livestock management was 
instrumental in helping farmers apply better livestock management practices. ‘I put what I learned 
from the project into practice and purchased two crossbreed cows,’ one model farmer responded 
when asked about the benefits he obtained from the training provided by the project. Many farmers 
and key informants also indicated that farmers are shifting from local to crossbreed cattle. “Most 
farmers are trying to shift from local to crossbred dairy cattle,’ a woreda agriculture office official 
also said. Indeed, analysis of qualitative data suggests that ownership of crossbreed cattle could have 
even been much larger if such animals were accessible and affordable to buy.
‘Before the project intervention we count our cows not their products and efficiencies, and most 
of us had local cows that gives us an average of 2 liters of milk per day, but now those who have 
hybrid milk cows and practically implemented the training received from the project are getting 
better yield and they are obtaining an average of 8 to 20 liters of milk per day’ a direct beneficiary 
farmer that participated in one of the FGDs said.
The proportion of households that owned local and crossbreed bulls/oxen also increased from 74% 
and 10%, respectively, at baseline, 77% and 17% at baseline. A household, on average, owned 1.86 
local and 0.28 crossbreed bull/oxen at endline, compared to 1.65 local and 0.15 crossbreed ox at 
baseline. The differences were statistically significant. In terms of overall cattle ownership, 
households that owned local and crossbreed cattle1 represented 89% and 47% of the endline survey 
respondents, compared to 93% and 27% at baseline. Beneficiary households on average owned 5.5 
local (5 at baseline) and 1.84 crossbreed (0.94 at baseline) cattle at endline. The results show that 
ownership of crossbreed cattle doubled over the intervention period which was also statistically 
significant.
Table 3: Percentage and Average Livestock and Poultry Ownership by Animal Category at Baseline and Endline
Type of
Animal
Type Baseline Endline
1 The cattle population is the sum of mature cows, mature bulls/oxen (> 2 years), heifers, male cattle (6 months to 2 
years), and calves (any sex < 6 months).
2 Mature cow refers to female cattle that have at least one calf.
3 Mature bull/ox refers to male cattle that are greater than 2 years old
10
# and % of HHs 
who own the
animal
Mean Animal
Count
# and % of
HHs who 
own the 
animal
Mean Animal
Count
Mature
Cow2
Indigenous 79% 1.5a 2871 72% 1.68b 3181
Crossbreed 23% 0.36a 694 38% 0.75b 1423
Total 89% 1.86 3565 92% 2.41 4604
Mature
Ox
3
Indigenous 74% 1.65a 3156 77% 1.86b 3526
Crossbreed 10% 0.15a 292 17% 0.28b 539
Total 78% 1.8a 3448 82% 2.12 4065
Cattle
Total
Indigenous 93% 5.11a 9785 89% 5.55b 10506
Crossbreed 27% 0.94 1802 47% 1.84b 3488
Sheep (local 54% 3.36a 6438 52% 4.09b 7741
Goat (local 17% 1.01a 1934 19% 2.15b 4079
Chicken Indigenous (66% 5.75a 11007 41% 3b 5682
Crossbreed 15% 3.08a 5888 62% 7.08b 13417
Total 72% 8.83 16895 79% 9.98 19099
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
Beneficiary households’ ownership of other animals such as sheep, goats and chickens also increased 
during the project intervention period. As also shown in Table 3 above, a household on average 
owned 9.98 chicken (both local and improved) at endline compared to 8.9 at baseline, a 12% growth 
from baseline. The change was particularly significant in regards to ownership of improved chicken. 
The proportion of households that owned improved chicken increased from 15% at baseline to 62% 
at endline. On average, direct beneficiary households that participated in the endline survey owned
7.1 improved chicken compared to 3.1 at baseline. In contrast, ownership of local chicken declined 
from 5.8 chicken per household at baseline to 3 chicken at endline. These findings suggest that direct 
beneficiary farmers have been reducing their local chicken ownership in favor of improved breeds, 
which can be considered a significant development in the right direction to improve poultry 
productivity. Qualitative data gathered from direct beneficiary farmers and government 
counterparts attributed the shift to the project intervention. ‘…some farmers are shifting from 
keeping indigenous chicken to improved hybrids for egg production,’ a woreda agriculture/livestock 
office expert said. The training provided by the project was said to have enabled farmers to gain 
knowledge about which animal breed are more productive and which aren’t.
11
2.1.2.2. Milk Production and Productivity
The vast majority (79%) of households reported having dairy cows at some point in time during the 
year preceding the endline survey, which was also exactly the same as the baseline results. Over the 
past five years, as shown in Table 4 below, the proportion of sample beneficiary households that 
owned local dairy cows remained the same at endline as was the baseline (59%). In contrast, there 
was a ten-percentage point increase in the proportion of households that owned crossbreed dairy 
cows (from 20% at baseline to 30% at endline). As also explained in the previous section of this 
report, these results suggest that beneficiary farmers have been increasingly buying improved cows 
so as to improve their milk production and productivity. Qualitative data analysis findings attributed 
the change to the project intervention.
Table 4: Percentage of Households that Own Dairy Cows by Region at Baseline and Endline
Types of
Dairy Cows
Amhara Oromia SNNPR Tigray Total
Baseline
478
Endline
479
Baseline
480
End
line
464
Baseline
477
Endline
480
Baseline
479
Endline
467
Baseline
1,914
Endline
1890
Local
(indigenous)
only
59 57.2a 55 38.6b 63.9 54.4a 58 43.5b 59 48.5
Crossbred 
only
8.6 9.2a 13.5 29.5b 17.6 22.5b,c 10.9 19.1c 12.6 20.0
Both types 4.8 19.6a 15.4 12.7b 6.5 4.0c 4.4 3.2c 7.8 9.9
Have no
dairy cows
27.6 14.0a 16 19.2a 11.9 19.2a 26.7 34.3b 20.6 21.6
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
About half (49%) of households owned only local/indigenous lactating cows, compared to 59% at 
baseline. In contrast, a fifth (20%) of endline survey respondent households (compared with 13% at 
baseline) owned only crossbred cows. One in ten households owned both indigenous and crossbred 
lactating cows. Average dairy cow ownership per household increased from 1.29 cows at baseline to
1.4 at endline, an increase of 8% (see Table 4 above).
In terms of milk production, as summarized in Table 5 below, the average maximum amount of milk 
produced (highest amount of milk produced per day) from 1,489 indigenous lactating cows in a 
typical day during the last year before the endline survey was 3485 liters (compared with 3,256.35 
liters from 1,701 cows at baseline) while the average minimum (lowest level milk production) was
2017 liters (compared with 1,569.03 at baseline). This provides an average maximum and minimum
milk production of 3.8 and 2.2 liters/day/cow, compared with 1.91 and 0.92 liters per day per cow 
at baseline, respectively. This suggests that milk productivity doubled over the past five years. The 
average maximum daily per cow milk production grew from 2.95 liters at baseline to 6.02 liters, a
100% increase from baseline. The average minimum daily milk production from a single cow also 
increased from 1.4 liters to 3.6 liters, a 154% increase from baseline.
Table 5: Average Maximum and Minimum Quantities of Milk Reported from Lactating Cows by Breed Type
12
Types of
Lactating cows
Number of
Lactating cows
Maximum total 
amount of milk 
collected from all 
lactating cows 
(liters/day)
Minimum total 
amount of milk 
collected from all 
lactating cows 
(liters/day)
Average maximum 
amount of milk 
collected 
(liters/cow/ day)
Average Minimum 
Amount of milk 
collected 
(liters/cow/ day)
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Indigenous/local 1,701 1489 3,256.4 3484.64 1,569.0 2017.4 1.91 3.8 0.92 2.2
Crossbred 371 616 2,744.8 3936.5 1,410.5 2525 7.4 10.4 3.8 6.7
Both types 399 513 1,278 1495 524 807 3.2 8.0 1.31 4.3
Total 2,471a 2637b 7,279.7 8916.1 3,503.5 5349.4 2.95a 6.02b 1.42a 3.6b
% change from 
baseline to 
endline (from 
total)
7% 22% 53% 104% 154%
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
Most farmers who participated in FGDs agreed on the project’s role in improving their livestock 
productivity. Participants stated that due to feeding better quality forages like elephant grass and 
oats/vetch mixed fodder and supplementing with manufactured dairy ration, the milk yield of 
especially crossbreed cows increased from 3 liters/day /cow to about 8 liters/day/cow, and in some 
high-grade cows, up to 15 liters/day/cow. This is also evident from the survey findings presented in 
Table 6 that showed large increase in use of concentrate mix over the past five years. The proportion 
of dairy cow owners that fed concentrate mix to their dairy cows increased from a mere 4% at 
baseline to 39% at endline. Qualitative data analysis results also suggest that the percentage of 
households that used concentrate mix could have been much higher than the reported figures if 
prices were affordable and products are widely available. For many households, the price of 
concentrate mix is high and considered less cost effective when used on local breeds. While there was 
sharp a increase in use of concentrate mix, the proportion of households that fed their dairy cows 
agro-industrial byproducts (wheat bran, oilseed cake, molasses, brewers grain, etc.); other high 
quality feeds (atela, grains); green forage, grass, hay, silage; and crop residue (teff straw, wheat straw, 
maize stover) has either remained the same or showed little changes during the period.
Table 6: Percentage of Dairy Cow Owners by Types of Feed Provided to Dairy Cows and Region at Baseline and Endline
Types of feed Amhara (%) Oromia (%) SNNPR (%) Tigray (%) Total (%)
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Concentrate mix
(factory formulated)
2.3 9.7a 3.7 46.4b 6.0 60.9c 4.3 39.1b 4.1 38.5
Agro-industrial 
byproducts
18.8 38.0a 36.0 26.4b 41.4 5.6c 22.5 29.0a,b 30.5 24.7
Other high quality 
feed (atela, grains)
93.4 86.0a 67.0 77.1b 47.6 18.7c 65.0 81.4a,b 67.2 65.1
4
There was an error in coding the survey questionnaire as related to local cows. As a result of the error, maximum and minimum milk production 
was captured on only 302 households that owned 500 cows. The maximum and minimum milk production data for local cows should have been 
collected for 917 households instead of 302. There was no error on data gathered for crossbreed and both types of cows. The evaluation team 
computed the minimum and maximum figures by assigning the average milk production of the 302 households to the other households (917-
302) from which this data was not captured by the tablet based survey
13
1
Green forage, grass, 
hay, silage 95.1 85.2a 75.2 86.9a 75.2 96.2b 92.0 98.4b 89.1 91.3
Crop residue 96.8 95.6a 96.o 95.2a 96.0 97.2a 94.8 98.0a 96.4 96.4
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
The training provided by the project dairy animal nutrition and management was one of the key 
factors behind the positive developments. As depicted in Figure 1 below, the proportion of 
respondents who received training on dairy animal nutrition and management grew from 13% at 
baseline to 88% at endline, and the difference was statistically significant. However, it is important 
to note here that almost all respondents are said to have participated in the training. The difference 
might have been partly caused by farmers’ inability to remember the training session in which they 
participated.
Figure 1: Percentage of households that received training on dairy animal nutrition and management
120
% of householdsthat received training on dairy animal nutrition and 
management
100 1 2 5 2 3 0 3 0 2 1
15 11
24
80
60
93 91
40
74 81 85
96 98
88
76 84
20
0
5 7
21 17 13
Amhara Oromia SNNP Tigray Total Amhara Oromia SNNP Tigray Total
Baseline Endline
Yes No No, but someone else in this HH has been trained
The evaluation team tried to examine if there was any relationship between amount of milk 
production and type of feeds used to dairy cows. As shown in Table 7 below, statistically significant 
differences were found in milk productivity between farmers that fed and did not feed concentrates, 
agro-industrial by-products, and other high-quality feeds. The maximum amount of milk collected 
from all dairy cows by households that fed these feeds to dairy cows during the reference period was 
significantly larger than those that did not. The use of concentrate feed was also found to play 
important roles in increasing the minimum as well as maximum amount of milk collected per day, 
and the project was credited for this. Apart from sensitizing farmers about the importance of 
concentrate feed, the project improved supply of and access to such feed products by establishing
14
union-based feed manufacturing enterprises and supporting backward and forward linkages. The 
head of a woreda livestock and fishery development in this regard said the following:
‘…The project trained farmers on modern animal feeding and other management, provided improved 
forage seeds and seedlings that used to be inaccessible, promoted and linked the manufactured feed 
supply and demand. The project initiated the manufactured feed supply and demand. Because of 
training and promotion, about 50% of beneficiary farmers buy manufactured feed for dairy and 
fattening. Farmers started focusing on better animal management as the awareness of farmers on 
modern animal management enhanced by the project through training, visit, field day and provide 
awards for best performing farmers.’
Indigenous dairy cows owned by households that feed concentrate feed produce milk for 7.5 months 
compared to 7 months for those that did not feed concentrates. In other words, the findings suggest 
that feeding concentrates make indigenous cows provide milk for longer period and the difference 
was statistically significant. But this was not the case for cross-breed cows (see Table 7).
Table 7: Milk Productivity and Type of Feed Fed to Dairy Cows
Concentrates Agro￾industrial by￾products
Other high
quality feeds
Green forage,
grass, hay, 
silage
Crop Residue
No Yes No Yes No Yes No Yes No Yes
What was the maximum amount of milk you
collected in one day from all of your cows during the 
past 12 months
5.01a 7.64b 5.76a 6.82b 5.52a 6.29b 4.85a 6.13a 6.12a 6.01a
What was the minimum amount of milk you
collected in one day from all of your cows during the
past 12 months
2.80a 4.91b 3.57a 3.75a 3.36a 3.74a 2.47a 3.72b 3.38a 3.62a
On average how many number of months did your
indigenous dairy cows produce milk (i.e. the typical
length of a single lactation)?
7.00a 7.48b 7.23a 6.86a 7.21a 7.12a 7.27a 7.13a 6.83a 7.16a
On average how many number of months did your
crossbreed dairy cows produce milk (i.e. the typical 
length of a single lactation)?
9.31a 9.12a 8.96a 9.66b 9.37a 9.13a 10.33a 9.09b 9.33a 9.21a
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
When we look at milk productivity against use of improved practices, as presented in Table 8, 
households that practiced drinking troughs to provide clean water for animals; feeding a total mixed 
ration; supplementation with urea molasses lick blocks; and factory produced formulated feed were 
more likely to collect more milk than those that did not use these improved practices. The impact on 
milk productivity is more pronounced among households that fed concentrates. The difference was 
statistically significant. Farmers as well as government officials indicated that the project enabled 
these changes by providing practical training; supplying initial improved forage seeds/seedlings as 
well as concentrate feeds; improving accessibility of concentrate mix; and supporting the provision 
of technical support. In this regard, one of the direct beneficiary farmers said the following:
‘The training providing by the project on livestock feeding and management was practical. After the 
training, starting materials were supplied in the form of foundation seed and planting material of 
forage crops; concentrate feeds of various kinds of formulations that they tested on their farm 
animals. We then realized the benefit of feeding better quality forage and -concentrate by the 
increased milk yield from three liters to more than eight liters. We built feed troughs exactly as
15
demonstrated during the training using locally available material and that proved to be handy in 
reducing wastage of feed material and avoidance of contamination with internal parasites. I also 
purchased a crossbreed heifer for ETB 14,000 and fed her with concentrate dairy ration; and the milk 
yield increased substantially from 8 lts to 15 lts/day.’
Table 8: Milk Productivity and Feeding Practices
Have you ever used any of the following improved practices in the past four years?
Feeding 
troughs to 
feed animals
Drinking 
troughs to 
provide clean 
water for 
animals
Feeding a total 
mixed ration
Baling
Urea 
treatment of 
crop residues
Supplementation 
with urea 
molasses lick 
blocks
Silage making
Factory 
produced 
formulated 
feed
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
Maximum
amount of milk
you collected in 
one day from
all of your cows
during the past
12 months
5.74a 6.21a 5.52a 6.63b 5.38a 8.74b 6.00a 6.11a 5.98a 6.22a 5.90a 7.52b 5.94a 6.99a 5.19a 8.04b
Minimum 
amount of milk 
you collected in 
one day from
all of your cows
during the past
12 months
3.29a 3.83b 3.18a 4.15b 3.15a 5.56b 3.57a 3.87a 3.58a 3.79a 3.54a 4.45a 3.54a 4.57b 2.99a 5.14b
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
2.1.2.3. Livestock Fattening
During the intervention period, the proportion of households that have been involved in fattening 
activities also showed modest increments from baseline. As depicted in Figure 2 below, the 
percentage of households that engage in fattening of cattle, sheep and/or goats during the 12 months 
preceding the survey increased from 43% at baseline to 51% at endline. As was the case at baseline, 
however, endline survey respondents might have considered any “feeding” of animals that would 
ultimately be sold for meat qualified as “fattening.”
Figure 2: Percent of Households that Fattened Cattle, Sheep and/or Goats during the Past 12 Months at Baseline and Endline
16
% of HHs who fattened livestock during the past 12 
months
80.0
70.0
60.0
50.0
40.0
30.0
20.0
10.0
0.0
74.2
52.1 53.1
38.6 38.4
51.1
26.3
38.8
42.5
50.8
Amhara Oromia SNNP Tigray Total
Baseline End line
Households involved in livestock fattening were asked about the type of feeds they feed their fattened 
ruminants. As shown in Table 9, 95% and 85% of the respondents used crop residues, and green 
forage, grass hay and/or silage, respectively, as major diets for livestock fattening across the study 
areas. The use of Formulated Concentrate Mixture (FCM) increased from 6% at baseline to 43% at 
endline. On the other hand, the proportion of respondents that fed ingredients of AIBP as 
supplements to their fattening livestock decreased from 42% at baseline to 35% at endline while 
little changed in use of other high-quality feeds, crop residue and others.
Table 9: Percentage of Respondents involved in Fattening by Types of Feed they Provided to Fattened Animals at Baseline and
Endline
Feed Types
Amhara (%) Oromia (%) SNNPR (%) Tigray (%) Total (%)
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Concentrate mix 3.7 14.0a 8.5 49.7b 9.1 77.7c 2.7 45.9b 6.3 43
Agro-industry 
byproducts
19.8 52.2a 62.7 40.2a,c 50.3 7.3b 31.9 35.9c 42.1 35
Other high￾quality feed 94.5 90.7a 68.7 78.8b 54.5 34.4c 70.9 86.7a,b 74 73
Green forage, 
grass, hay, silage
91.8 75.8a 75.6 86.0b 96.1 91.9b 80.8 91.7b 86 85
Crop residue 89.9 93.0a 94.5 93.3a 93.3 97.6a 96.8 96.1a 93.1 95
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
The large increase in proportion of households that fed concentrate mix to their fattened animals 
could be partly explained by exposure to fattening management training provided by the project as 
well as increased access to such feed. Over 93% of households received training on fattening 
management at endline, compared to just less than 12% at baseline. The training helped farmers to
17
engage in fattening and improve their fattening practices. ‘Most of us didn’t even know about 
concentrate feed and improved animal feeding practices. …hence the project can be considered a 
pioneer. There is improvement in livestock productivity due to the changes in our feeding practices, 
and the concentrate feed and other improved forages we have been feeding to our livestock,’ a farmer 
that participated in one of the FGDs said.
2.1.2.3.1. Cattle fattening
The project targeted to increase unit productivity of fattened cattle measured by number of cattle 
fattened/ year/100 HHs by 15%. At baseline, unit productivity of fattened cattle was 12.93 per year 
per 100 HHs. As illustrated in Table 10, across the project intervention areas and among those 
households that were involved in animal fattening, 92% of them (or 46% from total respondents) 
were involved in cattle fattening activities during the 12 months period preceding the endline survey 
compared to 68% (or 32% from total respondents) at baseline. The number of households that were 
involved in animal fattening as well as number of animals fattened have increased by 44% and 32%, 
respectively, from baseline. When computed from total number of survey participants, the average 
fattened cattle ownership increased from 0.68 cattle per household at baseline to 0.9 fattened cattle 
per household at endline, a 34% increase from baseline. Hence, when livestock productivity is 
measured by percentage increase in number fattened animals, it can be said that the number of 
fattened animals has increased by 34% from baseline. Unit productivity of fattened cattle measured 
by number of cattle fattened/ year/100 HHs also increased from 12.93 at baseline to 17.1 at endline, 
a 32% growth. Hence, the project has overachieved its target of increasing livestock productivity by
15%.
The proportion of households that engaged in cattle fattening at endline grew by close to 43% 
compared to the 613 HHs that were involved in cattle fattening at baseline. Government officials and 
farmers alike stated that farmers have been increasingly engaging in cattle fattening during the 
project intervention period that they believe was mainly due to the training provided by the project 
and increased awareness and access to improved animal feed. Recognizing the importance of 
engaging in fattening activities, many farmers are even fattening retired oxen to add value before sale 
which was not the case before the project intervention. In reply to the query as to whether there has 
been a change in livestock productivity due to the project support, a
replied: ‘We have observed good result in cattle fattening whereby farmers are feeding improved 
forage in combination with local beer byproduct, cereal straw and manufactured fattening 
concentrate feed obtained from the Union’s feed plant in the woreda.’ Another woreda government 
official also said the following which seems to be shared by farmers and other stakeholders:
‘… a significant number of farmers are working on fattening of bulls or oxen (4-20 heads) and 
sheep (12-15 heads). Time of fattening shortened to 3-4 months (earlier up to 1 year) and 
farmers started seasonal fattening by linking their activities to main holidays. Farmers have 
acquired important knowledge on manufactured formulated feed and started using for dairy 
and fattening. Generally, about 50% of beneficiary farmers have benefited either from dairy or 
fattening.’
Table 10: Percentage of Households by Region and Number of Cattle Fattened at Baseline and Endline
18
Number of Cattle Fattened
Total % change from baseline
Baseline Endline
Number of HHs that fattened cattle 613 874 43%
Percentage of HHs that fattened cattle (from total
respondents) 32 46 14%
Total cattle fattened 1,293 1709 32%
Number of fattened cattle per year per 100 HHs 12.93 17.1 32%
Mean number of cattle fattened (from total survey
respondents) 0.68 0.9 34%
Length (No. of days) of fattening per cycle 116.6 103 12%
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
As also shown in Table 10 above, the length of each cattle fattening period/cycle was reduced from
117 days at baseline to 103 days at endline. Analysis of qualitative data also confirms the same. Many 
farmers and government officials indicated that the fattening period has declined to 3 to 4 months. 
Direct beneficiary farmers and government officials associated the reduction in fattening period to 
increased use of improved animal feed and feeding practices. The intervention was found to be a key 
contributor to these positive developments by helping farmers learn improved livestock 
management and feeding practices, promoting improved forage and manufactured feed, making 
manufactured feed and improved forage seed/seedlings accessible to farmers. ‘… the feed we use to 
fatten livestock helps the oxen to gain weight in relatively shorter period of time,’ a farmer that 
participated in one of the FGDs said. Similarly, a said the following 
to explain how and why the reduction in fattening period happened.
‘Farmers started focusing on better livestock production as their awareness of farmers on modern 
animal management enhanced through training, visit, field day and awards for best performing 
farmers. The project distributed improved forage such as oats, vetch, elephant grass, desho and 
alfalfa at kebele level. Apart from supporting establishment of feed manufacturing enterprises, the 
project supported primary cooperatives supply manufactured feed from feed enterprise. The 
distribution of manufactured feed through dairy cooperatives increases accessibility of feed. Farmers 
have also got better access to balanced feed to increase productivity of their animals. As a 
consequence, the productivity of livestock is improving. Milk yield increase by about 2 to 4 kg in 
crossbred cows, fattening period reduced from 7 to 3 - 4 months.’
2.1.2.3.2. Sheep and Goat Fattening
The proportion of households that fattened sheep declined slightly from 20% at baseline to 17% at 
endline. The number of sheep fattened also declined from 2,126 at baseline to 1,547 at endline (Table
11). Although the proportion of farmers engaged in sheep fattening declined slightly, the project 
helped farmers to engage in sheep fattening activities. A woreda agriculture office expert in this
regard said the following:
‘The project encouraged sheep fatteners in the woreda whereby some farmers using manufactured 
fattening ration for sheep achieved good result that the sale of fattened sheep dramatically raised 
their household income; one of the fatteners has even accumulated enough capital to purchase a 
Bajaj taxi.’
19
Table 11: Percentage of Households by Region and Number of Sheep Fattened at Baseline and Endline
Number of Sheep
Fattened
Region Total
Amhara Oromia SNNPR Tigray
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
No sheep fattened (%) 71.3 59a 73.5 89b 85.5 93c 88.7 92b,c 79.8 83%
1-4 sheep fattened (%) 22.8 18.75a 13.5 1.94b 13.4 4.76b 4.4 0.00 13.5 6.44
5-9 sheep fattened (%) 3.1 16.88a 5 4.74a 0.8 1.86 3.8 2.57a 3.2 6.55
10-49 sheep fattened (%) 2.7 5.63a,c,d 7.9 4.09a,b 0.2 0.62c 2.9 4.28b,d 3.4 3.64
50 and + sheep fattened
(%)
No. HHs that fattened 
sheep
- 0.21a - 0.00 - 0.00 0.2 1.07a 0.1 0.32
137 199a 127 50b 69 35c 54 37b,c 387 321
Total sheep fattened 534 738 966 299 174 69 452 441 2,126 1547
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
As was the case at baseline, goat fattening was found not to be a common practice in the project 
intervention areas as only 3% of respondents reported to have fattened goats. At baseline, only 3.8% 
of respondents were involved in goat fattening.
2.1.3. Poultry Management and Productivity
2.1.3.1. Poultry size and breed kept by households
Direct beneficiary smallholder farmers that participated in the survey were also asked to provide 
information about their chicken ownership. As Table 12 below presents, about 35% of households 
owned egg laying local hens, which was a significant drop from baseline of 63%. On average a 
household owned 1.8 local egg laying hens at the time of the endline survey, compared to 3.53 at 
baseline. The decline in local chicken ownership was due to the fact that households were 
increasingly buying improved chicken.
Table 12: Households Ownership of Local and Improved Chickens during the 12 Months Preceding the Survey by Types and
Region at Baseline and Endline
Chicken Types
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Total egg laying local hens 1,018 1,049 1,392 488 781 277 1,049 546 4,240 2,360
Mean egg laying local hens
per egg laying local hen
owner HH
3.32 2.78a 4.24 1.53b 2.84 1.01c 3.59 1.58b 3.53 1.80
20
Number & % of HHs with egg 
laying local hens
307 
(64%)
277 
(58%)
328 
(68%)
139 
(30%)
275 
(58%)
96 
(20%)
292 
(61%)
146 
(31%)
1,202 
(63%)
658 
(35%)
Total HHs (n) 478 480 480 464 477 483 479 467 1,914 1,894
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
In contrast, ownership of improved egg laying chickens increased significantly among endline survey 
participant households compared to baseline. As shown in Table 13, about 53% of households owned 
improved chickens at the time of the endline survey, compared to 22% at baseline. The average 
number of egg laying improved chickens owned by a household increased from less than 1 at baseline 
to over 4 at endline. This change was also confirmed by government officials and farmers that 
participated in qualitative interviews. “Some farmers are shifting from keeping indigenous chickens 
to improved hybrids for egg production. About 30 - 40% of beneficiary farmers have started keeping 
improved chicken. Some farmers have also engaged in day old chicken rearing and egg production 
from improved chickens,” a woreda government official said.
Table 13: Total and Average Number of Improved Chickens by Region at Baseline and Endline
Chicken Types
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Total egg laying improved 
hens
284 1128 331 1332 364 848 502 2042 1481 5350
Mean egg laying improved 
hens per household that own 
such chicken breed
0.6 2.99a 0.7 4.19b 0.8 3.11a 1 5.90c 0.8 4.07
Total number & % of HHs 
with egg laying improved 
hens
92 
(19.2%)
246 
(65.3%)
92 
(19%)
247 
(53%)
85 
(18%)
217 
(45%)
149 
(31%)
299 
(64%)
418 
(22%)
1009 
(53%)
Total HHs (n) 478 480 480 464 477 483 479 467 1,914 1,894
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
Ownership of improved pullets and male chickens aged over three months was also observed 
compared to baseline. On average, a household owns 0.74 female and 1.15 male improved chickens 
aged over 3 months years/year/HH, compared with 0.4 at baseline. Ownership of male chickens 
varied significantly among the regions.
A local hen on average laid 124 eggs per year, an over 92% increase in egg production compared to 
baseline. Annual egg production from improved chickens was over twice as much as that of local 
chickens. On average, an improved chicken laid about 265 eggs per year, which was also an over 41% 
increase from baseline average of 187 eggs per year. The average egg laid by local and improved 
chickens seem to be very large (see Table 14 below). The national average for number eggs produced 
per year by local and improved hens was around 50 and 150 (World Bank, 2017). But this national 
average might be a bit old and might not be the current reality. In addition, the large increase in egg 
productivity was mainly associated with the use of manufactured poultry feed and improved poultry
21
management practices, according to government officials and many farmers that participated in
FGDs.
Apart from shifting to keeping improved chicken, farmers are using improved poultry feed and 
applying the improved poultry management practices, which are said to increase the egg productivity 
of local and improved chickens. The following is what one farmer that participated in one of the FGDs 
said “…we currently own more improved chickens than before. These chickens lay about 27-30 eggs 
per month.” This suggests that the annual amount of egg production stated above actually makes 
sense.
Table 14: Total and Average Egg Production by Breed Type and Region at Baseline and Endline
Egg producing hens
Amhara Oromia SNNPR Tig-ray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Total egg laying local hens 1,018 1049 1,392 488 781 277 1,049 546 4,240 2360
Mean egg production per 
year/local hen 66.2 131a 66.1 117b 62.5 115b 62.5 121a,b 64.4 124
Total egg laying improved 
hens
284 1128 331 1332 364 848 502 2042 1,481 5350
Mean egg production per 
year/improved hen 183.02 299a 182.6 220b 193.9 280c 189.7 2612d 187.3 265
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
2.1.3.2. Poultry management
The findings of the endline survey suggests improvement in improved poultry management 
practices. As shown in Table 15, the proportion of households that did not use housing or used 
unconfined shelter for poultry declined from 26% and 49% at baseline, respectively, to 12% and 13% 
at endline. In contrast, households that use confinement and shelter showed large increment. The 
percentage of households that use confinement and shelter at night or year-round increased from
23% and 2% at baseline to 56% and 19% at endline, respectively. The majority of households still 
use shelter and confinement at night. Less than a fifth of the households use confinement and shelter 
year-round and respondents from Oromia and SNNP are more likely to use these improved practices 
than others. It is also important to note that poultry housing system of households differed among 
some of the regions. Households in SNNPR are less likely not to have housing for their poultry than 
the rest of the regions, despite this fact, the practice of no housing in the same region has shown a 
decline at endline.
Table 15: Percentage of Chicken Holder Households by the Type of Housing System and Region at Baseline and Endline
Types of poultry housing 
system
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
22
No housing 32.0 18.5a 35.3 12.1a 27 13.1a 11.7 4.5b 26.4 12.1
Shelter but unconfined 58.4 7.1a 47.9 13.6b 40.6 7.2 a,b 46.6 21.7c 48.5 12.6
Shelter and confined at night 10.1 59.2a 15.1 45.7b 32.2 56.7a 36.9 62.6a 23.4 56.3
Confinement and shelter 
year-round
- 15.2 a,c 1.6 28.6b 0.3 23 a,b 4.7 11.2c 1.7 19.1
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
Positive changes were also observed in type of poultry feed used by households over the past few 
years. As shown in Table 16, scavenging home grain supplements has remained the dominant poultry 
feeding system (81%), which was also the case at baseline. But noticeable changes were observed in 
proportion of households that practice scavenging with concentrated feed supplements, full feeding 
with home grown grains as well as full feeding with manufactured feed, which were almost negligible 
at baseline. Scavenging with concentrated feed supplements is now a poultry feeding system among
35% of households at endline, compared to 5% at baseline. The proportion of households that 
practiceed full feeding with home grown grains or with manufactured feed also increased from 3% 
and 0.2% at baseline to 33% and 15% at endline, respectively. However, it is unclear why scavenging 
showed increment while at the same time an increased number of households adopted improved 
practices.
Table 16: Percentage of Chicken Holder Households by Types of Poultry Feeding Systems and Region at Baseline and Endline
Poultry Feeding system
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Scavenging 7.3 27.6a 8.0 64.7b 6.0 49.5c 5.8 12.0d 6.8 37.1
Scavenging with home 
grain supplements
91.9 72.9a 88.5 84.7b 82.7 68.7a 76.7 95.5c 85 80.9
Scavenging with 
concentrated feed 
supplements
0.8 38.0 a,b 3.0 31.5 a,c 9.9 45.0b 7.8 28.6c 5.3 35.3
Full feeding with home 
grown grains
- 39.2a - 34.7a 1.5 39.2a 9.4 19.3b 2.8 32.8
Full feeding with 
manufactured feed - 21.0a 0.5 19.1a - 19.9a 0.3 2.1b 0.2 15.3
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
Improvements in poultry management practices can be partly attributed to the training provided to 
beneficiary households. The percentage of households that participated in poultry nutrition and 
feeding training grew from less than 10% at baseline to 77% at endline. Households in Oromia and 
SNNP were more likely to participate in such training than those in Amhara and Tigray regions. It is 
also important to note that the proportion of respondents that received training on poultry nutrition 
and feeding differed from those that received training on improved fattening management (93%) 
which were provided at the same time. The difference might be partly due to difficulties faced by 
some respondents to remember training sessions they attended.
23
The evaluation team also tried to examine any relations that may exist between amount of egg 
production and poultry feeding practices. As shown in Table 17, the average egg production laid per 
local hen was higher among those households that practice scavenging alone or with home grain or 
commercial feed supplements, and full feeding with manufactured feed, compared with those that 
did not practice these methods. Average egg production laid per improved hen was higher among 
households that practiced scavenging with commercial feed supplements, full feeding with home 
grown grains, and full feeding with manufactured feed, compared with those that did not practice 
these methods. It is also interesting to observe that average egg production laid per improved hen 
would suffer when households feed their chickens by scavenging alone or with home grown 
supplements. In other words, improved hens were not responsive and productive when they are fed 
with these methods.
Table 17: Average Egg Production during the Past 12 Months Preceding the Survey, and Poultry Feeding Practices by Region
Did you feed your chickens
by scavenging?
by scavenging with 
supplements 
(home grains)?
by scavenging with 
supplements 
(Commercial feed)?
with full feeding 
with home grown 
grains?
with full feeding 
with 
manufactured 
feed?
No Yes No Yes No Yes No Yes No Yes
Average egg production (number 
of eggs) laid per local hen Per 
Clutch during the past 12
months
13.91a 13.63a 14.69a 13.59b 13.61a 14.25b 13.56a 14.33b 13.57a 15.15b
Average egg production (number 
of eggs) laid per local hen during 
the past 12 months
110.77a 126.92b 100.11a 120.56b 118.52a 112.04b 121.54a 105.86b 119.91a 97.41b
Average egg production (number 
of eggs) laid per improved hen 
during the past 12 months
267.23a 239.64b 267.59a 254.63b 254.70a 261.67a 248.06a 273.33b 252.45a 279.55b
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
2.1.4. Sustainable Forage Production Systems
Sustainable forage production was one of the strategies of the project to achieve its objectives and 
goals. In this section, the findings of the evaluation as related to forage production are presented in 
this section.
2.1.4.1. Improved forage production and cultivation at farmer level
There were different initiatives that were implemented by the project to develop sustainable forage 
production system. Among these included the provision of training, seeds/seedlings and technical 
support for model farmers to grow improved forage in their backyard. Almost all households that 
participated in the endline survey received training on improved forage production during the past 
five years. The proportion of households that received such training increased from less than 10% at 
baseline to over 96% at endline. Households were reached through government extension agents, 
forage experts who received TOTs and FEED II field staff. The training was conducted with 
theoretical/classroom sessions, practical demonstration and site observations in the field. During the 
site visits, best forage development activities from the established nursery sites, farmers’ training
24
centers, individual farmers and investors were viewed, and trainees shared experiences and lessons 
learned.
Baseline and endline percentages of households the cultivated improved forage along with the size 
of land allocated to that activity is presented in Table 18. Over 71% of households cultivated 
improved forage during the 12 months preceding the survey at endline, which was a more than 
double increase compared with the 30% at baseline. The surveyed direct beneficiary households 
stated that provision of seeds and seedlings along with training increased their adoption of improved 
forage cultivation practices. The findings also suggest that some of the households that received the 
training have not cultivated improved forage (96% of HHs received training but 71% of HHs 
cultivated improved forage). This could be partly associated with lack of land, shortage of water, and 
small amount of improved forage seed received from the project. On average, a household planted 
improved forage on 303m2 plot of land during the 12 months preceding the survey (or 40.9 hectares 
among all those surveyed who cultivated forage during the period), which was a three-fold increment 
from baseline of 101m2. Considering the 37,261 smallholder farmers that were trained by the project 
to establish and maintain sustainable forage plots, the findings could mean that the 1,129 hectares 
were planted to improved forage over the past five years. Many farmers also plan to increase their 
improved forage production in the future hoping that seed/seedlings would be available.
Table 18: Percentage of Households that Cultivated Improved Forage 12 Months Preceding the Survey by Region at Baseline 
and Endline
Have you 
Cultivated Forage 
During the past 12
Months?
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Yes 37.2 76.7a 26.2 61.4b 38.6 91.3c 17.5 54.6b 29.9 71.2
No 62.8 23.3a 73.8 38.6b 61.4 8.7c 82.5 45.4b 70.1 28.8
Total (n) 478 480 480 464 477 483 479 467 1,914 1894
Average land size 
planted to
improved forage 
(computed from 
total sample HHs_
142m2 630m2 124 m2 59m2 80m2 236m2 59 m2 280m2 101m2 303m2
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
Statistically significant variations were found among regions in plot of land they planted to improved 
forage, as shown in Table 18 above. Plot of land planted to improved forage is highest in Amhara and 
lowest in Oromia. Indeed, the amount of land planted to improved forage declined in Oromia 
compared to baseline. It is unclear why this decline occurred given the fact that many farmers were 
made aware of the improved forage varieties by the project intervention and started cultivating these 
forage plants after that. The fact that some of the nursery sites have not been operational for the last 
two years following withdrawal of the project support might have hampered availability of forage
seeds and seedlings, and ultimately cultivation of improved forage in the region. Shortage of 
water/drought also affected some of the project intervention areas such as woreda. A farmer 
that participated in FGD in this regard said ‘We have cultivated and adopted oats and vetch improved 
forage. However, due to land cracking during the dry season some improved forage such as tree
25
lucerne, elephant grass, desho couldn’t sustain in the area.’ The amount and timing of seed 
distributed to farmers, lack of land for forage production, and shortage of water were also mentioned 
as bottlenecks to improved forage cultivation. A farmer that dwells in one of the affected woredas in 
Oromia said the following:
‘…We have learned about improved forage but the amount of seed was too small to expand. We 
also lack land for forage production. Furthermore, shortage of water and free grazing during 
dry season destroyed the cultivated improved forage. The delay of seed distribution has also 
affected the adaptation of cultivated forage.’
The project was very successful in creating awareness among farmers and government experts about 
the different kinds of improved forages. Farmers are also increasingly allocating land for improved 
forage plantation. The demand for improved forage seed and seedlings has also increased 
significantly. Some have managed to produce their own seeds and even sell to other farmers or to the 
market. However, many beneficiaries complained about the unavailability or shortage of seed or 
seedlings. Forage nursery sites established by the project bridged some of these gaps by distributing 
seeds and cuttings free of charge. Initial seeds and seedling were supported by FEED II project and 
the nursery sites were producing the seeds and seedling for replacement and dispatch. But their 
capacity has remained limited and some are not functional. Some farmers tried to buy seed from the 
market but were not able to do so due to their high price and fear that the seeds might not be genuine 
seeds. As also depicted in Figure 3 below, only 6% of direct beneficiary households bought improved 
forage seeds, seedlings, cuttings or splits.
Figure 3: Percentage of respondents that bought forage planting materials (seeds, seedlings, cuttings, splits, etc) or got them for 
free
PERCENTAGE OF RESPONDENTS THAT BOUGHT FORAGE PLANTING 
MATERIALS OR GOT THEM FOR FREE
Bought them
6%
Got them free 
of charge
94%
26
Other challenges that hampered the cultivation of improved forage relates to adaptation of some 
forage varieties such as alfaalfa and desho. ‘The major challenges of improved forage production 
were shortage and delay of distribution of seed and seedlings. Some of the forage varieties such as 
alfalfa and desho couldn’t adapt to the area,’ a woreda livestock office official said. In other areas, 
there was also adaptation problem of some forage varieties such as desmodium, Rhodes, Desho and 
elephant grass, which hampered even a more improved forage development and extension than what 
was observed during the final evaluation.
Households that cultivated improved forage 12 months preceding the survey were asked about their 
use of improved forage planting materials. As shown in Table 19 below, the proportion of households 
that used improved forage planting materials (seeds, seedlings, cuttings, splits, etc.) more than 
doubled over the past 5 years. Among all surveyed households 52% of them used improved forage 
planting materials, compared to 24% at baseline. Region-wise, respondents from SNNPR were less 
likely to use improved forage planting materials than the rest three. The sources for these improved 
forage planting materials were government (35% at endline vs 17.6% at baseline), other farmers 
(1% at endline compared with 8% at baseline), purchases from private venders (2%), and NGOs 
(19% at endline versus 2% at baseline). This suggests that the proportion of households that received 
improved forage planting materials from government and NGOs increased significantly over the past 
five years. Although pure attribution is difficult given the non-experimental nature of study design, 
qualitative data analysis results suggest that the project (identified in the name of government or 
itself) was the major supplier of improved planting materials.
Table 19: Percentage of Households that Cultivated Improved Forage 12 Months Preceding the Survey by Region at Baseline 
and Endline
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
DID YOU USE IMPROVED FORAGE PLANTING MATERIALS (SEEDS, SEEDLINGS, CUTTINGS, SPLITS, ETC.) OVER THE PAST 5 YEARS?
Yes 32.8 58.1a 20 49.8a 31.7 57.1b 12.3 42.4a 24.2 52.0
No 4.4 18.5a 6.2 11.6a 6.9 34.2b 5.2 12.2a 5.7 19.3
No forage cultivation done
5 62.8 23.3 73.8 38.6 61.4 8.7 82.5 45.4 70.1 28.8
Total (n) 478 480 480 464 477 483 479 467 1914 1894
WHAT WERE THE SOURCES OF IMPROVED MATERIALS?
Government 17.8 50a 14.6 5.2b 26 41.6c 12.1 40.3d 17.6 34.5
NGO 1.3 6.0a 0.6 45.9b 5 15.3c 1.9 7.3a,c 2.2 18.5
Purchased from private
vender
0.6 2.5a 7.9 0.9a 2.9 1.2a 0.2 2.4a 2.9 1.7
Farmer exchange 16.3 2.1a 5.8 0.9a 10.9 0.6a 0.4 0.2a 8.4 1.0
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
5 No forage cultivation done” refers to missing values as the question was asked to households that said they have 
not cultivated improved forage. In other words, the ‘no forage cultivation row is indicated in the table to show that 
the percentages are computed from the total number of surveyed households rather than from the valid total as was 
also the case in the baseline study.
27
2.1.4.2. Establishment of Forage Development/Nursery Sites
FEED II in collaboration with government, and cooperative unions established union-based forage 
nursery sites. The government provided land for nursery site development and provided supportive 
supervision and other support to the nursery sites. In some regions such as Amhara, the government 
has also started allocating budget to pay salary of nursery site staff and operate the site after the
project stopped its funding.
Ligaba Nursery Site (Amhara Region)
FEED II assisted the establishment/development 
of 14 union-based nursery sites and some private 
forage enterprises by providing training, and 
technical assistance as well as provision of parent 
seed and seedlings. Moreover, the project supplied 
financial and in-kind supports for constructing 
fences and store. It also provided tools, irrigation 
pump, accessories and others necessary for 
cultivation. It is also the project that paid the 
salaries or provided top-ups for the site staff. 
Furthermore, the project capacitated union staff 
and government extension agents on nursery 
management. It also organized domestic learning 
exchange visits for forage production
stakeholders. The project has also developed good
nursery management capacity. Apart from union-based nurseries, FEED II also supported the 
establishment of community youth group-based forage enterprises in some areas.
Many of the nursery sites established through the support of the project have now become sources 
of seeds and seedling. Although functionality of nurseries seems to vary from one region to another, 
those that are operational have been producing seed/seedlings to support forage production by 
smallholder farmers. There also seems to be a joint ownership of nurseries by woreda 
agriculture/livestock office and unions. But it is usually the woreda governments that are playing 
important roles in the operation of the nurseries.
28
The woreda governments are running the 
nursery sites which is vital for their 
sustainability. After the project withdrew its 
support about two years ago (according to 
key informants), some regional 
governments started allocating budget to 
run the nurseries. While most nurseries 
often provide seed/seedlings free of charge, 
some have started generating income which 
are also important in reducing financial 
burden of government and sustaining their 
operation. But they are more like exceptions. 
One of such nursery sites is found in SNNPR 
which provides some seed/seedlings free of 
charge while charging for others. In the last
12 months that ended on July 7, 2018, the 
nursery site managed to produce 600,000
splits of Desho grass, 258,315 cuttings of
Nadugne Agam Forage Development Site (SNNPR)
Elephant Grass, 250,000 splits of Guatemala grass, and 400,000 pieces of Phalaries grass. While the 
site provided the Desho and Phalaries grasses that it produced during the period free of charge to 
farmers, it generated around ETB 210,000 from sale of Elephant Grass and Guatemala grass that it 
produced at reasonable price. The income generated from sale of seed/seedlings was shared among 
the woreda agriculture office, union and farmer training center (FTC), which all are said to have used 
to support the nursery site. For example, the woreda agriculture office used its share to purchase 
inputs for the nursery site.
Nursery sites have been playing vital roles in ensuring continued availability of improved forage 
seed/seedlings to farmers. They were also creating awareness among farmers in surrounding areas 
about the importance of improved forage. A nursery site manager in this regard said the following:
‘The seeds and seedlings distributed to farmers were free of payment. The distribution was 
conducted at kebele level though development agents and the project supported the 
transportation facilities and other costs. When the nursery site was operational, we distributed 
about 50,000 seedlings of Sesbania, 40,000 seedlings of tree lucerne, 16,000 seedlings of fodder 
beet from the nursery site to two woredas, including ours. In addition, the project distributed 
significant amount of oats, vetch, desho and elephant grass to farmers.’
29
The project has been very successful in creating demand for improved forage. There is high demand 
of seeds and seedlings/ cuttings from farmers and the demand is increasing over the years. However, 
this demand is not being met by the nursery sites due to
their limited capacity. Nursery sites in some areas such as 
in Tigray and Oromia are not also currently functional. 
While lack of budget allocation by government when the 
project ceased its support was the major reason identified 
for non-functionality of nursery sites in Oromia. They are 
still dependent on external support as they don’t generate 
income of their own or government does not allocate 
budget to run them. Woreda governments have also a 
budget too small to provide continuous monitoring and 
supervision of the nursery site. As a result of shortage or
lack of budget, nurseries are also being challenged by lack
Sodo Buee Nursery Site (SNNPR) of transportation services for collecting and delivering 
seeds/seedlings for farmers. Some also lack a focal person
for monitoring day to day activities of the nursery site. The involvement of unions was also found 
inadequate. As a result, their support to the nursery site has not been that strong which was 
associated with the fact that the nurseries are not generating income for the unions. Indeed, nursery 
sites lack clear ownership. There is weak relation between union and woreda agriculture/livestock 
and fishery development offices. A forage expert in one of the woreda Livestock and Fishery
Development offices in this regard said the following:
‘The project supported our office to establish of 
nursery site financially and technically. It also 
provided initial seed and support the 
distribution. FEED II covered the cost of daily 
labor, provided training for expert, purchased 
small generator to pump water, purchased 
small farm equipment, helped seed and seedling 
distribution and arranged experience sharing 
visit and train farmers. We have developed 
interest at farmers level on importance and 
contribution of improved forage and bring 
change on livestock productivity. In addition, we 
have developed capacity on nursery site 
management. The seeds and seedlings produced 
by the site were distributed for farmers free of
payment. The distribution was conducted at
Azedebo Forage Development Site
(SNNPR)
kebele level though development agents and the project supported the transportation facilities 
and other cost of transportation. The only source of fund for this nursery site was FEED II 
project. There is high demand of seeds and seedling/ cuttings from farmers and the demand is
30
increasing over the years. The nursery site is not operational in the last two years due to absence
of budget allocated by the project and government.’
In sum, the project was instrumental in creating awareness about improved forage among 
government experts and farmers. Farmers started allocating land for forage development and 
focusing on better livestock production as the awareness of farmers on modern animal management 
is enhanced through training, farm visits, field days and awards for best performing farmers. As a 
result, the productivity of livestock is improving (see relevant sections of this report for more 
details). Some of the respondents were also able to sell forage seed to generate additional income. 
The direct beneficiary households involved in FGD also indicated that the intervention helped in 
reducing mortality of calves. This qualitative result obtained from FGD is obviously not only a 
function of the improved forage. It is a result of a combination various intervention such as improved 
feeding practices like using formulated concentrate mixture. The interviewees in one of the project 
intervention woredas of Amhara region also said that the training and demonstration of improved 
forage crop production and utilization has brought about attitudinal changes among farmers in favor 
of cultivated forages rather than the free grazing system as indicated by substitution of maize with 
improved forage crop cultivation like Rhodes grass and oats-vetch mixture by some farmers who 
own mature crossbred dairy cows and relatively have better market access for dairy products. 
Rhodes grass cultivation in some areas of Amhara region is likely to continue. According to
in one of the woredas of Oromia regions, around 50% percent of 
beneficiary farmers cultivated and used improved forage varieties such as oats, desho and elephant 
grasses.
2.1.4.3. Impact of project forage production activities on forage milk production
Many households that participated in focus group discussions appreciated the introduction of the 
new improved forage varieties that they have never been aware of. Farmers are happy with the 
project as it introduced them to additional animal food sources. ‘The contribution of nursery site is 
high to expand improved forage and reduce dependency on crop residue,’
said. Another government official also said the following:
‘There are two major positive results from FEED II project. The first is improved forage. For the 
first time in decades of livestock extension history, unknown forage species have been introduced 
by VOCA projects and multiplied at a nursery site for subsequent distribution to farmers. Again, it 
was VOCA that established the forage nursery on a 1.5 ha of fenced irrigable land and the 
operation costs was being financed by the FEED II project until last June. At least ten perennial 
forage species have been multiplied at the nursery. By means of this activity the project availed to 
farmers’ alternative species of forages that suit different production systems and land holdings of 
the smallholder farmers.’
Direct beneficiary households that participated in FGDs also indicated that feeding improved forage 
to their dairy cows and other livestock increased their productivity. A case story presented below 
also illustrates this assertion. Government officials also had a similar opinion.
‘…now improved forage production has gained acceptance in the woreda. Farmers especially those 
having cross-bred cows have realized that feeding of improved forage increases milk yield. These
31
activities in turn encouraged improved forage crop production …feeding animals on a locally made 
feed trough which also has facilitated on-farm feeding and a retreat from free grazing tradition’ a 
government official said.
Survey findings also showed that households that used improved forage were able to increase the 
average number of months indigenous and crossbreed dairy cows produce milk to 7.45 and 9.6 
months, respectively, compared to 6.6 and 8.37 at baseline. The change was also statistically 
significant. However, statistically significant differences were not found in minimum and 
maximum milk productivity between households that used and did not use improved forage 
planting materials over the past five years (see Table 20).
Table 20: Use of improved forage and milk productivity
Did you use improved forage planting
materials(seeds, seedlings, cuttings, 
splits, etc) over the past 5 years?
No Yes
Maximum amount of milk collected in one day from 
all of cows during the past 12 months
5.76a 6.02a
Minimum amount of milk collected in one day from 
all of cows during the past 12 months
3.57a 3.49a
Average number of months indigenous dairy cows 
produce milk (i.e. the typical length of a single
lactation)
6.60a 7.45b
Average number of months did crossbreed dairy 
cows produce milk (i.e. the typical length of a single 
lactation)
8.37a 9.60b
The survey findings also showed that a small number of farmers benefited from sale of seed and 
seedlings. About 1% and 8% of respondents sold seed and seedlings, respectively, during the 12
months preceding the endline survey.
, who is married with five children, lives in Woreda in Amhara region. She 
received training from the project on production of improved forage, livestock nutrition 
management, improved feeding system, and improved poultry/dairy production She implemented 
improved livestock feeding and management practices like dairy production from a cross-bred cow; 
egg production from improved chicken breeds; and production of improved forage (Rhodes).
Apart from training, she also received farm supplies, including forage seed free of charge (even 
though it was not enough), including Rhodes grass that she planted at the backyard on about 800 
square meter plot. She also received concentrate feed from the project. She mentioned that before 
initiation of the projects, she never heard of such improved forage crops. After planting improved 
forage, she has been harvesting these forages and feeding tethered stock on feeder trough built from 
local wood materials. Feeding cattle using such home-made feeder troughs helped her minimize
feed wastage and protect animals from parasites.
32
When asked about the benefits her family obtained from the project support, said that the 
project helped her gain an understanding of improved animal feed and dairy/poultry production; 
increase forage from a piece of land (three times a year) by planting Rhodes; and increase her 
income by selling Rhodes seed (1500 ETB) and poultry products. Her Rhodes grass can feed her 
cattle from a small piece of land and she got money from sale of its seed. She also owns two 
crossbreed cows.
According to , her household livelihood has changed from a subsistence mixed farming to a 
moderately technology-oriented way of farming and improved livestock production through 
improved feeding and improved forage production. However, although she was able to increase her 
milk and egg production, has not been able to earn deserving income due to lack of access 
to markets. While she is very satisfied with what she gained from the Project support, she did not 
conceal her dissatisfaction with availability and accessibility of manufactured feed.
Plate -1: at her Rhodes farm Plate 2: , & her husband (from left 
to right) at her Rhodes farm
Plate -3: improved calfs eating Rhodes Plate 4: Utilization of Rhodes
2.1.5. On-Farm Feeding Practices and Adoption of Other Improved
Practices
Promotion of on-farm feeding practices was one of the strategies employed by the project to improve 
livestock productivity. In this section, the findings of the final evaluation as related to on-farm feeding 
practices are presented in comparison with the baseline results.
33
2.1.5.1. Farm management and improved livestock feeding and nutrition practices
As shown in Table 21, the percentage of endline survey participating households that practice cut 
and carry, supplementation of crop residues with green fodder, and formulated feed showed 
noticeable increment compared to baseline. In contrast, a decline was observed in proportion of 
households that use troughs to provide clean water for animals, practice rotational grazing, and hay 
making. The reduction in use of rotational grazing system and hay making practices by the 
households might be associated with the increased in cut carry system practiced by the respondents 
from 76% at baseline to 89% at endline. It is unclear why the percentage of households that provide 
free choice access to clean water dropped from 63% at baseline to 21% at endline.
Table 21: Percentage of Households that Adopted Improved Practices at Baseline and Endline
Improved Practices
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
Cut and carry 84 93a 70 84b 82 92a,c 66 88b,c 76 89
Feeding troughsto feed animals 53 40a 58 49 38 53a 74 84b 56 59
Using troughsto provide clean 
water for animals
43 35a 44 42a 51 39a 60 63b 49 45
Rotational grazing 28 19a 37 33b 48 32b 10 29b 31 28
Record keeping of any kind 6 21a 3 10b 6 12b 6 25a 5 17
Feed a total mixed ration 7 16a 21 28b 18 10a 19 24b 16 19
Forage chopping 36 42a 45 40a 66 68b 41 66b 47 54
Baling 4 14a,c 1 6b 40 14a 9 20c 13 14
Supplementation of crop 
residues with green fodder
23 62a 19 76b 53 58a 43 65a 34 65
Urea treatment of crop residues 3 15a,c 2 10a 13 21b 6 16b,c 6 15
Supplementation with urea 
molasseslick blocks 2 4a 2 7.1a 4 4.8a 4 16.5b 3 8
Hay making 79 84a 70 48b 67 45b 73 73c 72 63
Silage making 1 3a,b 2 7a 8 1b 2 20c 3 7
Feeding factory formulated feed 4 6a 4 49b 11 27c 5 35d 6 29
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
The proportion of respondents along with the number of improved practices they adopted is depicted 
in Figure 4 below. About 10% of the respondents adopted 1 to 2 practices. The majority (71%) of 
respondents adopted from 3 to 6 practices. Close to a fifth of respondents adopted over 6 improved 
practices.
Figure 4: Number of Improved Practices and Percentage of Households that Adopted the Practices at Endline
34
PPercent of respondents
NUMBER OF IMPROVED PRACTICES AND % OF HOUSEHOLDS 
THAT ADOPTED THEM
25.0%
20.0%
15.0%
10.0%
5.0%
0.0%
0 2 4 6 8 10 12 14
Number of improved practices adopted
2.1.5.2. Knowledge transfer /adoption of improved practices/technologies
The final evaluation included some questions to measure the extent to which model farmers and 
direct beneficiary farmers transferred their training/knowledge. Both direct beneficiary and 
follower farmers were asked about knowledge sharing. As depicted in Figure 5 below, about 43% of 
direct beneficiary households transferred their knowledge to other farmers. Knowledge sharing was 
highest in Amhara region and lowest in Oromia region.
Figure 5: Percentage of households that did/did not transfer knowledge they gained through the project
90%
80%
70%
60%
50%
Percentage of respondents that passed on the 
training/knowledge they gained/received from/through 
the project to other farmers
83%
63% 60% 57%
40%
30%
20%
10%
0%
37%
17%
49% 51%
40% 43%
Amhara Oromia SNNP Tigray Total
No Yes
Households that transferred knowledge to other farmers were also asked about the number of 
farmers to whom they transferred the knowledge. The findings showed that households transferred
35
their knowledge to close to 6.446 (median of 5) other farmers. The most widely knowledge and 
practices transferred by model farmers to followers were fattening management (62%), improved 
forage production (59%), improved poultry nutrition and feeding (45%) and hay making (46%). 
About 37% of the model farmers also transferred their knowledge on dairy animal nutrition and 
management to other farmers. Baling, supplementation with urea molasses lick blocks and silage 
making were the least transferred and adopted practices among follower farmers (see Figure 6
below).
Some model 
farmers were 
able to share
their
Figure 6: Percentage of households by type of knowledge/practices transferred and 
adoption of these practices
PERCENTAGE OF RESPONDENT FARMERS THAT TRANSFERRED
IMPROVED PRACTICE TO FOLLOWER FARMERS BY TYPE OF
PRACTICE
knowledge to 
hundreds of 
farmers that
came from
P R O V I D E F R E E C H O I C E A C C E S S T O C L E A N …
F E E D I N G F A C T O R Y P R O D U C E D F O R M U L A T E D …
S I L A G E M A K I N G 
H A Y M A K I N G
S U P P L E M E N T A T I O N W I T H U R E A M O L A S S E S …
14%
15%
3%
4%
46%
different 
areas. Model 
farmers also 
said that an 
average of 4 
follower 
farmers 
(median of 2) 
adopted the
U R E A T R E A T M E N T O F C R O P R E S I D U E S
S U P P L E M E N T A T I O N O F C R O P R E S I D U E S W I T H …
B A L I N G 
F O R A G E C H O P P I N G
F E E D I N G T O T A L M I X E D R A T I O N ( I . E . A L L …
R E C O R D K E E P I N G O F A N Y K I N D ( R E L A T E D T O …
R O T A T I O N A L G R A Z I N G
D R I N K I N G T R O U G H S T O P R O V I D E C L E A N …
F E E D I N G T R O U G H S T O F E E D A N I M A L S 
C U T A N D C A R R Y
D A I R Y A N I M A L N U T R I T I O N A N D M A N A G E M E N T
I M P R O V E D F O R A G E P R O D U C T I O N 
I M P R O V E D P O U L T R Y N U T R I T I O N A N D F E E D I N G
I M P R O V E D F A T T E N I N G M A N A G E M E N T
10%
6%
10%
13%
12%
17%
23%
26%
35%
35%
37%
45%
59%
62%
improved
practices 
transferred
to them.
F E E D I N G C O N C E N T R A T E F E E D T O L I V E S T O C K 21%
0% 10% 20% 30% 40% 50% 60% 70%
About 59% of model farmers indicated that their follower farmers adopted improved fattening 
management, followed by those that adopted improved forage production (54%). Over 43% also 
stated that the knowledge on poultry nutrition and feeding, and hay making that they transferred 
were adopted by other farmers. About a quarter of model farmers indicated that their followers 
adopted one to two improved practices. Close to 40% of follower farmers are said to have adopted 
three to four practices. About a third of follower farmers adopted 5 to 9 practices. In sum, the findings
suggest that the vast majority of follower farmers adopted two or more improved practices.
6 Some farmers that transferred their knowledge to over 50 other farmers were excluded from the determination of 
the mean they are exceptional and not reflective of most farmers, thereby distorting interpretation of the data if not 
removed.
.
36
The evaluation team also consulted around 73 follower farmers. All of them indicated that they 
received some knowledge transfer on improved practices from model farmers. As shown in Table 22
below, 89% of the follower farmers received transfer of knowledge on dairy animal nutrition and 
management during the past four years. Over half of follower farmers received transfer of knowledge 
on improved forage production, fattening management, and poultry nutrition and feeding. The 
findings could provide some confirmation that model farmers are actually transferring their 
knowledge to follower farmers.
Table 22: Percentage of follower farmers that received transfer of knowledge from model farmers over the past four years
Amhara Oromia SNNPR Tigray Total
N N N N N %
Have you received any transfer of knowledge 
on dairy animal nutrition and management?
Yes 13 14 20 18 65 89%
No 4 1 1 2 8 11%
Have you received any transfer of knowledge 
on improved fattening management?
Yes 11 9 11 6 37 51%
No 6 6 10 14 36 49%
Have you received any transfer of knowledge 
on improved poultry nutrition and feeding?
Yes 17 2 14 5 38 52%
No 0 13 7 15 35 48%
Have you received any transfer of knowledge 
on improved forage production?
Yes 17 4 11 8 40 55%
No 0 11 10 12 33 45%
As also depicted in Figure 7 below, most follower farmers said that they use at least five improved 
practices, including hay making, supplementation of crop residue with green fodder, feeding and 
drinking troughs, and cut and carry method.
Figure 7: Percentage of follower farmers and their use of different improved practices
37
PPERCENTAGE OF FOLLOWER FARMERS AND THEIR 
USE OF DIFFERENT IMPROVED PRACTICES
P R O V I D E F R E E C H O I C E A C C E S S T O C L E A N …
F E E D F A C T O R Y P R O D U C E D / F O R M U L A T E D …
M A K E S I L A G E 
M A K E H A Y
S U P P L E M E N T A T I O N W I T H U R E A M O L A S S E S …
U R E A T R E A T M E N T O F C R O P R E S I D U E S
S U P P L E M E N T A T I O N O F C R O P R E S I D U E S …
B A L I N G 
F O R A G E C H O P P I N G
F E E D I N G A T O T A L M I X E D R A T I O N
R E C O R D K E E P I N G O F A N Y K I N D 
R O T A T I O N A L G R A Z I N G ?
D R I N K I N G T R O U G H S T O P R O V I D E C L E A N …
F E E D I N G T R O U G H S T O F E E D A N I M A L S 
C U T A N D C A R R Y
19%
34%
56%
56%
56%
48%
56%
47%
75%
71%
99%
92%
95%
88%
86%
0% 20% 40% 60% 80% 100% 120%
2.1.5.3. Knowledge on animal nutrition
Minerals supplements are very critical for growing and lactating ruminants and poultry. For instance, 
limestone is known as calcium source which is an important mineral for lactating dairy cows and 
laying hens specially to reduce milk fever and soft eggshells respectively. In this regard, farmers were 
asked to identify which feeds/minerals could reduce the occurrence of milk fever in dairy cows 
(hypocalcemia; cow goes down soon after calving) and the occurrence of soft egg shells. Over 43% of 
them indicated that they don’t know which feeds can reduce milk fever in dairy cows and the 
occurrence of soft egg shells in poultry, compared to 64% at baseline.
Close to 31% said that salt reduces the occurrence of milk fever and soft egg shells, while this was 
the case for 14% of households at baseline. The proportion of households who knows that limestone 
can reduce the occurrence of milk fever or soft eggshells increased from 4.7% at baseline to 21.1%. 
In sum, relatively larger proportion of households that participated in the endline survey know the 
different types of feeds that reduce milk fever and soft eggshells compared to baseline. Although 
there were some changes, households in Oromia and Amhara are more likely to lack knowledge about 
the type of feed that reduces the incidence of milk fever and soft egg shells than those in Tigray and 
SNNP. This was also mainly the case at baseline.
The proportion of households that were not able to compare the impact of feeding one kg of noug 
cake on milk production with different kgs of cottonseed cake declined from 79% at baseline to 55% 
at endline. This suggests that target households have acquired some knowledge about the nutritional 
content or values of noug cake and cottonseed cake over the past five years.
Plant growth stage at harvest is one of the most important factors that affect the quality of forages. 
Usually forages harvested at early growth stage have better nutritive value for livestock and can give
38
more milk production than the forage harvested at later stage of growth. In this regard, households 
were asked about their knowledge on using these kinds of feeds and their impact on milk yield. The 
proportion of households that said that grass cut 30 days after the last cutting of the same plot can 
support more daily milk production when fed to a crossbred dairy cows as grass cut 60 days after the 
last cutting increased from 30.1 to 43.6%. On the other hand, on average the proportion of 
respondents who did not have any knowledge on the effect of grass growth stage at harvest on daily 
milk production of crossbred dairy cows reduced from 41.6% at baseline to 7.6% at endline, though 
they don’t seem to have the right understanding.
Both energy and protein sources are essential for improving dairy milk productivity. Generally, 
grasses like Elephant and Rhodes are sources of energy while vetch is a protein source. A dairy cow 
fed in combination of elephant grass with vetch forage can produce more milk per day than when fed 
elephant grass alone. The proportion of households with such understanding showed modest 
increment compared to the baseline. Households who did not have any knowledge on feeding Rhodes 
grass or vetch or teff straw in combination with elephant grass also declined from 62.5% at baseline 
to 31.6% at endline.
2.1.5.4. Business knowledge
Business knowledge is important for farmers to make viable investment decisions and engage in 
sustainable livelihood activities. For an enterprise choice, for example, farmers should compare 
market price minus input cost for each of the different enterprises they might consider investing and 
choose the largest of all. If they have the choice of investing their money in raising broiler chickens 
or fattening sheep for income, about 45% of households would decide by comparing market price 
minus input cost for each and choose the larger of the two, compared to 18% at baseline. This shows 
that farmers have gained some business knowledge that they need in making investment choices. 
The proportion of respondents that would not know how to decide declined from 32.9% at baseline 
to 4.2% at endline.
Target households were provided training on record keeping. Over 17% of households indicated that 
they adopted record keeping lessons learned through the project. Households were also asked about 
the kind of information that they need to collect to know their profit/loss if they plant a piece of their 
farm land to forage. The percentage of households that know some of the answers increased 
compared to baseline. For example, the percentage of households that said that they need to collect 
information about amount of money spent on input and labor to raise the forage and marketing 
expense and price received per bale of forage sold increased from 14% and 10% at baseline to 40% 
and 16% at endline, respectively. Households that don’t know what information to collect also 
declined from 51% to 18%. However, it important to note that the proportion of households that 
know the full answer to the question declined from 10% at baseline to 2% at endline. This suggests 
that though business knowledge of households showed some improvement, they still have partial 
understanding.
39
2.1.6. Manufactured Feed Development
In this section of the report, the project achievements and contributions toward the supply of and 
demand for formulated feed is presented. As depicted in Figure 8 below, the project strengthened the 
whole formulated feed value chain. On the input supply side, the project supported local feed mill 
manufacturers and ingredient suppliers and linked them with the union-based feed manufacturing 
enterprises. At the center, the project made significant investment to establish union-based feed 
manufacturing enterprises. On the demand side, the project created awareness and demand for 
formulated feed among livestock and poultry enterprises, smallholder farmers and dealers. In 
addition, the project created linkages between the feed manufacturing enterprises and buyers of the 
feed products. FEED II also strengthened regulatory capacity in the animal feed sector.
Figure 8: Formulated Feed Value Chain
In this section, the contribution of the project in each of the three components of the formulated feed 
value chain is discussed in some detail.
2.1.6.1. FEED II Contribution to Development of Feed Input/Ingredient
Supply Chain and System Strengthening
FEED II project has been strengthening the feed ingredient supply chain by providing cash grants and 
technical support to feed input/ingredient trading/producing enterprises. Project monitoring 
reports showed that FEED II disbursed USD 716,559a in the form of in-kind and cash grants to feed 
ingredient supply chain actors. The key actors in the feed ingredient supply chain supported by the 
project included feed ingredient producers and traders, feed processing and handling equipment
40
manufacturers, and animal feed associations and regulators. The project activities and contributions 
as related to each of these key actors is presented below.
Feed Ingredient Suppliers
Project progress reports showed that thirty feed ingredient supply enterprises and other input and 
service providers were supported with technical assistance. The project also provided cash grants to 
feed ingredient supply chain actors to foster supply. Market linkages were also established between 
private sector feed ingredient suppliers and feed manufacturing enterprises established under 
unions with the project support. Feed ingredient supplying enterprises that were consulted for the 
final evaluation also recognize the role played by FEED to foster linkages to suppliers in collaboration 
with relevant institutions. All the 22 feed ingredient suppliers consulted by the evaluation team 
confirmed that market linkages have been established mainly with concentrate feed manufacturers 
that are said to be their major customers. All but one of them said that there is adequate demand for 
their feed/forage ingredients.
Feed manufacturing enterprises on their part also acknowledged the contribution of the project in 
facilitating linkages with their ingredient suppliers while also mentioning the fact that most of them 
buy the ingredients through a bid process for purchases with a value above a certain amount. Union￾based feed manufacturing enterprises largely procure feed ingredients from the open market. The 
main sources of ingredient supply are other unions, primary cooperatives and other private suppliers 
(flour and oil mills, premix suppliers largely located in Addis, etc.).
Data gathered from government officials also showed growing investment in agro-processing sector 
that could improve the supply of feed ingredients. The growing trend was largely associated with the 
implementation of the FEED project. Some unions are also investing in flour mills to produce some 
of the inputs needed by their feed manufacturing enterprises. A regional agriculture bureau official 
in this regard said the following:
‘Investment in food processing agro-industries such as flour mills has increased availability of raw 
materials for compound feed manufacturing factories. For instance, Gozamin Farmers Union has 
integrated flour mills with feed manufacturing factory such that by products like wheat bran and 
middling are directly made available for the feed factory, a kind of backward integration. Therefore, 
as availability of feed ingredients improves, price setting can be revised to a level that farmers can 
afford to buy and use.’
However, inadequate and/or timely supply feed ingredient is still a challenge to many feed 
manufacturing enterprises, which sometimes causes disruptions in operations resulting in 
occasional failure to meeting demand at sales points. Seasonal variabilities in the amount, quality and 
price of ingredients are also prevalent. Availability of sufficient resources (cash and storage facilities) 
for bulk purchase at times of harvest when prices are lowest is also a common problem shared by 
many plants. Some unions have mitigated the challenge by investing in ingredient producing 
enterprises, such as those with flour mills. Others are also in the process of establishing flour factories 
to ensure continued supply of inputs. ‘The supply problem is very critical, it even made us
41
temporarily stop production. To resolve this problem, we are about to establish a flour factory and 
we believe that this will improve our production as it will give us sustainable input supply,’ a manager 
of one of the project-supported feed manufacturing enterprise said.
Union procurement policies are also restrictive and lengthy limiting capacity to respond to 
immediate input needs or opportunities of limited-time price reductions and when supplies are 
limited. Suppliers in many instances are not willing to participate in auctions/bid required by the 
purchasing regulation of unions that oblige them to bid out any items worth more than a certain 
amount and the lengthy purchase and payment procedures. Unions are also being challenged to find 
suppliers that can issue receipts/invoices for payments they make to buy inputs. But suppliers prefer 
to sell their product to other private retailers who usually don’t request receipt for the money they 
paid or with less stringent procedures. ‘There is input supply problem. The rules and regulation of 
union influence the purchase of input. There is no sufficient amount of input in the market to 
purchase through bid and we have shortage of purchaser to conduct continuous purchase,’ a feed 
manufacturing enterprise manager said. Weak coordination between the union board and feed 
enterprise manager was also a factor in delaying acquisition of inputs.
‘One of the main challenges that hinders the activity of feed enterprise is low input supply. This is 
mainly related to financial rules and regulation. The Performa based purchase system limits the 
amount of money to 50,000 ETB to purchase input and this causes low supply, repeated requisition 
and travel. The trader may refuse to provide Performa and sometimes change of price may disqualify 
the purchase order. (2) There is poor coordination between manager and boards which cause delay 
for approval as some of the traders give 2-3 day time limit for prices of input. (3) Sometimes there is 
shortage of truck to transport inputs as the vehicle serves both for the feed enterprise and union.’
Continued price increases, high dependence on a few suppliers for some ingredients limit the unions’ 
capacity to negotiate for better prices and quality. There is a relatively wider supply base for 
ingredients like grains that can, in some cases, be procured from member cooperatives sometimes 
with special concessions. Strengthening the capacity of primary cooperatives to produce ingredients 
that target the feed plants should be encouraged. Key challenges in relation to imported ingredients 
and equipment include limitations in access to hard currency, tax burden including VAT that 
ultimately reflect on the prices of such ingredients at the plant level.
Contractual agreements for ingredient supplies are also very limited to some unions and few inputs. 
Primary cooperatives that work under the Merkebe union, for example, have a contractual agreement 
with the feed plant to supply inputs. But the relation with other suppliers is said to be demand driven. 
FEED has worked to support forward contracting, but contextual factors such as continued price 
increases have limited the union’s success in contracting. A feed manufacturing enterprise manager 
in this regard said ‘The relation between suppliers and feed manufacturing is not bound with 
contract. We don’t have any specific contractual agreement with the factory to supply inputs.’
Feed Mill Equipment Manufacturers
FEED II project also supported five manufacturers with technical and financial assistance to design 
and manufacture feed manufacturing equipment to ultimately ease the supply of machinery even 
beyond the union feed plants. Supporting these local equipment manufacturers was considered
42
appropriate to build local capacity to manufacture, maintain and repair feed manufacturing 
equipment. The project worked closely with the grantees and the phased and performance-based 
grant release approach was seen as effective by the grantees. The approach and support provided by 
the project enabled grantees to manufacture prototype equipment as per requested specifications. 
In addition to building their capacity on manufacturing of feed mill (some received this support in 
FEED I), the project also helped the manufacturers to get customers for the products by linking them 
with the feed manufacturing enterprises and creating demand for formulated feed which in turn 
increased demand for feed mill. One of the equipment manufacturers in this regard said the 
following:
‘I have been involved with ACDI-VOCA starting from FEED I. I received capacity development support 
in FEED I. We have also got financial support and business opportunity for us to create new one and 
make modification on the existing feed mill. Our major customers are farmer unions at different 
regions, government agriculture research institutes, universities and sometimes individuals. I usually 
get orders through referral and ACDI-VOCA. …The demand for our feed mill product is growing, since 
the farmers are aware of the benefit of manufactured feed. We have now good capacity for producing 
feed mill and we are updating ourselves with the help of internet and the FEED project.’
Feed mill manufacturers learned new ways of production through project support. One of the 
grantees also said that his firm would not have thought of producing feed milling equipment if it 
wasn’t for the FEED project. Apart from technical support, the project linked the feed mill 
manufacturers with unions. While acknowledging increased demand for feed milling equipment over 
the past few years, the manufacturers were not able to utilize their production capacity well. ‘The 
knowledge transfer is good. But FEED II project’s support on market linkage was limited compared 
to its ability to create such linkages. It was only two feed equipment that my company has managed 
to sell ACDI/VOCA per year in comparison to our 60 to 70 equipment production capacity,’ an owner 
of one of the feed equipment manufacturers said. Data gathered from one of the equipment producers 
showed increased client base over the years. Apart from selling to FEED project supported unions, 
the feed milling equipment manufacturer has been selling to meat and dairy institute, universities, 
TVT colleges, capital goods supplier, agricultural research institutes, Ethiopian feed industry, 
Ethiopian poultry association, Chamber of Commerce, and other investors and individuals. Although 
attribution is difficult, revenue and profit of one of the equipment manufacturers increased from Birr
899,553 and 151,061 in 2015 to 7,184,047 and 599,092 in 2018, respectively. According to the feed 
equipment manufacturers, their production and income could have been much better if importation 
of necessary metals and access to finance to sell the equipment on credit was possible.
Capacity Building and System Strengthening
As part of its input supply strengthening initiative, the project trained mechanics to repair feed 
manufacturing equipment. This was considered important by feed enterprises and other
stakeholders to ensure continued operation of feed producing machineries.
FEED II also provided technical and financial assistance to the Ethiopian Animal Feed Industry 
Association. The technical assistance involved international and local experience sharing visits, and 
training on feed formulation, feed manufacturing process, basics of animal nutrition and others. It 
also received grants for different activities that included publication of bulletins and consultation 
with stakeholders. The association’s capacity has also been studied and enhanced through the project 
support, which enabled it to engage in feed policy issues, conduct studies on Ethiopian tax system on 
animal feed, and produce operation manuals on animal feed manufacturing processes. The 25 feed 
manufacturing enterprises established by the FEED project are also members of the association. The
43
project also supported the Ethiopian Meat and Dairy Technology Institute to collect, analyze and 
distribute transaction data from feed and livestock industries.
FEED II has also strengthened feed quality control and standards enforcement mechanisms by 
providing training and supporting staff of the Veterinary Drug and Animal Feed Administration and 
Control Authority to participate in the international experience sharing visits and conferences. The 
authority worked closely with the project as a partner. Officials consulted for the final evaluation also 
indicated that the project support helped them develop necessary skills and knowledge that they 
needed to formulate relevant directives that govern the animal feed sector. The project support also 
came at the right time as the authority was a recent establishment that needed all rounded support 
to develop capacity needed to control and regulate quality and safety of feed manufactured by this 
union. Some encouraging improvements were also observed during the project period. Quality has 
improved. National feed standards covering an array of different feeds and various regulations 
guiding feed quality and safety inspection are now in place to serve as the basis of the regulatory 
framework to help ensure feed quality and safety. The feed regulatory body has started to inspect the 
feed at production, storage and distribution stages to help uphold quality and safety (Alemu and 
Seyoum, 2019).
2.1.6.2. FEED II Project’s Contribution to Production/Supply of
Manufactured Feed
i. Trends in production of manufactured feed
The Ethiopian government has identified the transformation of the livestock sector as a source of 
economic growth in both the first and second Growth and transformation Plans (GTP I and II; 2010-
2015 and 2016-2020, respectively) through intensification and commercialization of production. 
The growth of the animal feed sector, specifically commercial feed production, will be key to bring 
about this transformation.
Survey of the commercial feed sub-sector conducted during the period August 2016 to March 2017 
(Seyoum, 2017; commissioned by FEED II), reported that the distribution of feed processing plants 
in the country is skewed with most (68%) of the private feed processing plants located in Oromia 
and Addis Ababa where there is good business/demand for formulated feed. The Amhara and SNNP 
regions each account for 13% while Tigray accounted for 6% (Seyoum, 2017; FAO 2018). The 
cooperative union-based feed manufacturing plants are more evenly distributed across SNNP, 
Oromia, Amhara, and Tigray in locations deeper in the countryside for broader access to farmers and 
some are, as a result, located in places private feed processing plants may not want to operate.
Production of compound feed in Ethiopia has almost doubled during the last five years as a result of 
the relatively fast growth in the number of feed mills and an increase in the level of production by 
existing ones. The level of technology use has also shown advances. Nevertheless, the quantity being 
produced currently does not match the country’s huge livestock population.
The types of feeds produced in the country have become more diverse. New additions are, for 
example, pig feeds, aquaculture feeds, etc. The rations are also becoming more diverse in terms of 
the form. Feed was produced in the form of mash earlier but other forms like pellets, crumbs etc. are 
now available. The production of complete feed was the norm. Recently, concentrates of different
44
formulations for mixing with grains to come up with complete feed are coming. New trends in the 
distribution mechanism are also emerging. Delivery at farm gate, packaging of different sizes to meet 
customer requirements, etc. are emerging. Laboratory service (public and private) is becoming more 
readily available.
ii. Production of formulated feed by FEED II supported feed manufacturing 
enterprises
FEED II project contributed to the changes described above in 
many ways. One of these ways included establishment of feed 
manufacturing enterprises. FEED II project supported the 
establishment of 12 union-based feed manufacturing 
enterprises and provided technical assistance to these newly 
established enterprises and those that were formed under 
FEED I project in business plan development and marketing,
and accounting and reporting. Project progress reports showed that FEED II disbursed USD
2,164,378 in-kind and cash grants to unions to establish feed manufacturing enterprises. It also 
provided training to union staff to establish and operate feed manufacturing enterprises. Training of 
staff on feed mill management and feed manufacturing, simple repairs, and upkeep of mill facilities 
helped to reduce down times of the factories. FEED II also facilitated domestic and international 
learning visits to hundreds of feed manufacturing stakeholders. the project also awarded grants to 
the 12 feed manufacturing enterprises established under FEED I to expand their businesses. Training 
on Peachtree accounting was also offered to facilitate financial management.
FEED’s choice of establishing feed plants for supplying concentrate feed as an intervention was 
considered to be appropriate and innovative by stakeholders from the regional level down to the 
beneficiaries surveyed from the perspective of 1) the key role concentrate feed supply plays in 
market orientation/intensification of smallholder livestock production, and 2) the increasingly 
prevalent shortage of land for extensive management of livestock and protection of the natural 
resource. This is also the direction of livestock development in the national livestock masterplan.
Overall, all but one plant assessed for the final evaluation were operating 
well (were functional) despite differences in the level of performance as a 
result of challenges specific to their circumstances. Production and gross 
sales/revenue and profitability of the union feed plants have been increasing 
steadily since their establishment commensurate with the increase in 
awareness of the benefits by smallholder farmers. Twenty-five union-based 
feed manufacturing enterprises produced and sold 11,286.73 metric tons of 
concentrate feed from April-September 2018. Of the total, 51.63%, 32.12%, 
and 13.96% were dairy, fattening and poultry concentrates, respectively. 
(FEED II Year V Semi-Annual report).
All union-based feed plants surveyed asserted to have an active business plan, mission statement, 
annual action plan, prepare annual budgets, produce financial statements and have a clear and 
transparent governance system. The evaluation team consulted 12 feed manufacturing enterprises 
that have been supported by the FEED II project (but three of them were established under FEED I).
45
Average daily production capacity of these enterprises was around 105 quintals in the period July
2016 to June 2017 and July 2017 to June 2018. But their actual production rate collectively, was 46 
quintals, which represented around 44% capacity utilization based on an eight hour per day, five 
days a week basis.
Capacity utilization among FEED II supported plants increased 
significantly from 36 quintals per day in 2017 to 56 quintals per 
day in 2018. The average annual production of the plants 
increased from 4,440 quintals in 2017 to 6,625 in 2018. This 
represents an increment of 67%. Each plant on average produced 
and sold around 5,500 quintals per year for the two years. A lower 
(34.9%) increment in production and a 39.27% increment in sales
was reported by the project for a 6-month period (April 2018-September 2018) relative to an earlier
6-month period of Oct 2017 – March 2018 (FEED II Year V Semi-Annual report).
Despite the 44% of capacity utilization, the annual increment of
67% indicates a significant increment through time. This shows 
further increments would be expected given expected change in 
awareness of the benefits of compound feed, expansion of 
distribution channels, and improvements in storage and 
transport. Ingredient supply and power interruptions are also 
mentioned by some plants as limitations in this regard. Internal
management problems/gaps that hinder the plants’ from producing at full capacity like the presence
of an adequate number of skilled personnel, the stringent purchasing procedures, and weak 
leadership and management are also mentioned.
The union-based feed plants that are slightly lower in number (29 including 25 established by FEED) 
represented about 16% relative to the privately-owned plants (32) that contributed 84% of annual 
compound feed production (Seyoum et al. 2017; FAO, 2018). This seemingly small contribution in 
view of the small difference in the number of plants is due to the larger average capacity of the private 
plants (especially the very big capacity of one private plant inflating the average) and the larger sales
due to their locations near places of higher demand warranting larger production levels. The 
potential, i.e. the production and returns on investment are, however, substantial for the union plants 
that are currently serving the rural markets. These plants also go beyond the profit objective to 
impact livelihoods, food security, and nutrition among their members, the rural households. Prices 
charged by union-based feed enterprises is a bit lower than other competitors. Unions also put in 
place mechanisms to ensure that the prices charged by their distributors are not high. ‘The feed 
manufacturing enterprise supplies feed with more moderate price than traders,’ a
said.
The contribution of the project-supported feed enterprises is well recognized by stakeholders 
consulted for the final evaluation. ‘By organizing, empowering and providing grants to the unions 
and establishing feed manufacturing units in the unions the project has increased the production, 
process and supply of livestock and poultry feed,’ a regional agriculture bureau official said. 
Concentrate feed manufacturing factories installed by the project are contributing to the supply of
46
different rations for specialized animal farms such as poultry, fattening and dairy. The demand for 
these products has also increased because animals that are fed with such better quality feeds have 
become more productive thus motivating farmers to buy more.
There is also agreement that the union-based feed plants are playing a catalytic role in the expansion 
of concentrate feed manufacturing in the country. ‘Motivated by high demand for manufactured feed, 
private entrepreneurs are being involved in feed manufacturing business,’ a regional agriculture 
bureau official said. Many union-based plants seem to have recognized the significant opportunity 
for growth, given the growing awareness resulting in substantial potential demand in their 
distribution areas, untapped sales channels to meet existing demand among member PCs and 
beyond. Consequently, some have taken or are planning to take additional measures in further 
improving their production and sales capacities like investing in additional equipment and input 
procurement and distribution networks, dedicating vehicles for input supply and formulated feed 
delivery by some.
Despite the encouraging efforts, faster growth at the various stages of production and distribution 
seems to be constrained by 1) limited marketing and point-of-sale education capacity, 2) access to 
credit which constrains investments in storage, procurement, transportation, and sales/marketing, 
and 3) weak business/financial and operational management capacity.
iii. Return on Investment and Gross Margin of Feed Manufacturing Enterprises
Apart from meeting social goals, union-based feed manufacturing enterprises established with the 
project support also have financial objectives. They would like to make a profit from the enterprises, 
which is also vital for the sustainability of the initiatives. The evaluation team also tried to look at the 
profitability of the feed enterprises by determining their return on investment (RoI)
7 and gross 
margin (GM)
8. ROI measures the gain or loss generated on an investment relative to the amount of 
money invested, while gross margin represents the amount of sales revenue that an organization 
retains after incurring the direct costs associated with producing the goods and services it sells.
The determination of RoI and GM was, however, challenged by the fact that many unions did not 
produce separate financial statements for their feed manufacturing operation. They produced 
consolidated financial statements covering all their business activities including feed manufacturing. 
Hence, collection of data pertaining to revenue, costs and expenses of feed manufacturing enterprises 
was not possible as doing so would require processing their transactions that went beyond the scope 
of the evaluation. It was even much harder to get reliable financial information from 
livestock/poultry/forage enterprises that were supported by the project. As a result, the evaluation 
team utilized financial data gathered by the project from each of the 12 FEED II supported unions 
and livestock/poultry enterprises supported by the project. This data was largely gathered on 
monthly basis that was also validated through phone calls, and physical visits. The financial data 
maintained by the project staff about grantees was also verified by data gathered by the evaluation
team from a sample of the grantees. The evaluation team, for example, gathered information about
7
RoI is determined by dividing return by total investment and then multiplying the result by 100
8 Gross margin is the difference between sales revenue minus its cost of goods sold (COGS). Gross margin is often
calculated in percentage terms by dividing the gross margin number by net salesrevenue.
47
annual feed production and sales in units, average selling prices per quintal of feed, and other 
relevant information. This information was then used to validate the financial records maintained by 
ACDI/VOCA about the feed manufacturing operations of the 12 unions. The verification process 
showed that the financial information maintained by ACDI/VOCA was very close to the data gathered 
by the evaluation team. In light of this, the return on investment and gross margin figures were 
determined based on the financial information provided by the project staff.
As far as feed manufacturing enterprises are concerned, as shown in Table 23 (see Annex 7 for more 
details), the FEED II project invested Birr 15,605,565.45 to buy equipment and support construction 
of shed, which represented 82% of the total investment of Birr 19,025,565.45 for establishing the 12 
feed manufacturing enterprises. The remaining was contributed by the unions. Out of 12 feed 
enterprises, four of them started production in 2016 and managed to earn income of Birr
8,259,167.25 in 2016 while incurring Birr 7,524,408.86 resulting in a total return of Birr 734,758.39. 
Considering a total investment of Birr 6,154,478.30 in the four enterprises, their RoI and gross 
margin in 2016 was 12% and 9% respectively. This can be considered an encouraging start given the 
fact that they utilized less than 54% of their production capacity. Indeed, most businesses often 
struggle to survive let alone to make a profit during the first few years of operation due to 
inexperience, lack of established client base, weak financial capacity, and so on. The feed enterprises 
managed to earn some profit in their first year of operation due to several factors. The integrated 
engagement of the FEED II project was one of those factors. Apart for supporting the establishment 
of the enterprises, the project linked them to ingredient suppliers and buyers of their products. It 
also worked significantly to create awareness among smallholder farmers about formulated feed. As 
a result, demand for manufactured feed has been growing. Drought that occurred in 2016 was the 
other notable factor that contributed to increased demand for manufactured feed which led to 
profitability of feed enterprises in the first year of operation.
In 2017, all 12 new (FEED II) feed manufacturing enterprises were operational. They generated 
revenue of over Birr 27 million and managed to get 10% RoI and gross margin of 7%. This financial 
performance can also be considered very encouraging given that most of the feed enterprises started 
operation during 2017. Capacity utilization among a sample of enterprises visited by the evaluation 
team was below 40% during the period. Financial data was only available for five enterprises in 2018 
that generated revenue of over 13.6 million, obtained RoI and gross margin of 23.4% and 13.3%, 
respectively. If data on all the enterprises was available and assuming similar growth pattern, one 
would expect RoI and gross margin higher than the previous years.
The RoI is expected to grow significantly in the coming years given the growing demand for 
formulated feed in the country, increased capacity utilization and economies of scale that comes 
along with larger production, and other factors. Indeed, some of the unions are expanding their 
manufacturing capacity which will also increase their return. Hence, the payback period is expected 
to be much shorter. Even at 12% RoI in the initial year of operation, the project can pay back the total 
investment in around 8 years. In sum, feed manufacturing investment was a viable initiative that 
generates profits for unions while also increasing supply and access to such feeds to the community.
Table 23: Return on Investment in Feed Manufacturing Enterprises Figure 9: RoI and Gross Margin Growth Trends in
Feed Manufacturing Enterprises
48
2016 2017 2018
25.00%
23%
20.00%
15.00%
13.30%
11.90%
10.00% 10%
8.90%
7%
5.00%
0.00%
FY 2016 FY 2017 FY 2018
ROI (%) GM (%)
FCU9 Total 
investment
ROI 
(%)
GM 
(%)
ROI 
(%)
GM 
(%)
ROI 
(%)
GM 
(%)
Enterprise 1 1,549,288 5.8 2.6 6.3 4.5 9.6 4.5
Enterprise 2 1,502,955 26.7 12.4 3.0 3.0 59.1 26.7
Enterprise 3 1,549,288 13.8 10.7 -9.3 -
37.9
Enterprise 4 1,765,140 25.6 8.1
Enterprise 5 1,552,947 4.9 5.5 13.2 8.7
Enterprise 6 1,556,002 11.1 14.6 11.3 9.4
Enterprise 7 1,725,878 10.4 6.9
Enterprise 8 1,562,529 17.5 10.8 31.2 14.8
Enterprise 9 1,562,529 22.0 11.5
Enterprise 10 1,617,791 8.1 6.5 7.3 6.5
Enterprise 11 1,531,963 4.2 4.2
Enterprise 12 1,549,254 5.8 3.6
Total 19,025,565 11.9 8.9 10.0 7.0 23.4 13.3
Source: ACDI/VOC FEED II Project
2.1.6.3. FEED II Project’s Contributionto IncreasedAccess, Demand and Use
of Formulated Feed, and Impact on Users/Buyers
i. Impact of FEED II on improving farmers’ access to formulated feed
The emergence of the new union-based feed plants established with the support of the FEED project 
in recent years has made feed more accessible to smallholder farmers far from the service areas of 
private feed plants. ‘The project has brought a significant impact especially on the livelihood of some 
model farmers, since it helped them to obtain manufactured feed at their vicinities. The availability 
of the manufactured feed has improved the productivity of livestock and this in turn has changed the 
livelihood of some beneficiary farmers,’ a regional agriculture official said.
Hence, apart from supporting the production of concentrate feed at farmer owned unions, the project 
increased access to formulated feed to rural communities by strengthening or creating supply 
networks. Project supported union-based feed manufacturing enterprises often sell their produce to 
agro-dealers, primary cooperatives, individual farmers, and livestock (dairy, fattening and poultry) 
enterprises.
Most respondents said that the supply of formulated feed in their kebele is increasing. As shown in 
Table 24 below, close to 44% of households have access to manufactured feed within a 16 Km radius 
to a source of factory formulated feeds compared to only 18 percent at baseline. Only less than 13% 
of respondents do not know where the factory or formulated and mixed feed distribution center is,
compared with 54% at baseline. This is a substantial improvement. There are now 253 feed sales
9 The evaluation team chose to not to identify enterprises by name as the financial data are confidential.
49
outlets (including the 25 feed enterprise unions’ factory outlets) with almost 2,918,914 agriculture￾dependent households within a 16km radius of a sales point.
Table 24: Percentage of Households by Distances to the Nearest Animal Feed Distribution Center and Region at Baseline and
Endline
Distance to the nearest
factory formulated
and mixed 
distribution center
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
<16 km 15.7 10.8a 20.9 69.6b 27.2 52.8c 9.6 41.5d 18.2 43.5
17-30 km 6.7 10.6a 1.4 6.3a 7.8 7.0a 7.1 32.1b 5.8 13.9
31km & more 6.7 44.2a 1.9 5.8b,c 3.9 4.3b 6.7 10.5c 4.9 16.3
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
FEED helped the union-based plants to promote the formulated feed to potential customers. The 
project also supported unions to build shops for distribution of manufactured feed. Many 
respondents have indicated that it is through the support/effort of FEED that many smallholder 
farmers have accessed formulated feed. Almost all the areas currently supplied by the union-based 
plants were either not aware of the use of formulated feed and/or did not have access as formulated 
feed from the private enterprises did not reach them. The supply from the union plants is the sole 
source for many areas.
Encouraged by the increased demand, regional governments are also taking measures to scale up 
formulated feed production by establishing additional plants, which could in turn increase 
availability and accessibility of manufactured feed to farmers and livestock and poultry enterprises. 
Here is what a regional agriculture bureau official said.
‘Following the project intervention and establishment of the six feed manufacturing units, farmers 
are becoming more aware of the concentrated feed, as a result the increase in the demand for the 
manufactured feed motivated other unions to see the opportunity of installing additional plants. The 
regional cooperative office has provided loan for the procurement and installation of five additional 
units. This shows our commitment to scale up the project. Currently, we are accepting additional 
requests from other unions for installation of processing plants, though we have deferred the process 
till we are able to create the needed demand for the manufactured feed, as the shortage of demand 
may make the unions to be discouraged by the business.’
ii. Impact of increased availability and accessibility of manufactured feed on downstream 
economic activity (e.g. by farmers, poultry raisers, milk processors and retailers)
Impact on Livestock/Poultry Productivity of Farmers
Increased availability and accessibility of concentrate feed contributed to increased livestock 
productivity of farmers. Many farmers that participated in FGDs said that the change in the farmer’s
50
feeding practices, and use of concentrated feed and other improved forages have brought significant 
improvement on the productivity their livestock. The survey findings also proved the same. As shown 
in Table 25 below, the maximum and minimum amount of milk collected per day per cow by 
households that fed concentrates to their dairy cow was 2.63 and 2.1 liters higher than those that did 
not feed this mixed feed. Compared to households that did not feed concentrates, households that fed 
concentrates to their dairy cow was able to increase their maximum and minimum daily milk 
production from all dairy cows per day by 52% and 75%, respectively, and the difference was 
statistically significant. Average number of months indigenous dairy cows produce milk was also 7% 
higher for those that fed concentrates to their dairy cows, while no statistically significant difference 
was found on cross-breed cows. Households that fed concentrates to dairy cows sold milk that was
33% higher than those that did not feed such feeds. Similarly, income earned from sale of livestock 
and livestock products was 19% higher for households that fed concentrates than those that did not. 
The amount of milk and butter sold and income generated from sale of these products was also 
significantly higher for households that fed concentrates than those that did not. In light of this, it can 
be concluded that the supply and accessibility of concentrate feed enabled by the FEED II project has 
contributed to increased smallholder farmers’ dairy productivity and income.
Table 25: Impact of feeding/not feeding concentrates on dairy cow productivity
Have you ever fed concentratesto your dairy 
cow?
No Yes
Mean % Mean % %
difference
Maximum amount of milk collected in one day from 
all of cows during the past 12 months (liters/day)
5.01a 62% 7.64b 38% 52%
Minimum amount of milk collected in one day from 
all of cows during the past 12 months (liters/day)
2.80a 62% 4.91b 38% 75%
Average number of months indigenous dairy cows
produce milk (i.e. the typical length of a single 
lactation)
7.00a 70% 7.48b 30% 7%
Average number of months crossbreed dairy cows
produce milk (i.e. the typical length of a single 
lactation)
9.31a 48% 9.12a 52% -2%
Number of liters of milk sold per month
(liters/month)during the last 12 months
174a 38% 231a 62% 33%
Total amount of income earned from sale of 
livestock/products
24,348a 63% 28,946b 38% 19%
Total annual sales collected from butter sale (birr) 1996a 61% 3310b 39% 66%
Amount(killogram)of butter sold in a year. 11.7a 62% 16.1b 38% 38%
Amount (Liters) of milk sold in a year. 210a 62% 737b 38% 252%
Total annual sales collected from milk sale (birr) 2731a 62% 11288b 38% 313%
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
It is also interesting to observe that smallholder farmers that reside close (less than 16 kms) to the 
nearest distribution point for formulated feed collected more milk than those that are live in areas 
that are far away. Average number of months indigenous and cross-breed dairy cows produce milk 
was also higher for respondents that live in less than 16 kms from the nearest distribution point for 
factory formulated. This might be because households that live close to distribution sites have the
51
incentive to buy and use concentrates. This is also evident from the fact that the majority of 
households that bought manufactured feed where those that live in less than 16 kms away from the 
distribution point (see Table 26).
Table 26: Distance from nearest distribution point for factory formulated and milk productivity
Distance from nearest distribution 
pointfor factory formulated and mixed
feed
<16 
km
16-30 
km
30 km 
or
more
No 
distribution 
point in the 
area
Maximum amount of milk collected in one day
from all of cows during the past 12 months
7.23a 5.14b 5.39b 4.71b
Minimum amount of milk collected in one day 
from all of cows during the past 12 months
4.52a 3.05b 2.92b 2.70b
Average number of months indigenous dairy
cows produce milk (i.e. the typical length of a 
single lactation)
7.36a 6.36b 7.19a,b 7.32a
Average number of months crossbreed dairy 
cows produce milk (i.e. the typical length of a
single lactation)
9.36a 8.75a,b 9.72a 8.86a,b
Do you feed your livestock concentrates
(factory formulated and mixed feed)?
No 317a 110b 221c 134b
Yes 336a 64b 53c 68b
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
Qualitative data analysis results also support quantitative findings. Increased supply and 
accessibility of concentrate feed encouraged many farmers to engage in livestock fattening and 
achieve better results. Compared to prices charged by local traders, the price of concentrate feed 
supplied by project supported feed manufacturing enterprises was also considered a moderate one. 
Here is how a government official described the impact:
‘The feed manufacturing enterprise supplies feed with more moderate price than traders. The 
distribution of manufactured feed through dealers increases availability of feeds to farmers as the 
feed enterprise is established far from our district (50km). It creates job opportunities for dealers 
and feed enterprise. The union, flour factory and feed enterprise benefitted through input and output 
supply. Farmers have also got better access to balanced feed to increase productivity of their 
animals.’
The distribution of manufactured feed through dealers and cooperatives increased accessibility of 
feed, which in turn encouraged use of such feed to improve productivity of livestock and poultry.
‘Oxen fattening and improved chicken breed rearing based on manufactured feed have raised
household income of farmers,’ a government official said. These activities in turn encouraged 
improved forage crop production and use of concentrate feed; feeding animals on a locally made feed 
troughs which also has facilitated on-farm feeding and a retreat from free grazing tradition. The 
project has brought a significant impact especially on the livelihood of some model farmers, since it 
helped them obtain manufactured feed in their vicinities. The availability of the manufactured feed
52
has improved the productivity of livestock and this in turn has changed the livelihood of many 
beneficiary model farmers,’ a said. Availability of concentrate feed 
might also have also promoted self-employment among unemployed youth. Some unemployed youth 
were said to be encouraged to engage in animal fattening, dairy and poultry activities after they 
observed the significant productivity gains achieved by other neighboring farmers with the use of 
concentrate feed. In connection with this, a government official said the following:
‘The project has brought a significant impact on farmers, livestock enterprises and input suppliers, 
especially smallholder farmers and enterprises that are engaged in livestock, dairy and poultry 
farming, as the availability of the feed has created additional opportunity to the farmers and 
jobless youngsters to consider the urban agriculture as business and means of self-employment.’
Farmers are said to buy manufactured feed mainly for fattening of sheep and cattle. This was mainly 
because the use of concentrate feed reduced fattening cycle time for farmers. Many farmers that 
participated in FGDs agreed that the fattening period was reduced to 3 to 4 months, which according 
to them used to take up to seven months before use of concentrate feed and other improved practices.
It is, however, important to note that the increase productivity was a function of a number of factors 
including those that relate to the project intervention. The training providing by the project on animal 
nutrition and feeding practices, ownership of cross-breed cows, and others were largely credited for 
a substantial increase in livestock and poultry productivity. ‘The improvement in livestock 
productivity is due to the change in their feeding practices, and the concentrated feed and other 
improved forages they used to feed their livestock. Besides, they have the hybrid cows and hens that 
have better productivity than the local ones,’ a woreda agriculture official said.
As much as farmers are motivated to use concentrate feed, they are also being challenged by the price 
of the feed. Most farmers and other stakeholders consulted for the evaluation agree that the price of 
concentrate feed is a bit high for farmers to buy. It was also considered less cost effective when such 
feed is used on local breeds. The case story below could illustrate the impact of the project
intervention on a direct beneficiary farmer.
Alemitu (pseudo name), married with nine children of different age group above school age; all did 
attend school; the eldest is now a in junior high school. Household income of 
Alemitu before 5 years was from sale of food crops like finger millet, maize and wheat. Now after 
adopting improved animal production technologies largely introduced by ACDI/VOCA, the household 
income was from cattle fattening, sale of poultry products (egg and cockerels), and sale of milk from a 
cross-bred cow. There is a clear indication that the livelihood of Alemitu has improved over the last five 
years. She is a relatively well to do farmer as can be judged by the iron sheet-roofed house, half the living 
room filled with a pile of sacs containing recently harvested grain.
According to Alemitu contribution of ACDI/VOCA to improved livestock production included trainings on 
improved feeding and management of livestock and poultry; construction of feeding trough using local 
construction material; improved housing and sanitary handling of animals; planting improved forage 
crops and feeding techniques. She acknowledged the continual training and advisory role of the 
development agents (DAs) that led to successes in her activities.
53
Out of the above practices that Alemitu received training, she implemented improved livestock feeding 
and management practices like dairy production from a cross-bred cow; fattening of local oxen using 
concentrate ration; egg production from improved chicken breeds.
According to Alemitu, Feed II project of ACDI/VOCA has been providing farm supplies, including, forage 
seed of improved forage crops free of charge, including Rhodes grass that she planted at the backyard on 
about 20 square meter plot; sesbania and elephant grass planted along the fence of her compound. She 
mentioned that before initiation of the ACDI/VOCA projects in the past five years, she has not heard of 
such improved forage crops. Based on what she learned from the project training she has been
harvesting these forages and feeding tethered stock on feeder trough built of local wood materials. 
Feeding cattle using such home-made feeder troughs helped her to minimize feed wastage and protect 
animals from parasites.
Alemitu said that she received initially free of charge compound (concentrate) feed of various specialized 
composition for chickens, milk cows and for fattening cattle produced by ACDI/VOCA-installed feed 
manufacturing factory. As an incentive for her active participation in improved livestock management 
practice, she received a gift of one quintal dairy concentrate feed. Having realized the benefit, she 
purchased concentrate feeds manufactured by the union’s feed factory until recently when the price 
became beyond her financial capacity and quit purchasing compound feed. Instead, she started feeding 
animals with home-made rations composed of raw materials most of which are home-produced and
some purchased from dealers.
As regards to benefits gotten from FEED II project, Alemitu has successfully put into practice 
ACDI/VOCA’s training on improved feeding and management practice. As a result, her household income 
has significantly increased. Her cross-bred cow is now on her second calf, and the milk produced, about 4 
liters per cow per day, is rented to school teachers. Similarly, the egg produced from improved chicken 
breed is sold to school teachers and to market in a nearby town. The income from sale of these products 
has been used to cover living expenses, clothing and school expenses for children, etc. From her stories 
and observations in and out of her house, Alemitu is well satisfied with what she gained from the
Project’s development activities, knowledge and material support. However, she did not conceal her
dissatisfaction as regards to the price of manufactured feed which nowadays could not afford to buy and 
so quitted using the same from the union’s product.
Regarding her future plan, Alemitu said that she has saved and allocated birr 3,000 to purchase egg type 
chicken pullets through a local dealer. She as well is sure to continue fattening oxen and dairy
production although the high price of manufactured feed has become prohibitive.
Regarding challenges to sustenance of using improved livestock production systems, Alemitu mentioned 
that the price of the manufactured animal feed has become unbearable that many farmers avoided buying 
the manufactured ration and instead began buying raw materials like wheat bran (frushka) and oilseed 
cake (fagulo) individually to produce home-based rations prepared in a traditional way. Surely, the price 
set by the union is so expensive to the extent that it is likely to affect the sustainability of using introduced 
technologies. Other problems include lack of capital to build improved cattle shed, poor transport, lack of 
electric power supply and absence of market linkage for livestock products.
By way of recommendation, Alemitu said that the union has to consider the current price of raw 
materials in setting the price of their manufactured feed. She also recommended that the union improve 
packaging of the products which has become substandard- use of poor quality sacs that are often torn, 
and the quantity lacks uniformity probably due to using faulty balance.
54
Regarding knowledge transfer, Alemitu has been appointed as the in 
the sub-kebele. She has been advising the cooperative members on improved poultry production, 
fattening and dairy. She estimates forty women have taken up her advice and taken up their preferred 
activities. There is positive challenge among members in being involved in improved chicken breeds 
rearing. It is common to see at least 5 to 10 chickens in every household in the sub-kebele.
According to Alemitu, the outstanding challenge in the acceptance and sustained use of improved 
practices is high price and lack of supply chains for manufactured feed, lack of market linkage for animal 
products.
Plate 1. Crop harvested and conserved at home Plate 2. Cross-bred cow of Alemitu
Plate 3. Feeder trough used by Alemitu
Impact on livestock and poultry enterprises
The project targeted to support the establishment of 30 feedlot, poultry and dairy enterprises. 
Progress reports showed that the project only managed to establish 15 dairy, fattening and poultry 
related enterprises. About half of these enterprises are private while the rest are cooperatives. Most 
of them are dairy enterprises. Different kinds of support were provided for these enterprises 
including in kind and/or fixed obligation grant with leverage requirement, training on business 
management and improved nutrition and feeding practices along with technical support and others. 
Value of in-kind and cash grants budgeted for disbursement for 30 feedlot, poultry and dairy
55
enterprises was USD 675,000. The project actually disbursed USD 544,075 for 15 such enterprises, 
according to project progress reports.
The evaluation team consulted 8 livestock and poultry enterprises and gathered some primary and 
secondary data to gain insight about their financial status, impact of the project on their operation, 
and so on. While some qualitative and quantitative data was gathered, the evaluation team found it 
difficult to get financial reports of many of the enterprises. This was largely because many of them 
did not process their financial transactions and produce financial statements. In light of this, the 
evaluation team had to base its financial analysis on data that has been collected by ACDI/VOCA on 
monthly basis from 12 enterprise in 2016, 2017 and 2018. Before putting the data into use, however, 
the evaluation team verified many of the figures maintained by ACDI/VOCA by comparing financial 
and non-financial data gathered by the evaluation team from the enterprises.
As shown in Table 27 below, the project and enterprise owners invested Birr 11,377,035 over the 
past three years. The RoI of the 12 enterprises averaged 26% which can be considered very attractive 
by any standard. This means that for every Birr 100 invested, the livestock enterprises earn Birr 26 
as profit. However, it is important to note that the two fattening firms incurred a loss in the past three 
years, which might be due to the nature of their business. In other words, fattening firms incur a lot 
of costs and start generating revenue with the fattened animals are sold which could take some time. 
Among the three categories of enterprises, poultry farms earned the highest RoI followed by dairy 
firms. When we look at the operation of individual enterprises in each category (See Annex 8), it 
becomes clear that much of the income was actually generated by few private enterprises, while some 
others have either earned low income or experienced losses. Low or negative profit is 
understandable particularly for newly established enterprises until they manage to produce and sell 
in larger quantity that offsets their large initial costs. Their return is expected to grow in the future. 
In conclusion, the results suggest that the enterprises that the FEED II project supported are 
financially viable.
Table 27: Return on investment (RoI) of livestock and poultry enterprises supported by the project
Year Item
Dairy Total 
(7 
enterprises)
Poultry 
Total (3 
enterprises)
Fattening
Grand Total (2
enterprises) Total
Investment 3,605,998 2,555,010 472,746 6,633,754
CASH IN (Revenue) 2,538,396 1,023,945 339,000 3,901,341
2016 CASHOUT (expense) 2,486,375 695,919 290,340 3,472,634
NET CASH 52,021 328,026 48,660 428,707
ROI 1.4% 13% 10% 6%
New investment 2,922,734 - 1,490,000 4,412,734
T-Investment 6,528,732 2,555,010 1,962,746 11,046,489
2017 CASH IN (Revenue) 6,071,586 1,846,298 840,240 8,758,124
CASHOUT (expense) 4,606,104 1,185,494 1,292,972 7,084,570
NET CASH 1,465,482 660,803 (452,732) 1,673,553
56
ROI 22% 26% -23% 15%
2018
New investment 117,880 - 117,880
T-Investment 2,584,321 1,490,000 4,074,321
CASH IN (Revenue) 1,846,953 284,000 2,130,953
CASHOUT (expense) 981,538 343,487 1,325,024
NET CASH 865,416 (59,487) 805,929
ROI 33% -4% 20%
Sum
Total Investment 6,859,279 2,555,010 1,962,746 11,377,035
CASH IN (Revenue) 10,456,935 2,870,242 1,463,240 14,790,418
CASHOUT (expense) 8,074,017 1,881,413 1,926,798 11,882,228
NET CASH 2,382,918 988,829 (463,558) 2,908,189
ROI 35% 39% -24% 26%
Data source: ACDI/VOCA, 2019.
All livestock and poultry enterprises consulted by the evaluation team use concentrate feed for their 
livestock/poultry. The union-based feed plants established by the project were also identified as key 
sources of concentrate feed. With use of concentrate feed, the enterprises improved their 
productivity and income. ‘Concentrate feed manufacturing factories installed by the project are 
contributing to the supply of different rations for specialized animal farms such as poultry, fattening 
and dairy. On the other hand, the demand for these products has increased because animals that are 
fed with such better quality feeds have become more productive,’ a government official said.
Feeding fattened animals concentrate feed was considered very effective in reducing the fattening 
cycle time. The feed development initiative of the project is also said to be catalyzing the expansion 
of livestock and poultry enterprises. A regional agriculture bureau official in this regard said the 
following.
‘Another commendable activity of the Project is the installation of a compound feed manufacturing 
machine which is an innovative undertaking, now playing a catalytic role in the expansion of modern 
poultry, fattening and dairy enterprises. By using fattening rations obtained from Farmers Union￾based feed manufacturers, it has become possible to fatten cattle in less than three months.’
Use of concentrate feed has also increased livestock and poultry productivity. One of the dairy 
enterprises supported by the project, for example, has been using concentrate feed for dairy cows 
since 2013. The enterprise source the feed from nearby union-based feed manufacturing enterprise 
and other private ones. According to the owner of the enterprise, the supply of manufactured feed is 
growing. With the use of concentrate feed, the enterprise was able to increase average milk yield per 
day per crossbreed cow to 16 liters from 12 liters in the past. The enterprise produces cheese, butter, 
yogurt and milk and sells them to farmers and urban dwellers. It buys milk from farmers. When asked 
about the gaps and weaknesses in feed manufacturing supply chains, the enterprise owner identified
57
high price and poor quality of manufactured feed. The respondent referred to manufactured feed in 
general. The findings showed that union produced feed is of better quality and price compared to 
other suppliers though it is still considered high from buyers point of view. Many other livestock and 
poultry enterprises also shared similar opinions. A case story below illustrates the impact of the 
intervention on the operation of a poultry enterprise supported by the project
Bagerish 10Poultry Enterprise was one of the private firms that 
received support from the project. The main products of the 
enterprise are pullets (young and below one-year old female 
chicken that has not started laying egg), broiler chicken for 
meat, and 45 days chicken. Its major customers are government 
agriculture office to distribute for farmers; and unemployed 
youth to start poultry enterprise. In the past five years, the 
enterprise’s revenue grew from ETB 228,036 in 2014 to ETB
1,091,700 in 2018. But it started earning a profit since 2018 which is expected to grow in the coming 
years with increased production and sales.
The Poultry Farm had 6,120 one-week old chickens at the time 
of the visit. The enterprise has been buying concentrate feed 
since 2014 after receiving advise from development agents. 
According to the enterprise , the supply of manufactured 
feed has been increasing in the woreda/kebele. Significant 
improvements have been made in poultry productivity after the 
enterprise received assistance from the FEED II project. 
According to of the enterprise, average egg laid per 
chicken per month increased from 18-20 eggs to 30 eggs. 
Average egg laid per chicken throughout its life was also said
to increase from 480 to 720.
Impact on Dealers
Concentrate feed dealers are often primary cooperatives that are members of unions that 
manufacture concentrate feed. These primary cooperatives are ideal for making concentrate feed 
accessible to farmers. They also offer the feeds at lower prices than traders. Some of the cooperatives 
that sell manufactured feed indicated that their engagement in sale of such feed has increased their 
income and diversified their income sources. But they were not able to give figures showing amount 
of income earned as they tend to mix the finance relating to concentrate feed with other commodities 
they trade. Most importantly, they are happy with their engagement in concentrate feed dealership 
as doing so allows them to make such feed accessible to their members. Hence, apart from getting an 
additional source of income, their involvement in sale of manufactured feed is contributing to their 
overall purpose of improving the productivity and lives of their members. ‘Dealers of feed 
manufacturing enterprises supply feed to farmers and others with more moderate price than traders.
Apart from reducing prices, the distribution of manufactured feed through dealers increased
10 Evaluation team chose to use pseudo names instead of real ones to protect privacy and confidentiality.
58
availability of feeds to farmers. It also created job opportunities for dealers,’ a government official 
said.
Dealers and particularly cooperatives, however, seem to believe that awareness about the 
concentrate feed is yet low. As a result, demand for such feed is not as much as they expected. Apart 
from awareness, however, the high price of concentrate feed and inadequate ownership of improved 
breeds for which farmers often prefer to use manufactured feed were identified as constraints to low 
demand for such feed.
Like cooperatives, private agro-dealers that are involved in retailing concentrate feed are also happy 
with the results they obtained. The dealers indicated that their involvement in sale of manufactured 
feed increased their income. They mostly sale the feed to farmers by working closely with 
development agents. The case story presented below illustrates the impact of increased concentrate
feed supply on an agro-dealer.
An agro-dealer in one of the project woredas started retailing concentrate feed manufactured 
by union-based plants in early 2017. It sells rations for fattening, poultry and dairy that it often 
procures from Bokoji Union Feed Manufacturing plant. But before 2017, it used to buy the feed 
from private firms. On average, the dealer sells 192 quintals of concentrate feed per month.
The dealer believes that there is no adequate supply of concentrate feed. The major customers 
of its concentrates are farmers, other agro-traders, livestock fattening enterprises, poultry 
farms/ enterprises and dairy farms. According to the owner, the agro-dealer sells about 70% of 
its concentrates to farmers, who are often reached by working with government and 
development agent as well as cooperatives. The dealer also said that potential demand for 
concentrate feed is large. The average number of quintals of concentrate feed that the dealer 
sells per month along with the price is as follows.
The dealer observed positive changes in income after retailing concentrate feed. According to 
the owner, average monthly income increased from Birr 60,000 to Birr 153,600 after he started 
retailing concentrates. Profit also grew from Bir 2700 per month to 10,500 per month. The 
dealer shared that his customers are happy with the quality and results they achieved with the 
concentrate feed. As a result, their number is growing from time to time. Attracted by the return, 
the dealer is planning to construct his own manufacturing feed enterprise and expand his shops 
to make them accessible to buyers.
When asked about the challenges the dealer often faces in retailing concentrate feed, the owner 
identified lack of adequate and reliable supply of the feed from unions as well as problems of 
transportation.
iii. Impact of project-supported livestock and poultry enterprises on downstream economic 
activity of traders and consumers
Smallholder farmers are among the suppliers of inputs such as milk to livestock enterprises. The 
support provided by the project to cooperatives to engage in dairy processing activities played an 
important role in creating some market access to farmers to sell milk and milk products. In this
59
regard, one direct beneficiary farmer that participated in one of the FGDs said ‘The project 
established a dairy cooperative and supported the construction of a dairy shop. Thus, we have good 
access to buy manufactured feed and sell milk.’ Government bodies also had a similar opinion. Apart 
from creating demand for concentrate feed manufactured by unions, the establishment of livestock 
and poultry enterprises was said to be a smart approach to improve market access to farmers sell 
their produce to and buy products from the enterprises. However, access to markets has remained 
one of the major challenges of smallholder farmers. As shown in Figure 10, only 38% of respondents 
sold their milk to dairy enterprises/cooperatives. About 44% of them are still travelling to the local 
market to sell their milk. Some 38% of the respondents are also selling their milk to neighbors.
Figure 10: Buyers of milk produced by smallholder farmers
% of respondents andy buyers of their milk 
products
Other Specify 5%
Nearby Market 44%
Neighbors 38%
Dairy enterprise / Cooperatives 38%
0% 10% 20% 30% 40% 50%
The qualitative data analysis results also support quantitative findings. While appreciating the 
changes that have been happening in their production and productivity, beneficiary smallholder 
farmers that participated in FGDs expressed their frustration about the lack of access to market to 
sell their increased livestock production.
‘We are benefiting from sale of eggs in a better way. We are producing high amount of milk although 
there is not that much demand for it. We convert the milk into butter, which sells at a price of birr
150 per kilogram. It is a very good income. We also buy day-old chickensfor Birr 70 and earn a profit
from selling them at Birr 250. All of those benefits are very high when compared with the situation 
five years ago. However, access to markets has remained a critical problem particularly during the 
fasting seasons. During this season, the price of our products gets lower and we can’t hold our 
products for longer period of time as they are perishable’ a beneficiary farmer that participated in 
one of the FGDs said.
60
Market access is particularly a major problem for farmers that are involved in fattening activities. 
Unlike dairy and poultry products, it is not easy to find buyers for fattened livestock given their high 
value. As a result, farmers are often forced to go to central or larger market to sell their fattened 
livestock which are sometimes located hundreds of kilometers away from their residence. A 
development agent in this regard said the following:
‘It is very good in providing fodder crop. Although the problem is the farmers sell them in the market 
they found after fattening the cattle. For example, they take them straight (directly) to Mekele city 
[capital city of Tigray region] or to Abiy Gidy city to sell them. There is no one that buys them going 
to their houses. Besides, fattened animals are only demanded when it is holiday time.’
Poultry enterprises that have been supported by the project also created some market access for 
beneficiary households to buy poultry products. As shown in Table 28 below, about 45% of direct 
beneficiary households bought poultry and poultry products from nearby a poultry enterprise. A 
significant majority (71%) of respondents said that poultry products have become available in their 
areas in the past five years. About half of the respondents also said that poultry products have become 
more affordable than before in their areas or access to buy improved chickens/hens has increased 
over the past five years. Increased egg production, increased income and increased production of 
chickens were identified as major changes in livelihood over the past five years by 72%, 58% and
52% of households, respectively.
Table 28: Beneficiary households’ access to poultry market
Questions Response Categories N %
Have you bought poultry 
products in the past five
years?
No 549 29%
Yes 1345 71%
Where did you buy the 
poultry products from 
during the past five years?
Nearby Poultry Farm/Cooperative 604 45%
Nearby Market 985 73%
Others (specify) 19 1%
What has changed in poultry
product availability and 
affordability in the past five 
years?
Poultry products have become
available in my area
1353 71%
Poultry products have become more 
affordable than before in our area
897 47%
Access to buy improved chickens/hens 
has increased
1005 53%
Others (specify) 120 6%
What has changed in your 
household's lives in the past
five years?
Increased egg production 1370 72%
Increased number/production of 
chickens
993 52%
Increased income 1092 58%
Others (specify) 225 12%
The project also had an indirect impact on traders that buy products from enterprises it supported. 
Consultations made with restaurants and retail shops that buy products of enterprises showed that 
the enterprises created opportunities for them to buy dairy and poultry products in nearby areas. 
Some traders also found the prices charged by project-supported enterprises a bit lower than others.
61
The quality of dairy or poultry products supplied by the enterprises was also considered good 
compared to other suppliers. The case of a coffee shop that buys from Mestewal dairy enterprise may 
illustrate the benefits and impact very well. Chaltu (pseudo name) Coffee shop buys milk from 
Mestewal dairy enterprise over the past six months. When asked about the benefits she gained from 
buying dairy products from the dairy enterprise, Chaltu identified access to buy dairy products, 
buying dairy products at cheaper/lower prices, improved household nutrition and health, and
improved quality of dairy product.
2.1.6.4. Challenges in production, distribution, and use of formulated feed
The production and marketing of formulated feed in being challenged by different factors. Some of 
these are national level challenges that are beyond the control of feed producers. The overall 
challenges at country level include: 1) escalating prices of feed ingredients driving up compound feed 
production costs; 2) unsupportive multiple taxations; 3) periodic feed ingredient shortages; 3) feed 
ingredient safety and quality, 4) unreliable demand for compound feed, 5) lack of research and 
extension support for the commercial feed sub-sector, 6) lack of feed analytical services at 
commercial scale 7) unreliable power supply; 8) shortage of qualified and dependable labor; 9) 
shortage of hard currency for companies engaged in importing specialty feed ingredients (e.g. pre￾mixes); and 10) organizational capacity of the Ethiopian Animal Feed Industry Association to serve 
the needs of its members and the sector at large. In addition, awareness about formulated feed has 
remained inadequate and mostly project intervention areas specific. Areas other than those not 
covered by project have still limited awareness about the value of using formulated fee. Moreover, 
farmers using the formulated feed sometimes do not get the intended benefit due to poor targeting 
of the most responsive animals like selective feeding of crossbred cows, feeding the appropriate 
levels to get the desired responses, minimization of wastages, etc.
In addition to the overall country-level challenges that affect both private and union feed plants, the 
following are additional challenges at the union factories:
i. Input supply: These include (1) inflexible financial rules and regulation (Proforma based 
purchase system); (2) poor coordination between managers and boards which cause a delay 
for approval as some of the input suppliers give 2 to 3 days’ time limit for prices; (3) shortage 
of transport for inputs. Some inputs especially premixes for poultry rations, having to be 
procured from Addis Ababa. 4) Input price fluctuations 4) storage and financial constraints 
limit the purchase of inputs at times of harvest when prices are low.
ii. Machinery maintenance and upkeep: Some unions indicated occasional breaks in 
operations due to machine breakdowns, but said they typically resolve these within a few 
days (e.g. trained staff repair), or enact work arounds (e.g., replace automation with 
manual processes). The occasional problem of availability of spare parts is also reported.
iii. Power supply: Power interruption has an adverse effect on feed manufacturing. Some unions 
have reported production interruptions of 2-4 days a week due to power supply. Installation 
of standby generators is proposed to avoid production interruptions.
62
iv. Other challenges: Some unions are managing many different enterprises and there is some 
lack of giving adequate attention to the feed enterprise. Shortage of skilled workforce in feed 
production and management as a result of high staff turnover has been mentioned by some 
unions.
There are also challenges in the distribution and sale of formulated feed of the union plants. The main 
distribution channels of union feed plants are through PCs, agro-dealers and direct sales to farmers. 
The general challenges related to distribution include:
• Marketing plan: none seemed to have a clear or comprehensive marketing plan.
• Limited access points: The distribution networks have limited access points/retail shops that 
are not accessible enough to many users in outlying areas resulting in additional transport 
costs.
• Inadequate storage facilities: Storage facilities for safe storage of ample quantities at many 
distribution sites to meet demand is a problem. This forced distributors to acquire small 
quantities more frequently creating gaps in supply and additional costs of transport. The 
available storage space in many cases is committed to other established products handled 
rendering the feed to possible contamination;
• Transport: Most distributors lack their own means of transport and generally depend on 
rented vehicles causing delays and high prices.
• Archaic payment system at some factory outlets: Long and bureaucratic payment system like 
the Gozamen union where buyers can’t pay in cash to the factory and access the concentrate 
feed. Instead, buyers must deposit the payments in the bank (either the Commercial Bank or 
Abay Bank) and come up with the slip to access the concentrate feed. This discourages 
customers. Arranging a one-stop payment system is critical under such situations.
• Formulated feed supply/demand: some PCs and agro-dealers that are in the network complain 
about inadequate feed supply from feed mills. Some 25% of retailers indicated shortage of 
supply from feed plants as a problem. Lack of adequate demand was mentioned as a problem 
by 50% of retailers surveyed. Most (63%) attribute the poor demand to lack of awareness 
among feed buyers/farmers. A lot needs to be done in raising awareness on the necessity and 
benefit of feeding concentrated feed including feeding to the most responsive animals to 
realize the benefits more visibly.
Weak organizational capacity of PCs: Many unions use PCs as distribution channels. PCs are
not well organized to effectively distribute to farmers. Most PCs do not have dedicated store 
staff or salespersons and may not keep regular sales hours. Point-of-sale education is very 
limited, though some PCs and agro-dealers have received training. PCs are bound by 
restrictive policies relative to agro-dealers like bureaucratic decision-making,
organizational structures and leadership teams that are less responsive to market changes 
and opportunities quickly.
• Unaffordability of price: Average price of manufactured feed at union-based feed 
manufacturing enterprises indicates that the most expensive formulated feeds across regions 
are poultry feeds (average of 1049 ETB/quintal) while fattening feed was the least expensive 
(699 ETB); formulated feed across feed types was on average most expensive in the Tigray 
region and least expensive in the SNNP. Fish feed was produced in the SNNP region only and 
had the lowest price of all formulated feeds. The variability across regions largely has to do
63
with ingredient prices. But most farmers that participated in the FGDs and government 
officials alike complained about the price of concentrate feed. They believe that the price is 
too high for many farmers to buy and use it. Cost reduction measures along the value chain 
need to be explored to further reduce prices.
2.1.7. Income from Livestock and Livestock Products
One of the key objectives of the project was to increase target smallholder farmers income from 
livestock and poultry. The target was to increase income from livestock and poultry by 15%. In this 
section, the findings of the final evaluation on income earned by smallholder farmers from livestock 
and poultry sales are presented.
2.1.7.1. Income from Sale of Fattened Oxen/Bulls
Close to 43% of the respondents sold fattened oxen and/or bulls 
during the 12 months before the endline survey, an increase from
28% at baseline. The biggest increase in number of households that
fattened and sold oxen was observed in Amhara region, while a slight 
decline was seen in Oromia. Households that participated in the 
endline survey sold 1,450 fattened oxen and earned close to Birr 21 
million. While the number of oxen sold at endline was 35% higher 
than the baseline, the value earned increased by over 150%, which 
might be due to improved body weight gained by the oxen as a result 
of improved feeds. On average, a household earned Birr 11,053 from 
sale of one fattened oxen, compared to Birr 4,366 at baseline, a 153% 
increase from baseline. Households in Amhara region earned the 
highest income from sale of fattened oxen than other regions. Income
from sale of fattened oxen was lowest in SNNPR (see Table 29 below).
Table 29: Distribution of Households by Number of Fattened Oxen/Bulls Sold and the Value of those Sales and Region at
Baseline and Endline
Number of fattened 
bulls/oxen Sold
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
None 66 35a 65 70.3b 72 55c 84 69b 72 57
1-2 head 30 59a 30 27.2b 25 42c 9 21b 23 38
3-4 head 3 4a,b 5 2a 3 2a,b 4 5b 3 3
5-9 head 1 2a,b 0.2 .4a - .2a 2 4b 1 2
10+ head 0.2 .4a - .2a - .4a 2 2a 0.4 0.6
Total HHs (n) 478 480 480 464 477 483 479 467 1,914 1894
Number of fattened oxen sold 255 471 261 216 198 312 362 451 1,076 1,450
Total birr collected from sale
of fattened oxen
1,538,320 6,985,459 2, 373,756 4,882,900 1,344,459 3,859,924 3,100,200 5,206,010 8,356,726 20,934,293
Average proceeds per
household from sale of 
fattened oxen (birr)
3,218 14,553a 4,945.30 10,524 a,b 2,818.60 7,992b 6,472 11,148 a,b 4,366 11,053
64
Note: Values in the same row and sub table not sharing the same subscript are significantly different at p< .05
As few as 7% of households fattened and sold cow over the past year. As shown in Table 30 below, 
households earned close to 2 million Birr from sale of 195 fattened cows. While the number of 
fattened cows grew by 57% from baseline, income increased from sale of these cows increased by
112%. Average income earned from sale of a fattened cow also grew from 492 Birr to 1,055 Birr.
Table 30: Distribution of Households by Number of Fattened and Sold Cows and the Corresponding Sales and Region at
Baseline and Endline
Number of cows sold Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
No fattened cow sold 90 89a 93 95b 94 92a,b 97 95b 94 93
1-2 fattened cows sold 9 10a 6 5b,c 6 8a,b 2 4c 6 7
3+ fattened cows sold 1 .6a 1 .2a 0 .4a 1 1a 1 1
Total HHs (n) 478 480 480 464 477 483 479 467 1,914 1,894
Number of fattened cows sold 63 69 46 30 43 49 26 47 124 195
Total sales collected (birr) 281,198 792,348 230,970 377,800 246,500 444,000 182,200 383,000 940,868 1,997,148
Average earnings per HH from 
selling fattened cows (birr) 588 1,650.7a 481 814.2a 517 919.3a 380 820.1a 492 1,055
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
The proportion of households that fattened and sold sheep did not change much over the past five 
years. Indeed, it declined slightly. Only 16% households fattened and sold sheep, compared with 18% 
at baseline. These households earned Birr 3.3 million from the sale of 1,272 sheep over the past year, 
compared with 2.1 million from sale of 1,882 sheep at baseline. Hence, although the number of 
fattened sheep sold declined slightly, the income earned from the sale of sheep increased by over Birr
1.2 million. This might be due to inflation and use of improved feed to fatten sheep. On average,
income of Birr 1,759 was obtained from the sale of a sheep last year, compared with Birr 1,102 at 
baseline.
As also mentioned in previous sections, few households (2%) owned and fattened goats. These 
households sold 414 fattened goats and earned Birr 743,615. The average proceeds obtained from 
sale of a fattened goat was Birr 393, compared with 142, a 174% increase from baseline.
2.1.7.2. Quantity of Dairy Products Sold and their Value
The percentage of households that sold milk declined slightly over the past five years from 19% at 
baseline to 14% at endline. Similarly, the proportion of households that sold 1,000 liters or below 
per year declined to 5% at endline compared to 17% at baseline. In contrast close to 7% of the 
households sold over 1,000 liters of milk over the past year before the endline survey, compared with 
just 2% at baseline. This may suggest that even if the proportion of households that sold milk 
declined, the amount of milk sold at endline was much higher than the baseline. Indeed, endline
65
respondents sold 504,756 liters of milk and earned over 7.4 million Birr, which was 2.5 and 5.4 times 
the baseline results, respectively. On average, a household earned Birr 3,954 during the year 
preceding the survey, compared to just Birr 610 at baseline (see Table 31 below). It is important to 
also note that price of milk increased by over 83% over the past five years. Hence, a significant 
portion of the change in value of income earned from sale of milk is attributed to the increase in price 
of milk. Even at baseline prices, however, average income from sale of milk increased by over 200%. 
In sum, increased production and productivity of milk as well as some market access and increased 
price were said to have contributed to the significant increase in income from sale of milk. Qualitative 
and quantitative findings also associated the dramatic increase in productivity and income from milk 
to the integrated interventions of the project as explained in the different sections of this report.
Table 31: Distribution of Households by Quantity of Milk Sold and Corresponding Value of Sales from Milk at Baseline and
Endline
Amount of milk 
sold in liter
Amhara Oromia SNNPR Tigray Total %
change
Amount of milk 
sold in liter
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
No milk sold 88 96.5a 84.6 79.7b 63.6 83.4b 87.5 85.4b 81 86.3 -5.3
<=1000liter/ year 10.7 2.1a 12.5 6.7b,c 36.2 11b 9 5.8c 17 6.4 -10.6
> 1000liter/ year 1.3 1.5a 2.9 13.6b 0.2 5.6c 3.5 8.8b,c 2 7.3 5.3
Total HHs (n) 478 478 480 462 477 471 479 464 1,914 1,875
Liters of milk sold 
last year
16,950 13,272 63,101 260,100 8,478 65,208 57,243 166,176 145,773 504,756 246
Total sales 
collected from 
milk sale (birr)
91,732 154,920 498,336 3,418,098 73,970 991,080 504,002 2,848,860 1,168,042 7,412,958 535
Average 
proceeds per HH 
from milk sale
192 324.1a 1,038 7398.5b 155 2104.2 a,b 1,052 6139.78b,c 610 3,954 548
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
The proportion of households that sold butter did not change from baseline. As summarized in Table
32 below, only a third of household sold 22,170 kgs of butter and earned close to Birr 4 million at 
endline. Amount of butter sold and income earned from it was much lower. On average, a household 
earned Birr 2,106 from sale of butter at endline, compared to 667 at baseline, a 215% increase over 
the past four years.
Table 32: Distribution of Households by Quantity of Butter Sold and the Value of Sales and Region at Baseline and Endline
Amount of butter sold (kg)
last year
No butter sold (% of HHs)
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
60.6 52.5a 70.9 61.6b 68.2 79.1c 67.8 69.8b 66.8 65.8
<100 kg/year (% of HHs) 37.5 45.6a 28 34.9b 31.5 19.9c 32 29.1b 32.3 32.4
>100 kg/year (% of HHs)
1.9 1.9a 1.1 3.4a 0.2 1.0a 0.2 1.1a 0.9 1.8
66
Total HHs (n) 478 480 480 464 477 483 479 467 1,914 1,894
Amount of butter sold (kg)
last year
3,819.9 7,111 3,867.3 6,857 1,298.8 5,035 2,277.3 3167 11,263.1 22,170
Total sales collected from
butter sale, birr 419,706 1,105,038 437,256.5 1394305 153,769.5 808,760.0 266,584 680615 1,277,316 3,988,718
Average proceeds per HH
from butter sales
878 2,302.2a 911 3,005a 322.4 1,674.5a,b 556.5 1,457.4b 667.4 2,106
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
2.1.7.3. Sales of Poultry and Poultry Products
As few as 12% of households sold 1,236 pullets during the 12 months before the endline survey, and 
earned income of Birr 178,294, compared to Birr 413,207 at baseline. The number of pullets sold 
declined three-fold at endline and income from sale of pullets also declined by 57%. The decline is 
attributed to the fact that one respondent at baseline sold 67% (3,181 pullets of 4741) of the total 
number of pullets sold by all respondents, which was not observed at endline. This was also noted in 
the baseline report (see Table 33 below).
Table 33: Distribution of Households by Number of Pullets Sold and the Value of Sales and Region at Baseline and Endline
Number of
pullets sold
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
No pullets sold
(% of HHs)
85.1 73.5a 81.5 95.3b 86 96.7b 79.1 86.7c 82.9 88.0
1-4 pullets sold
(% of HHs)
9 21.9a 13.5 3.4b 10.9 1.9b 14.4 4.9b 12 8.1
5-9 pullets sold
(% of HHs)
4 2.9a,b 4 .9a 1.9 .8a 4.4 4.7b 3.6 2.3
10+ pullets sold
(% of HHs)
1.9 1.7a,b 1 .4a 1.3 .6a 2.1 3.6b 1.6 1.6
Total HHs (n) 478 480 480 464 477 483 479 467 1914 1894
Total number of
pullets sold last 
year
387 352 3181 87 291 140 882 657 4741 1236
Total sales
collected from 
sale of pullets 
(birr)
24,629.4 41,224 270,348 14,320 20,089 21,610 98,141 101,140 413,207 178,294
Average
proceeds per HH 
from pullet sales 
(birr)
51.5 85.9 563.2 30.9 42.1 44.7 204.9 216.6 215.9 94.13
Note: At baseline one individual in Oromia sold 2,850 pullets.
Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
The proportion of households that earned income from sale of broilers declined from 23% at baseline 
to 11% at endline. The number of broilers sold during the 12 months before the endline survey 
(1,550) was also lower than the baseline (1,941). Even so, endline respondents earned Birr 275,449, 
which was much higher than the income of Birr 190,093 at baseline. Households on average earned
67
income of Birr 145 from sale of broilers, compared to Birr 99 at baseline, an over 46% increase over 
the past five years (see Table 34 below).
Table 34: Distribution of Households by Number of Broilers Sold and Value of Sales and Region at Baseline and Endline
Number of broilers sold Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
No broilers sold (% of HHs) 77 86.7a 79.6 84.5a 80.3 98.1b 70.6 85.4a 76.9 88.8
1-4 broilers sold (% of HHs) 16.5 10.2a 14.6 8.8a 16.8 .8b 22.5 8.4a 17.6 7.0
5-9 broilers sold (% of HHs) 5.2 1.9a,b 4.2 3.9a 1.7 .4b 4.2 2.4a,b 3.8 2.1
10+ broilers sold (% of HHs) 1.3 1.3a,b 1.7 2.8a,b 1.3 .6a 2.7 3.9b 1.7 2.1
Total HHs (n) 478 480 480 464 477 483 479 467 1914 1894
Total number of broilers sold last 
year
439 260 439 429 292 69 771 792 1941 1550
Total sales collected from sale of 
broilers (birr)
37,854.3 46327 44,715 100108 27,825 7995 79,699 121019 190,093 275449
Number of HHs sold broilers 110 64 98 72 94 9 141 68 443 213
Average proceeds per HHs from 
sales of broilers (birr)
79.2 96.5a 93.2 215.8b 58.3 16.6c 166.4 259.1a,b 99.3 145.4
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
2.1.7.4. Sale of Spent Hens (Old hens)
About 19% of households earned income from sale of spent hens at endline, compared to 13% at 
baseline. Households, on average, earned Birr 203 from sale of spent/old hens, compared to Birr 29 
at baseline. Income from sale of old hens also grew by seven-fold over the past five years (see Table
35 below).
Table 35: Distribution of Households by Number of Spent Hens Sold and Value of Sales and Region at Baseline and Endline
Number of spent (old)
hens sold
Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
No spent hens sold 93.3 68.5a 84.4 84.7b 82 95.9c 89.6 76.2d 87.3 81.4
1-4 spent hens sold 5.9 27.5a 14 8.8b 15.3 2.9c 7.5 13.3b 10.7 13.1
5-9 spent hens sold 0.6 2.5a 1.5 3.4a 2.3 .4b 2.3 5.4a 1.7 2.9
10+ spent hens sold 0.2 1.5a 0.2 3.0a,b 0.4 .8a 0.6 5.1b 0.4 2.6
Total HHs (n) 478 480 480 464 477 483 479 467 1914 1894
Total number of spent 
hens sold last year
82 446 187 465 231 107 170 998 670 2016
Total sales collected 
from sale of spent hens 
(birr)
6,491 72,633 16,656 101,095 18,847 18,000 13,100.75 193,395 55,093.8 385,123
68
Average proceeds per 
HH from spent hen sales 
(birr)
13.6 151.3a 34.7 217.9 
a,c
39.5 37.3b 27.4 414.1c 28.8 203.3
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
2.1.7.5. Sale of Eggs
Close to 58% of households sold eggs during the 12 months period before the endline survey, an 
increase of 7% from baseline. Income earned from sale of eggs rose over 13-fold. But it is important 
to note here that part of the increase in income was caused by increase in price by 74% in the past 
five years (inflation). On average, a household received Birr 4,075 from sale of eggs at endline, which 
was an over 26-fold increase from baseline. Such dramatic changes can be largely attributed to 
improved productivity caused by increased ownership of improved chickens and use of improved 
feed and poultry management practices. Hence, the changes can be mainly attributed to the FEED II 
project interventions (Table 36).
Table 36: Number of Eggs Sold and Value of Sales from Eggs in the Last Year by Region at Baseline and Endline
Sale value of eggs Amhara Oromia SNNPR Tigray Total
Baseline Endline Baseline Endline Baseline Endline Baseline Endline Baseline Endline
HHs that sold 
eggs last year
(%)
56.7 77.1a 55.4 61.6b 40.3 30.8c 49.7 61.5b 50.5 57.7
Number of eggs 
sold
36,254 557,856 24,065 242,052 38,375 185,160 54,666 324,912 153,351 1,309,980
Sale value of eggs 60,405 1,718,166 47,045 780,465 74,334 652,647 100,017 1,033,719 281,802 4,184,997
Average 
proceeds per HH 
that sold eggs
126 5083.3a 98 2858.8b 156 4763.8a,c 209 3705.1b,c 147 4,075
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
Overall, there were significant changes in income from livestock and poultry over the past five years. 
Many beneficiaries have experienced increased household income from sale of milk, eggs and 
fattened animals in the past five years largely due to the projects intervention in raising livestock 
productivity such as better-quality forage and manufactured feeds and improved animal 
management.
2.2. Relevance
Relevance of the project was mainly assessed considering the level of participation of stakeholders 
throughout the project implementation cycle and their satisfaction with its results, and alignment of 
the project with government and target group priorities.
2.2.1. Alignment of project initiatives with government and target 
groups priorities
Ethiopia has the largest livestock population in Africa and the fifth largest in the world. The livestock 
subsector contributes nearly 20 percent of total GDP and foreign exchange earnings of the country, 
and some 35 to 40 percent of agricultural GDP. However, Ethiopia has not experienced significant
69
productivity gains in the livestock sub-sector. Meat and milk production and productivity have 
remained low. Milk yield averages 1.5 liters per day, which is about one‐eighth of the milk yield for 
improved dairy breeds. Average daily milk yield from crossbred cows was also 8 liters per cow per 
day in 2014/15.
Similarly, there is a substantial productivity difference between local and improved poultry breeds. 
The local poultry breeds lay only 50 eggs per year, which is one‐third the number laid by modern 
breeds. As a result of these low productivity levels, the livestock sector is characterized by relatively 
high greenhouse gas (GHG) emissions per unit of product. Average GHG emissions are 19 kg CO2 
eq/kg milk among mixed crop‐livestock systems in Ethiopia, against an average of ca. 9 kg CO2 eq./kg 
milk in Sub‐Saharan Africa11.
According to the World Bank, the key challenges that continued undermining the performance and 
potential of the livestock sub-sector include: reduced availability and access to communal grazing 
and natural pasture; insufficient access to forage, forage seeds and feed supply; poor animal health 
due to disease prevalence; and low livestock genetic make‐up. Limited adoption of improved 
livestock practices and poor provision of livestock support services are also major sources of low 
productivity levels. The public sector dominates livestock support service delivery, which is weak 
for animal health, breeding, feed, and extension services. In addition, the sector has a low commercial 
market off‐take due to inadequate processing and marketing infrastructure. Most farmers do not 
participate in the livestock market and household livestock production is ultimately consumed 
within the household or sold on the local market (World Bank, 2017).
Cognizant of its challenges and potential, the livestock sub-sector has been made a priority in the 
government’s Second Growth and Transformation Plan (GTPII) from 2016‐2020. In the GTP 2 period, 
the livestock sub-sector is seen as critical in achieving priority goals of the government including 
contributing to overall economic growth, contributing to poverty reduction, improving food security, 
and supporting the country’s green growth priorities. These objectives are reflected in the recently‐ 
approved Livestock Master Plan (LMP). The investments proposed in the LMP include appropriate 
combinations of genetic, feed and health interventions and related policy changes to improve 
livestock productivity and the performance of the value chains. The interventions are meant to 
transform traditional family farms into improved market‐oriented systems, to improve household 
incomes, food security, livestock product consumption and nutrition, and to contribute to national 
economic growth.
As clearly stated in the GTP 2 and the 2015-2020 Ethiopia Livestock Master Plan12, the high cost and 
low availability of good quality animal feed from forage and fodder is one of the major constraints, if 
not the most critical constraint, to increasing productivity of livestock in dairy farms and feedlots, 
improved family and specialized poultry, and smallholder mixed crop-livestock and extensive 
livestock production systems. Against this background, the government has aimed to increase
average daily milk yield from crossbred cows from 8 liters per cow per day in 2014/15 to 12 liters
11 The World Bank. 2017. Livestock and Fisheries Sector Development Project (P159382)
12 International Livestock Research Institute (ILRI). 2015. Ethiopia Livestock Master Plan: Roadmaps for Growth 
and Transformation (2015-2020)
70
per cow per day by 2019/20. GoE has also targeted to increase average meat production/cattle 
carcass yield from 107 kg in 2014/15 to 138 kg by 2019/20. Total eggs production is also projected 
to increase from 163 million in 2014/15 to 3,938 million by 2019/2013.
To deal with the challenges and achieve targets set for GTP II, different implementation strategies 
have been devised and implemented, including improving the genetics of livestock, and improving 
supply of livestock feed, integrated implementation of livestock value chain efficiency. In the GTP 2 
period, the GoE has targeted to increase the number of cattle with improved genetics from 902,390 
in 2014/15 to 4,902,000 by 2019/20. The findings showed that farmers are increasingly adopting 
improved practices such as investing on crossbreed livestock and poultry while reducing their stock 
of local livestock and poultry.
The 2015-2020 Ethiopia Livestock Master Plan also recognized shortage and unavailability of quality 
purchased concentrate feed and roughage; and lack of effective feed quality control and standards 
enforcement mechanisms as critical specific challenges to dairy productivity and strategized to deal 
with the challenges by making land available for commercial forage production by investors; 
enforcing feed quality standards, quality monitoring and control; promoting the establishment of 
flour mills to make more concentrate ingredients available; and promoting domestic production of 
cooking oil to replace the importation of cooking oil, and limiting or banning the exportation of 
oilseeds that affect availability of concentrate feeds. In this regard, the FEED II project was also found 
well aligned and highly relevant to the execution these government strategies. The project supported 
the establishment and expansion of feed manufacturing at cooperative unions, which increased the 
supply and affordability of formulated feed. It also strengthened the capacity of regulatory bodies to 
enable them formulate and enforce animal feed laws and standards.
With the aim to improve production and productivity of livestock and poultry, the government has 
been working on improvement of supply of livestock feed. In this regard, the government has 
targeted to increase from 68 million tons in 2014/15 to 184 million tons by 2019/20. Livestock feed 
from industrial by-products was also projected to increase from 1.5 million tons in 2014/15 to 3.3 
million tons by 2019/2014. Given that the goal of FEED II is to increase the incomes of Ethiopian 
smallholder livestock producers by improving access to, and use of consistent, affordable, high 
quality animal feed that can support greater livestock productivity and efficiency, it is clear that FEED 
II is very supportive of this target of the government. The project improved supply of improved 
forage and formulated feed through capacity building, provision of seeds and seedlings, 
establishment of union-based feed manufacturing plants and nursery sites, promotion of improved 
on-farm feeding practices, and others.
The regions and woredas in which the project was implemented were also considered appropriate 
as they are among those woredas with higher potential for livestock and poultry production. Most 
importantly, livestock being the major source of livelihood for most smallholder farmers in the
project regions and woredas, the project was well received by farmers. Government bodies as well
13
Federal Democratic Republic of Ethiopia, National Planning Commission (NPC). 2016. Growth and
Transformation Plan II (GTP II) (2015/16-2019/20); Volume I: Main Text
14
ibid
71
as target beneficiaries found the project very relevant and well aligned with their priorities. Below is 
what an official of a regional agriculture bureau said which also represents the views of most 
stakeholders:
‘Project activities are well aligned with the national as well as the regional government’s policies,strategies 
and initiatives for the livestock sector. The regional livestock development initiative recognizes animal feed 
improvement both in quality and quantity as the primary area of development intervention. There is acute 
feed shortage in the region as determined by feed balance surveysthat indicated a deficit of 25%. The region 
has long standing policy to reduce free grazing of livestock in favor of on-farm feeding with improved 
quality of feedstuff thereby reducing the intensity of environmental degradation. Therefore, the activities 
of the FEED II project are in support of these policies and initiatives of the regional government. The project 
is also well aligned with the priority needs of the farmers. With the current shrinking grazing resource, 
farmers are showing ….interest in better quality feeds that can be fed on-farm.’
The project activities are directly related with the routine plan and duties of agriculture/livestock 
offices at regional and woreda levels. Their development strategy aims to improve livestock 
productivity and increase the contribution of livestock to family nutrition and income generation. 
FEEDII project supported livestock development by facilitating knowledge transfer, supporting 
adoption of improved technologies, improving supply and access to formulated feed and improved 
forage and other mechanisms. An official of another regional agriculture bureau said the following 
in this regard.
‘The project activities were highly or smartly aligned with our existing programs, policies and 
strategies to solve the existing bottleneck in feed development and animal feeding practices. Farmers 
in our region are highly interested in livestock development, but the critical problem we face in 
livestock development in our region is the shortage of feed both in quality and quantity. As the main 
objective of the project is to improve the livelihood of the farmers through increasing the productivity 
of their livestock by the improving the feed availability and quality, and animal feeding practices of 
the farmers. Therefore, I believe that the Feed-II project has addressed the basic limiting factor in 
livestock development of farmers in our region and the country at general level by providing grant 
to unions to establish for feed processing plants and forage development, training the farmers and 
organize experience sharing events.’
Considering smallholder farmers’ limited access to land for production of forage seed and forage, the 
project promoted backyard forage cultivation, and equipped farmers with the knowledge and skills 
needed to improve their forage growing systems. The project also provided improved seeds and 
vegetative materials to model farmers that helped them understand the value of using improved 
forage and obtain additional sources of feed. The capacity building support provided to farmers has 
also enabled them to utilize their feed efficiently, and properly feed and manage their animals. 
Alongside the promotion of improved forage at smallholder farmers, the project has also established 
forage nursery sites that could serve as sources of forage seed and seedlings to model and follower 
farmers over the long-term. These project activities contributed to improved livestock and poultry 
production and productivity, which has remained the critical priorities of the Government of 
Ethiopia. Farmers that participated in FGDs were also consistently stating how valuable and
72
important the project was to improve their productivity of their livestock on which their lives mainly 
depend.
The establishment of feed manufacturing plants under the unions was also accepted as both 
appropriate and innovative. Unions are associations of producer farmers established to improve the 
livelihoods of its members by closely working with them. The establishment of the feed plants under 
the jurisdiction of the unions is considered by all respondents from their perspectives as taking this 
new dimension of promoting concentrate feed production and utilization close to the beneficiaries. 
Farmers unions have gained the confidence of members thus making the process of adoption easier. 
The unions also have experience in business management by engaging in other enterprises. Farmers 
also have experience in using the services of unions. This was considered to be a recipe for success.
2.2.2. Stakeholders’ participation in project implementation and
satisfaction with the its results
The implementation approach was viewed as comprehensive and participatory. The project’s 
mutually supportive/ complementary intervention components that included provision of training 
(feed production/plant management, business/accounting skills), follow up technical support, 
material support, advisory support, support to poultry/dairy/fattening enterprises that utilize the 
manufactured feeds, promotion of improved on-farm feeding practices, creation of market linkage 
for input supply and product sales, exchange visits to facilitate knowledge exchange/peer learning, 
capacity building of government and association support staff at different levels were also 
appreciated.
The project engaged stakeholders at all levels in the initiation and implementation of the project. 
Development of the knowledge horizon at different levels including the level of the technical staff 
(DAs) cascaded down to the level of the beneficiary farmers was considered to be a prime 
contribution. Job creation and encouragement of private sector participation were addressed by the 
project. Besides, the project has been operating in conformity with the communities’ cultural, 
economic, and social context.
Almost all stakeholders consulted for the final evaluation expressed their satisfaction with the 
project results. ‘Generally, the project contribution can be rated as excellent in feed manufacturing 
and increasing of livestock productivity in our region,’ a said. Farmers 
that participated in the FGDs were also unanimous in stating their satisfaction with the project 
support. ‘We are very satisfied with the support we got from the project. It has given us adequate 
training on concentrate feed and improved forages. We are also very happy as we are practically 
using the training we got from the project,’ farmers that participated in one of the FGDs said. This 
opinion was shared by beneficiaries across project intervention areas. As shown in Figure 11 below,
93% of direct beneficiary farmers were satisfied or very satisfied with the project support. But 
beneficiary farmers’ satisfaction seems to vary from one project intervention to another. Some are 
satisfied with the improved forage related activities of the project while others are happier with the 
improved livestock/poultry and feed management techniques they learned from the project.
Figure 11: Level of satisfaction of direct beneficiary farmers with the project support
73
% of HHs and their level of satisfaction with 
the project support
Neutral
6% Dissatisfied
3%
Satisfied
38%
Very satisfied
53%
Very satisfied Satisfied Neutral Dissatisfied
2.3. Sustainability
Sustainability is about whether the positive outcomes of the action and the flow of benefits are likely 
to continue after the project funding ends. The fact that the project was very aligned with the 
government priorities implies that its outcomes and initiatives are more likely to be sustained. 
Project activities are well aligned with the national as well as the regional government’s policies, 
strategies and initiatives for the livestock sector of the national economic development program. The 
regional livestock development initiative recognizes animal feed improvement both in quality and 
quantity as the primary area of development intervention. There is acute feed shortage in the regions 
as determined by feed balance surveys. The regions have long standing policy to reduce free grazing 
of livestock in favor of on-farm feeding with improved quality of feedstuff thereby reducing the 
intensity of environmental degradation. Therefore, the activities of the FEED II project are in support 
of these policies and initiatives of the regional government. The alignment has also helped 
government officials own the project interventions. Intervention woredas have been providing 
necessary support to nursery sites established with the support of the project and many continued 
supporting the sites by even allocating budget. Some have already started replicating the project 
initiatives in non-project kebeles. Experience sharing sessions among model farmers have also been 
recorded in videos, with the aim to use it to train other farmers in the future.
The project is also well aligned with the priority needs of the farmers. With the current shrinking 
grazing resource, farmers are showing attitudinal change in a favorable way, which implicates an 
interest in better quality feeds that can be fed on-farm. The project also built capacity of government 
experts to provide necessary support to smallholder farmers. The trainings given to government 
experts will contribute to sustainability. The development of training support materials like videos
74
by FEED II will go a long way to pursue capacity building effort by the extension service and possibly 
by the feed plants themselves to sustain building capacity beyond the project.
Most stakeholders, including farmers and government stakeholders, identified improved feeding 
techniques and use of feeder troughs; awareness created on production and use of improved forage; 
improved dairy, poultry and fattening practices; and manufacturing of compound feed and continued 
use of manufactured feeds as most likely to be sustainable and scalable across regions. Knowledge 
transferred to farmers on improved forage production and use, and other technologies promoted by 
the project could also be equally sustainable as the farmers start practicing what they have learned.
‘The community has a good understanding about animal nutrition, so better livestock nutrition 
management will be sustained even if the project ends. In addition, farmers realize the benefit 
of concentrate feeds, especially to dairy and fattening. Therefore, improved fattening and dairy 
feeding systems also will be sustained,’ a woreda agriculture officer said.
‘Some activities are easy to adopt and implement and brought more return. From FEED II 
project livestock nutrition management, improved feeding systems, improved fattening and 
dairy are well understood by the community. The community can sustain these activities.’ 
another woreda government official said.
Among project activities and outcomes, concentrate feed production and supply through union￾based feed enterprises is the most sustainable and innovative, according to most people consulted 
for the final evaluation. Unions expressed that they will pursue the feed manufacturing operations 
post project. Many have shown this commitment through concrete steps of additional investments 
already in place and plans to do so. Some even intend to expand the operations beyond the current 
status by resolving the problems such as input supply shortages, power interruptions that limit 
production to fully utilize feed production capacity. A few expressed intentions to upgrade the 
current production capacity to produce more feed and a union like the Sidama Elto has already done 
so. Moreover, many unions also have the financial capacity and experience in running similar
business enterprises to not only sustain but also expand the feed manufacturing operations.
The progressive increase in demand and income from enterprises have encouraged many unions to 
make further investments and give more attention to feed production. Some are working to establish 
complementary enterprises like flour mills that help to solve the problem of input supply. Unions like 
the Sidama Elto, are, for example, planning to establish complementary livestock production 
enterprises and a small factory that produces various sizes of packaging materials. The Bokra Union 
is engaged in livestock, poultry and dairy production as complementary enterprises and is about to 
establish a flour mill as part of an expansion plan.
The unions are also taking advantage of their member primary cooperatives to distribute their 
concentrate feed produces to farmers. The integrated approach of the project is also instrumental in 
sustenance of the concentrate feed manufacturing initiative. The project’s approach covered the 
whole value chain ranging from building local capacity to manufacture feed mills and supply 
ingredients to linking input suppliers and output buyers with feed manufacturing enterprises to 
supporting livestock enterprises and creating awareness among farmers about the feed. Strong work
75
has been undertaken on both the demand and supply side that could sustain the project initiative and 
outcomes.
However, sustainability of some project activities and results may vary among the project regions. 
Nursery sites in Amhara region are being taken over by the regional government. The region has also 
started allocating budget to the nursery sites. When asked about this, one woreda official in the 
region said
“FEED II has already stopped its support. Currently, livestock office takes all responsibility to 
manage the nursery site. So far, the government office is able and willing to sustain the nursery 
site.”
In contrast, forage development was considered least sustainable in Oromia region. This is attributed 
to many factors; shortage of budget at woreda agriculture offices did not allow them to allocate funds 
to keep the nursery sites established by FEED II. Nursery sites visited by the field team were not 
functional at the time of the visit. Indeed, government officers indicated that the nursery sites have 
not been in operation during the last two years, which was the time the FEED II project has stopped 
providing funding for the sites. Stakeholders seem to agree that nursery sites would only be 
sustainable if the local government allocates budget to run them. Lack of clear ownership of nursery 
sites was also mentioned as another factor that challenges sustainability of the nursery sites. They 
are often established under unions that are supposed to work with woreda agriculture offices to 
manage them. However, there is no strong working relationship between unions and woreda 
agriculture/livestock offices. Unions are not also said to be that involved in the management of the 
nursery sites. These suggest that sustained access to improved forage seed and seedlings in the 
region would be challenged.
“The established nursery site has not been functional in the last two years because of lack of 
budget. They (FEED II) stopped funding the nursery site before the project phase-out. The 
annual budget allocated for the office can’t sustain the nursery site. There is also poor relation 
between union and livestock and fishery development office. But forage development can easily 
scale up if budget allocated.” A woreda government official in one of the project’s woredas in 
Oromia region said.
Farmers also consistently identified lack and/or unaffordability of improved forage seed/seedlings 
as a threat to the sustenance of growing improved forage in their plots. Although farmers are made 
aware of the value of improved forage, some were not also able to grow them as much as they wanted 
due to drought. The climatic condition of some of the project woredas was not that suitable for forage
production without irrigation. Some improved forages also failed to germinate in some project areas. 
Among forage varieties promoted by the project, planting of improved forage crops especially Rhodes 
grass (Chloris gayana) was considered relatively more sustainable than others.
Although most of the project outcomes and activities are more likely to be sustainable, there are some 
risks that may challenge their continuation. Most of these risks were also identified in the mid-term 
review of FEED II. The costly nature of inputs used to manufacture concentrate feed has remained to
76
be a major challenge for unions to produce as much as they wanted and reduce the cost of feed to 
farmers. Some are trying to deal with the challenge by producing some inputs in house.
Others, and particularly those in Tigray region, are still struggling to find ingredients locally. 
Although the project did a great job in linking feed manufacturing enterprises with ingredient 
suppliers, some are still transporting important ingredients from Addis Ababa, which increases their 
cost of production and reduces competitiveness of their feed product. The price of concentrate feed 
was also considered not affordable to many farmers. This was mainly associated with the use of such 
improved feed on less productive local breeds and limited access to central markets to sell livestock 
products at better prices. Farmers also lack capital to adopt certain technologies such as purchasing 
cross-bred dairy cows (which nowadays cost birr 60,000 – 80,000) that respond favorably to feeding 
of improved forage crops introduced by the project. This was also mentioned as a risk to continued 
adoption of improved feeds and practices promoted by the project.
There are also some risks associated with dealers of manufactured feed. Some dealers of feed are 
shifting to selling raw materials such as noug cake rather than compounded feed mainly because of 
limited promotion by union, misunderstanding on feed quality, and unattractive return from sale of 
union-manufactured feed. Dealers find more profit from selling ingredients than concentrate feed as 
demand for ingredients are high due to relatively cheaper prices. Additionally, the distribution 
channel established by feed manufacturing enterprises is still very weak. Unions are still not 
adequately utilizing their member primary cooperatives to distribute their feed and make them 
accessible to farmers. In some areas, farmers buy the feed directly from feed manufacturing plants, 
which were found to have bureaucratic processes to manage sales.
Drought and shortage of land have been challenging the sustainability of improved forage production 
in many areas. In some areas, the drought did not allow farmers to cultivate improved forage on their 
land. The perennial grass could not resist shortage of water during the dry season. Most farmers also 
own land that is too small to grow improved forage. As a result, farmers in these areas were not able 
to observe changes that could be brought about by feeding improved forage to their livestock. Many 
farmers have also identified unavailability of improved forage seed as a major problem to continued 
cultivation of forage in their land. Those that were also able to source the seeds in the local market 
also found it unaffordable for them to buy. Like concentrate feed, many stakeholders also indicated 
that use of improved forage on local breeds would not be cost effective as their productivity is very 
low. The fact that some nursery sites have ceased to function could also mean that sustained access 
to forage seed supply will be curtailed. Although there are regional variations, some nursery sites 
were not operational during the field visit due to lack of budget. Some particularly in Oromia region 
ceased their operation immediately after the project stopped its funding. Hence, many stakeholders 
question whether nursery sites will continue in the absence of a clear owner and continued project 
support. In sum, interruption of forage seed supply which is likely when the project ends will be a 
major risk to sustainable production and use of improved forage.
The fact that the project was implemented through the existing government structure and staff 
created strong sense of ownership, which is also instrumental for sustainability of project outcomes.
77
However, the government staff are not providing the required follow-up and support to smallholder 
farmers. Some training sessions particularly on improved forage were not also followed by provision 
of seed/seedlings. To many farmers, the provision of training and materials may not guarantee 
achievement of desired results unless the project ensures that the farmers have put the knowledge 
and technologies transferred through the project into practice. These limitations have hindered 
adoption of some improved practices and technologies by some farmers.
Limited access to markets might also have negative impact on sustainability of some project 
initiatives. Many farmers that were able to improve their livestock productivity and production with 
the use of improved forage were unable to sell their livestock products at better prices. The project’s 
work on livestock enterprises was an initiative in the right direction. But the scope of the initiative 
was not found to be large enough to create access to market to livestock farmers.
2.4. Cross-Cutting Issues
The FEED II project took gender as a cross-cutting issue. It aimed to ensure that at least 30% of its 
beneficiaries are women. Gender disaggregated data has also been gathered to track progress 
towards the target. Project progress reports show that 34% of the 43,607 direct beneficiary 
households were women. Analysis of survey data also suggests that some of the responsibilities that 
used to be left for women are increasingly getting shared by men, and women are also increasingly 
involved in some livestock and poultry activities that used to be left for men. Although key household 
decisions such as livestock feed purchasing and selling of cows and calves are still mainly made by 
men, the findings showed that some activities such as feed collection and feeding that used to be 
undertaken by men or women alone have become more of a joint responsibility of men and women. 
A relatively similar pattern was also observed in poultry management. In sum, many responsibilities 
that used to belong to either men or women alone are increasingly becoming joint responsibilities,
which is a positive development towards gender equality.
3. LESSONS LEARNED, CONCLUSIONSAND 
RECOMMENDATIONS
i. The establishment of forage nursery sites under cooperative unions while also ensuring 
woreda agriculture/livestock office provide technical support and provide land could be 
considered an appropriate strategy to run the sites. This strategy worked well in most cases. 
However, unions are not as involved in the nursery site management and seem to lack 
ownership. This is partly because nursery sites in most cases are currently providing services 
free of charge to farmers and does not generate income for unions. This is a disincentive for 
union management to get involved in the management of nurseries. The project also tried to 
transfer the ownership of nursery sites to government and/or unions depending on their 
level of involvement and ownership. Governments in some regions have allocated budget to 
run the nursery sites. While this makes sense, it is likely that nursery sites managed by these
78
entities will be challenged by inadequate financial capacity of local governments or lack of 
commitment from officials. The lesson that might be taken here is probably the need to run 
nursery sites as commercial entities under the ownership of private enterprises that will be 
fully committed to sustain and keep them operational.
ii. Establishing forage seed multiplication nurseries at strategic locations and entities plays 
pivotal role in sustained supply of new forage varieties. Nursery sites established in Tigray 
region were not successful. That was said to relate to the entities under which they were 
established. Unlike other regions in which nurseries were established under cooperative 
unions, nursery sites in Tigray were established under primary cooperatives that were 
established by unemployed youth upon the request of the regional government. These 
primary cooperatives were, however, not that strong and able to sustain the nursery sites. As 
a result, three nursery sites established in the region have failed. While the need to establish 
nursery sites as commercial entities is strongly suggested, it can also be inferred that 
attaching nurseries to more stable organizations such as cooperative unions is vital.
iii. The project was considered much different from other projects in the practical training and 
support it provided. Stakeholders were impressed with the behavioral change brought about 
among farmers following the training. Beneficiaries put what they learned into practice and 
achieved greater results. Hence, though it sounds a general truth, the lesson drawn here is 
the learning that it was ‘possible to change the attitude of the farmer through practical 
training from the traditional beliefs such as keeping many unproductive animals towards 
keeping lesser number of productive animals commensurate with available feed resource,’ as 
one government official put it.
viii. The implementation of integrated initiatives made the project very successful in achieving its 
targets and helping target groups improve their livelihood. Even so, it was learned that 
increasing livestock and poultry productivity through improved forage and feed cannot be a 
standalone objective unless the intervention includes or partners with initiatives that 
improve access to market and improved breeds. FEED III may need to work more on 
improving access to improved breeds and market to sell products by including these 
activities in its package or partnering with other organizations that work on artificial 
insemination (AI), animal health, and markets. There is a need to consider linking the feed 
distribution and farmer capacity building effort with areas where such complementary 
interventions like breed improvement (AI service) and health interventions to benefit from 
enhanced benefits from the combined interventions. More work is also needed in the future 
to create awareness among farmers about manufactured feed to create demand for these 
products while at the same time improve their livestock and poultry productivity.
iv. Establishment of feed manufacturing enterprises under cooperative unions was an effective 
strategy that can be emulated elsewhere. This strategy worked well in improving accessibility 
of manufactured feed by using a network of member primary cooperatives that are close to 
smallholder farmers. Being farmer owned, unions are also discharging their membership and 
social goals by producing quality feed and selling them at prices lower than those charged by 
traders.
v. Price of technology promoted by the project and its cost-effectiveness was found to have 
significant influence in adoption of the technologies. Many farmers and stakeholders agreed 
that the current price of manufactured feed is too high and less cost effective for farmers that 
largely herd local breeds which are less responsive to improved feed. As a result of the high
79
price, farmers are resorting to other options. Hence, efforts need to be exerted in the future 
to make price of concentrates more affordable and cost effective by supporting feed
enterprises to increase their capacity utilization and expansion, and implementing strategies 
to reduce prices of ingredients. Among others, this might include the need to consider 
integration of feed manufacturing with other complementary factories like oil and flour mills 
to be established where feasible/appropriate. In connection with this, it may be important to 
explore/promote the possibility of union member PCs to engage in the production of feed 
ingredients like maize, soybean, etc. possibly based on a contractual agreement with the feed 
plant where feasible/appropriate;
vi. Government bodies are very much interested in scaling up feed manufacturing and nursery 
sites in the future. But it is likely that they may be constrained by financial and technical 
capacity limitations. Hence, future interventions are advised to support scaling up of the 
project initiatives inside and outside the FEED II project interventions. Part of this effort may 
focus on documenting and disseminating the project activities and achievements to relevant 
policy makers and other executive organs. Future interventions are also recommended to 
create mechanisms to link existing feed manufacturing enterprises and nursery sites to 
woredas with no feed manufacturing enterprise or nursery sites.
vii. The project catalyzed increased private sector investment in manufactured feed production.
While this is a positive development in terms of increasing supply and ultimately reducing 
price of such feed, it also brings challenges along with it particularly for union-based feed
manufacturing plants established by the project. The increased private sector investment will
make the market very competitive, which might not be easy for unions to win. Hence, there 
is a need for continued support to unions to enhance the capacity of their management to be 
able to compete with the growing private investors that are expected to be flexible and 
market oriented while at the same time more efficient with increased production and 
economies of scale.
viii. The evaluation study design did not allow adequate attribution of outcomes to the project 
intervention. This was because there was no control group at baseline and endline. Hence, 
future interventions may need to consider the use of quasi-experimental evaluation design 
to better measure achievements and draw conclusions.
vi. The adaptation and poor germination problems of some of the forage varieties observed in 
some of the study area suggests that suitable varieties (make adaptation trial first) and 
technologies needs to be carefully selected before popularization to beneficiaries.
ix. Free supply of seeds by NGOs for farmers could create dependence at farmers level and 
farmers becoming reluctant to buy seeds and seedling which also the concern raised by the 
respondents and needs to be considered.
x. For forage production, good quality land with better fertility, access to irrigation and other 
facilities such as road and proximity to market are required. Investment in seed production 
business may need access to finance. But the major challenge in forage seed multiplication is 
lack of information on the future demand of seeds and planting materials. Alternatives are 
possible to meet the high demand for forage seeds/seedlings: 1) expand the nursery 
operation in other places with abundant space and irrigable water 2) out-source/ engage 
farmers in seed production with technical backstopping and quality control, 3) increase
80
awareness of farmers about forage seeds, splits and cuttings is better way to sustain the 
nursery sites, 4) ensure access to finance.
xi. Government bodies need to allocate sufficient budget to nursery site management and follow 
up of other interventions. Unions should also own the nursery sites that are established under 
them and work closely with government to operate them.
xii. Strengthen the capacity of primary cooperatives so that they can distribute formulated feeds 
to their members and beyond; and engage agro-dealers more intensively in the distribution 
system.
xiii. Foster contractual agreements with input suppliers to avoid intermittent supply of inputs.
Foster linkages with financial institutions to help cater for the capital limitations expressed
by some feed plants to expand to meet increasing demand.
xiv. Provide support to ease the bureaucratic operations of the procurement, distribution and 
routine management through promoting more autonomy to the feed plants commensurate 
with the growth in their operations to foster faster decision making which may include, but 
not limited to, revisiting or modifying some procurement rule and regulation of the union to 
allow the feed plant to buy inputs on its own; and assigning professional managers, 
accountant and other staff dedicated to the feed manufacturing operation as warranted by 
the level of development of the feed plant.
xv. Adequate dedicated storage both at the factory and distribution sites has been a serious 
limitation. Fostering expansion of storage capacity will substantially improve the volume of 
input purchase at appropriate times, continuous availability of formulated feed at points 
demanded.
81
ANNEXES
82
Annex 1) FEED II Project’s Achievement on Outcome Indicators and Targets
Outcome-1: Increased
Agricultural Productivity
Definition
Indicator￾1
Percent increase 
in unit
productivity of
fattened cattle
Percentage increase in productivity of fattened cattle per head over the 
measured baseline.
Unit: Percent
Total Percentage
Cattle Fattening
Baseline Endline Change
Total cattle fattened 1,293 1709 32%
Cattle fattened per 100 households 12.93
15
17.1 32%
Indicator￾2
Percent increase 
in unit
productivity of 
poultry
Percentage increase in productivity of local and exotic/ improved poultry, by 
counting the total number of poultry (all types) raised over baseline.
Unit: percent
Average number of poultry
owned by a household by 
breed
Baseline Endline % Change
Local 5.8 3 -48%
Improved 3.1 7.08 128%
Total number of chickens 
owned per household 8.9
9.98 12%
Indicator￾3
Percent increase
in unit 
productivity of 
milk
This indicator will measure the percentage increase in productivity of milk,
by counting liters of milk produced and calculating the percentage increase 
over the measured baseline.
Unit: Percent
Milk Productivity Baseline Endline % Change
Average maximum amount 
of milk collected 
(liters/cow/ day)
2.95 6.02 104%
Average Minimum Amount 
of milk collected 
(liters/cow/ day)
1.42 3.6 154%
15 The baseline report stated 17.2 instead of 12.93
83
Average milk collected
(liters/cow/ day)
Outcome-2: Expanded Trade of Agricultural Products
2.185 4.81
120%
Indicator￾1
Percent increase 
in sales of
agricultural 
products
(fattened cattle, 
poultry, feed,
milk/milk 
products and
eggs) by project 
beneficiaries
This indicator will calculate the gross sales of these five products by 
beneficiaries at various points over the life of project to compare.
Unit: Percent
Sales items Baseline Endline
Total birr collected from sale of fattened oxen 8,356,726 20,934,293
Total sales collected from sale of fattened cow
(birr) 940,868 1,997,148
Total sales collected from milk sale (birr) 1,168,042 7,412,958
Total sales collected from butter sale, birr 1,277,316 3,988,718
Total sales collected from sale of pullets (birr) 413,207 178,294
Total sales collected from sale of broilers (birr) 190,093 275449
Total sales collected from sale of spent hens
(birr) 55,093.80 385,123
Sale value of eggs 281,802 4,184,997
Total 12,683,148 39,356,980
Indicator￾2
Return on 
investment from
livestock feed
related activities
Sales per year per household 6,626.51
16 20,562.69
% change 210%
This indicator will calculate the change in gross revenues as a percent of 
expenditures associated with a commercial activity.
Unit: Percent
Investment category RoI 
Feed Manufacturing 23.3
Livestock and Poultry
Enterprises
Average RoI
26
24.65
16 The baseline report reported ETB 9,744.14 as the value of products/year/ HH (birr) while the evaluation 
team computed ETB 6,626.51 from the baseline dataset.
84
Annex 2) References
Alemu Yami and Seyoum Bediye. 2019. Training Manual on Implementation of Ethiopian Animal Feed
Standards. VDFACA/International Finance Corporation. 50Pp.
Azage Tegegne, Berhanu Gebremedhin and Hoekstra D., 2010. Livestock input supply and service 
provision in Ethiopia: Challenges and opportunities for market oriented development. IPMS 
(Improving Productivity and Market Success) of Ethiopian Farmers Project Working Paper 20. 
ILRI (International Livestock Research Institute), Nairobi, Kenya. 48 pp.
Demissie Negash. 2017. Review on Compound Animal Feed Processing in Ethiopia: Condition, 
Challenges, and Opportunities. J Nutri Health. 2017;3(2): 5
FAO,2018. National and Regional Feed Inventory and Feed Balances in Ethiopia. FAO Country Office, 
Addis Ababa, Ethiopia.
Federal Democratic Republic of Ethiopia, National Planning Commission (NPC). 2016. Growth and
Transformation Plan II (GTP II) (2015/16-2019/20); Volume I: Main Text
FEED II project. 2014. Baseline Report. ACDI/VOCA. 81Pp.
FEED II.2018. Year V second semi-annual report (April-September 2018).
International Livestock Research Institute (ILRI). 2015. Ethiopia Livestock Master Plan: Roadmaps for
Growth and Transformation (2015-2020)
Mesfin Dejene, Getnet Assefa, Gezahegn Kebede and Karta Kaske. 2012. Forage Seed Production and 
Supply Systems in the Central Highlands of Ethiopia. In: Getnet Assefa, Mesfin Dejene, J. Hanson, 
Getachew, A., Solomon M and Alemayehu M. (Eds). Forage Seed Research and Development in 
Ethiopia. pp 59-70. ISBN: 978-99944-53-84-9. EIAR, Addis Ababa, Ethiopia.
Mesfin Dejene, Seyoum Bediye, Dawit Alemu, Getu Kitaw, Aemiro Kehaliw, Getnet Assefa and Getaw 
Tadesse. 2014. Livestock Feed Marketing in Ethiopia: Challenges and Opportunities for Livestock 
Development. Journal of Agricultural Science and Technology A 4 (2014) 155-168.
Seyoum Bediye. 2017. The Status of the Commercial Feed Sub-Sector in Ethiopia. Feed Bulletin No.007/
2017. Ethiopian Animal Feed Industry Association, Addis Ababa, Ethiopia. Pp. 39.
African Business Council N. (2010) Livestock in Ethiopia and opportunity analyses for Dutch investment.
Netherlands: pp1-27.
CSA. (2018) Agricultural Sample Survey, Report on livestock and livestock characteristics of private 
peasant holdings for the year 2017/18, Volume II Federal democratic republic of Athiopia, 
Central Statistical Agency (CSA), Addis Ababa, Ethiopia.
Delgado C., Rosegrant M., Steinfeld H., Ehui S., Courbois C. (1999) Livestock to 2020: the next food 
revolution., Washington, DC: International Food Policy Research Institute, Food, Agriculture, and 
the Environment Discussion Paper No. 28, International Food Policy Research Institute, 
Washington, DC:. pp. 83.
Diao X., Fekadu B., Haggblade S., Taffesse A.S., Wamisho K., Yu B. (2007) Agricultural growth linkages in 
Ethiopia: Estimates using fixed and flexible price models International Food Policy Research 
Institute Washington, DC.
FAO. (2018) Ethiopia: Report on feed inventory and feed balance, 2018. Italy. 160 pages. Licence: CC BY￾NC-SA 3.0 IGO.
Farm-Africa. (2017), https://www.farmafrica.org/ethiopia/ethiopia, Farm Africa. Accessed on 19
September 2017.
85
Gebremedhin B., Fernandez-Rivera S., Hassena M., Mwangi W., Ahmed S. (2007) Maize and livestock: 
Their inter-linked roles in meeting human needs in Ethiopia, ILRI Research Report 6. ILRI 
(International Livestock Research Institute), Nairobi, Kenya. 103 pp.
Gebremedhin B., Hirpa A., Berhe K. (2009) An appraisal of fodder marketing in Ethiopia, Proceedings of 
the 16th Annual Conference of the Ethiopian Society of Animal Production (ESAP) on 
Commercialization of Livestock Agriculture Addis Ababa, Ethiopia pp. Part I, pp. 1-20.
Halderman M. (2004) The political economy of poor-poor livestock policy-making in Ethiopia, Poor-Poor
Livestock Policy Initiatives, FAO, Rome, Italy.
Herrero M., Grace D., Njuki J., Johnson N., Enahoro D., Silvestri S., Rufino M.C. (2013) The roles of 
livestock in developing countries. Animal 7:3-18.
Jackson M. (1978) Treating straw for animal feeding. An assessment of its technical and economic 
feasibility, FAO Animal production and health paper 10, FAO, Rome, Italy.
McDermott J., Staal S., Freeman H., Herrero M., Van de Steeg J. (2010) Sustaining intensification of 
smallholder livestock systems in the tropics. Livestock science 130:95-109.
Negassa A., Rashid S., Gebremedhin B. (2011) Livestock production and marketing, International Food
Policy Research Institute (IFPRI) – Ethiopia Strategy Support Program II (ESSP II), Working Paper
26. IFPRI, Addis Ababa, Ethiopia. pp. 35.
NPC. (2016) Growth and Transformation Plan II (GTP II) (2016). Federal Democratic Republic of Ethiopia, 
GTP II (2015/16-2019/20), Volume I. National Planning Comission (NPC), May, 2016, Addis 
Ababa, Ethiopia. 236p.
Reddy B., Reddy P.S., Bidinger F., Blümmel M. (2003) Crop management factors influencing yield and 
quality of crop residues. Field Crops Research 84:57-77.
Schiere J., Joshi A., Seetharam A., Oosting S., Goodchild A., Deinum B., Van Keulen H. (2004) Grain and 
straw for whole plant value: implications for crop management and genetic improvement 
strategies. Experimental Agriculture 40:277-294.
Schiere J., Nell A. (1993) Feeding of urea treated straw in the tropics. I. A review of its technical 
principles and economics. Animal Feed Science and Technology 43:135-147.
Schiere J.B., Ibrahim M.N.M. (1989) Feeding of Urea-ammonia Treated Rice Straw Pudoc, Wageningen, 
the Netherlands, 125pp.
Shapiro B.I., Gebru G., Desta S., Negassa A., Nigussie K., Aboset G., Mechale H. (2015) Ethiopia livestock 
master plan. ILRI Project Report. Nairobi, Kenya: International Livestock Research Institute (ILRI).
Shapiro B.I., Gebru G., Desta S., Negassa A., Nigussie K., Aboset G., H. M. (2017) Ethiopia livestock sector 
analysis. ILRI Project Report. Nairobi, Kenya: International Livestock Research Institute (ILRI).
Stapleton J. (2016) Unlocking the potential of Ethiopia’s livestock sector: growth, jobs and 
environmental sustainability, https://news.ilri.org/2016/02/04/unlocking-the-potential-of-the￾livestock-sector-ethiopia/.
Steinfeld H., Gerber P., Wassenaar T., Castel V., de Haan C. (2006) Livestock's long shadow:
environmental issues and options Food and Agriculture Organisation., Nairobi, Kenya. Sundstol F.
(1985) Procedures for research into the treatment of crop residues and agro-industrial by￾products in developing countries, in: T. R. Preston, Kossila, V.L., Goodwin, J., Reed, S.B. (Ed.), 
Better utilization of crop residues and by-products in animal feeding: research guidelines. 1. 
State of knowledge. , FAO, ILCA Headquarters, Addis Ababa, Ethiopia.
World-Bank. (2009) Minding the stock: bringing public policy to bear on livestock sector development. 
World Bank. 2017. Livestock and Fisheries Sector Development Project (P159382)
86
Annex 3) Sample Beneficiary Households’ Sex, Age, Education Status, Family Size and
Membership in Cooperatives Disaggregated by Region
Tigray Amhara Oromia SNNP Total
N % N % N % N % N %
Gender of
the 
household 
head
Male 388a,b 83 422a 88 405a 87 385b 80 1600 85
Female 79a,b 17 58a 12 59a 13 98b 20 294 14
Gender of
the 
respondent
Male 359a 77 373a 78 297b 64 311b 64 1340 71
Female 108a 23 107a 22 167b 36 172b 36 554 29
Age of the
household 
head
Less than
30 years
32a 7 21a,b 4 12b 3 20a,b 4 85 5
30 – 40
years
127a 27 137a 29 154a 33 159a 33 577 31
41 - 50 
years
138a 30 192b 40 151a,b 33 163a,b 34 644 34
Over 50
years
170a 36 130b 27 147a,b 32 141a,b 29 588 31
Education
status of 
the Head
None 173a 37 142a,b 30 111b,c 24 97c 20 523 28
Read &
Write
53a 11 187b 39 54a 12 139c 29 433 23
Primary 197a 42 127b 27 178a 38 181a 38 683 36
Secondary 39a,c 8 23a 5 100b 22 54c 11 216 11
Above 
secondary
5a,c 1 1a 0 21b 5 12b,c 3 39 2
Membership
in 
cooperatives
Member 301a 65 377b 79 373b 80 297a 62 1348 71
Non-member 166a 36 103b 22 91b 20 186a 39 546 29
Cooperation
Type
Livestock 17a 6 2b 1 153c 41 71d 24 243 18
Multipurpose 278a 92 364a 97 186b 50 217c 73 1045 78
Both 6a 2 11a 3 34b 9 9a 3 60 5
Mean N Mean N Mean N Mean N Mean N
Average 
Household Size 
of Households 
Disaggregated 
by Region
Male 3.15a 467 3.14a 480 3.30a 464 3.96b 483 3.39 1894
Female 2.92a 467 2.92a 480 2.91a 464 3.65b 483 3.10 1894
Total 6.07a 467 6.05a 480 6.21a 464 7.61b 483 6.49 1894
Note: Values in the same row and subtable not sharing the same subscript are significantly different at p< .05
87
Annex 4) Percentage of Households and their Ownership of Mature Cows by Size of Holdings, Region and Breed Type 
at Baseline and Endline
Regions Amhara Oromia SNNPR Tigray Total
N
Baseline 478 480 477 479 1914
End line 480 464 483 467 1894
Mature 
cows Term Indigenous Crossbreed Indigenous Crossbreed Indigenous Crossbreed Indigenous Crossbreed Indigenous Crossbreed
No mature 
cow
Baseline 22.8 84.1 19 67.9 23.3 74.6 20 83.3 21.3 77.5
Endline 15.2 67.1 34.7 50.9 26.7 59.4 35.3 70.2 27.9 61.9
1-2 head Baseline 57.1 14.9 62.1 24.2 65.2 23.5 65.3 15.2 62.4 19.4
Endline 53.8 31.5 48.3 32.8 54.0 31.5 45.8 23.6 50.5 29.8
3-4 head Baseline 16.5 1 16.3 6.9 9.4 1.7 9.8 0.8 13 2.6
Endline 26.9 1.3 11.0 11.6 12.6 8.1 12.0 4.9 15.7 6.4
5-9 head Baseline 3.6 - 2.5 0.8 1.7 0.2 3.5 0.6 2.8 0.4
Endline 4.0 0.2 5.6 4.1 5.8 1.0 4.7 0.9 5.0 1.5
10-19 head Baseline - - 0.2 0.2 0.4 1.3 0.5 0.1
Endline 0.2 0.0 0.4 0.6 0.8 0.0 1.5 0.2 0.7 0.2
20 and 
more
Baseline - - - - - - - - - -
Endline - - - - - - 0.6 0.2 0.2 0.1
Cow count
(number)
Baseline 754 110 752 340 636 174 729 127 2871 751
Endline 942 213 660 551 785 357 794 302 3181 1423
Mean cow
holding
Baseline 1.57 0.2 1.56 0.7 1.3 0.4 1.7 0.3 1.5 0.4
Endline 1.96 0.44 1.42 1.19 1.63 0.74 1.70 0.65 1.68 0.75
88
Annex 5) Percentage of Households by Size of Bull/Ox Holding and Region at Baseline and Endline
Regions Amhara Oromia SNNPR Tigray Total
N
Baseline 478 480 477 479 1914
End line 480 464 483 467 1894
Heads of
Mature 
bull/ox
Term Indigenous Crossbreed Indigenous Crossbreed Indigenous Crossbreed Indigenous Crossbreed Indigenous Crossbreed
No mature
bull/ox
Baseline 24.9 91.4 17.1 80.4 35.8 93.9 27.8 95.4 26.4 90.3
End line 12.3 84.0 21.8 71.6 37.1 93.2 21.8 84.6 23.3 83.4
1-2 head Baseline 52.1 7.9 51.7 17.1 54.5 5.9 57.6 3.8 54 8.7
End line 54.4 15.4 48.7 22.0 55.1 6.6 49.5 11.1 52.0 13.7
3-4 head Baseline 19.5 0.4 24.4 2.5 6.9 0.2 11.3 0.4 15.5 0.9
End line 27.5 0.6 23.1 5.2 7.0 0.2 19.5 3.4 19.2 2.3
5-9 head Baseline 3.1 0.2 6.5 - 2.5 - 2.1 0.4 3.6 0.2
End line 5.6 - 6.5 1.1 0.6 - 7.1 0.9 4.9 0.5
10-19 head Baseline 0.4 0.4 0.2 0.8 0.5 -
End line 0.2 - - 0.2 0.2 - 2.1 - 0.6 0.1
20 and
more
Baseline - - - - - - 0.4 - 0.1 -
End line - - - - - - - - - -
Bull/Ox 
count
Baseline 803 56 1020 157 555 37 778 42 3156 292
End line 1054 93 935 258 537 38 1000 150 3526 539
Mean
bull/ox 
holding
Baseline 1.7 0.1 2.1 0.3 1.2 0.1 1.6 0.1 1.6 0.2
End line 2.20 0.19 2.02 0.56 1.11 0.08 2.14 0.32 1.86 0.28
89
Annex 6) Percentage of respondents who have access the different feed items in terms of how far is the nearest 
distribution point
Regions
Amhara Oromia SNNPR Tigray
How far is the 
nearest distribution 
point the feed
items? <16 km
16-30 
km
30 km 
or 
more
No 
distribution 
point in the 
area
I do 
not 
know
<16 
km
16-30 
km
30 km 
or 
more
No 
distribution 
point in the 
area
I do 
not 
know
<16 
km
16-30 
km
30 km 
or 
more
No 
distribution 
point in the 
area
I do 
not 
know
<16 
km
16-30 
km
30 km 
or 
more
No 
distribution 
point in the 
area
I d 
no 
kno
FEED ITE
MS
Concentrates
(factory
formulated 
and mixed 
feed)
6.5a,b 25.6a 10.3a,b 4.7a,b 5.5b 52.7a 23.8a,b 44.0a,b 34.6a,b 25.6b 64.9a,b 42.9a 44.4a,b 53.5a,b 78.9b 43.7a 43.0a 37.8a,b 11.8b 40.0
Agro￾industrial by￾products 
(wheat bran,
oilseed cake, 
molasses, 
brewers 
grain, etc)
45.7a,c,d 23.1a,b,e 52.6c 25.6b,d,e 15.4e 24.2a 47.6a 24.0a 46.2a 17.9a 6.7a 10.7a 11.1a 3.0a .0
1 31.9a 12.8b 29.7a,b 67.6c 6.7
Other high
quality feeds
(atela, grains
)
82.6a 94.9a 89.2a 76.7a 81.3a 79.5a 61.9a 76.0a 61.5a 79.5a 16.3a 10.7a,b 44.4b 18.2a,b 26.3a,b 86.7a 80.2a,b 73.0a,b 88.2a 46.
Green
forage, 
grass, hay, 
silage
93.5a 97.4a 71.1b 97.7a 100.0
1 86.4a 85.7a 96.0a 84.6a 87.2a 96.6a 100.0
1 100.0
1 91.9a 100.0
1 100.0
1 96.5a 94.6a 100.0
1 100.
Crop Residue
(teff straw, 
wheatstraw,
maize 
stover)
93.5a 94.9a 93.8a 97.7a 100.0
1 95.8a 100.0
1 88.0a 96.2a 92.3a 98.6a 89.3b 100.0
1 94.9a,b 100.0
1 99.3a 94.2b 100.0
1 100.0
1 100.
Note: Values in the same row and sub table notsharing the same subscript are significantly different at p< .05 in the two-sided test of equality for column proportions. Cells with no subscript are not included in the test. Tests assume equal v
1. This category is not used in comparisons because its column proportion is equal to zero or one.
2. Tests are adjusted for all pair wise comparisons within a row of each innermostsub table using the Bonferroni correction.
17 The evaluation team chose to not to identify enterprises by name as the financial data are confidential.
90
Annex 7) Return on Investment and Gross Margin of Feed Manufacturing Enterprises
FCU17 Total 2016 2017 2018
investment Sales Total cost Return ROI 
(%)
GM 
(%)
Sales Total cost Return ROI 
(%)
GM 
(%)
Sales Total cost Return ROI 
(%)
GM 
(%)
Enterprise 1 1,549,288 1,657,425 1,614,080 89,362 5.8 2.6 2,169,135 2,072,063 97,072 6.3 4.5 3,357,250 3,207,809 149,441 9.6 4.5
Enterprise 2 1,502,955 3,230,789 2,829,048 401,741 26.7 12.4 1,498,458 1,453,970 44,488 3.0 3.0 3,322,644 2,433,905 888,739 59.1 26.7
Enterprise 3 1,549,288 1,996,220 1,782,838 213,382 13.8 10.7 379,421 523,098 -143,677 -9.3 -
37.9
Enterprise 4 1,765,140 5,578,214 5,126,801 451,412 25.6 8.1
Enterprise 5 1,552,947 1,374,733 1,298,443 76,443 4.9 5.5 2,339,258 2,134,801 204,457 13.2 8.7
Enterprise 6 1,556,002 1,181,430 1,009,158 172,272 11.1 14.6 1,876,698 1,700,282 176,416 11.3 9.4
Enterprise 7 1,725,878 2,595,967 2,416,780 179,187 10.4 6.9
Enterprise 8 1,562,529 2,537,826 2,264,625 273,201 17.5 10.8 3,299,510 2,812,774 486,736 31.2 14.8
Enterprise 9 1,562,529 2,986,328 2,642,189 344,139 22.0 11.5
Enterprise 10 1,617,791 2,018,054 1,886,952 131,102 8.1 6.5 1,818,550 1,700,409 118,141 7.3 6.5
Enterprise 11 1,531,963 1,544,148 1,479,205 64,943 4.2 4.2
Enterprise 12 1,549,254 2,465,974 2,376,437 89,537 5.8 3.6
Total 19,025,565 8,259,167 7,524,409 734,758 11.9 8.9 27,294,211 25,386,079 1,908,132 10.0 7.0 13,674,652 11,855,179 1,819,473 9.6 13.3
Investment
by the project
Investment 
by unions
15,605,565.45
3,420.00
Source: ACDI/VOC FEED II Project
18 The evaluation team chose to not to identify enterprises by name as the financial data are confidential.
91
Annex 8) Return on Investment of Livestock/Poultry/Forage Enterprises
18
YR ITEM
NAME OF ENTERPRSE
G.Total
DAIRY POULTRY FATTNING
Firm 2 Firm 3 Firm 4 Firm 5 Firm 6 Firm 7
Firm 1 Total Firm 8 Firm 9 Firm 10 Total Firm 11 Firm 12 Total
2016
Investment 582,713 785,282 529,175 605,160 704,609 399,059 3,605,998 937,093 816,393 801,525 2,555,010 472,746 - 472,746 6,633,754
CASH IN
(Revenue) 1,024,036 812,355 115,689 40,100 223,468 322,748 2,538,396 554,879 176,400 292,666 1,023,945 339,000 339,000 3,901,341
CASHOUT
(expense) 930,417 729,559 285,631 107,795 127,168 305,807 2,486,375 365,600 139,662 190,657 695,919 290,340 290,340 3,472,634
NET CASH 93,619 82,797 (169,942) (67,695) 96,301 16,942 52,021 189,279 36,738 102,009 328,026 48,660 48,660 428,707
2017
ROI 16% 11% -32% -11% 14% 4.2% 1.4% 20% 5% 13% 13% 10% 10% 6%
New
investment 153,000 128,500 59,463 450,000 1,983,000 64,390 84,381 2,922,734 - - - - - 1,490,000 1,490,000 4,412,734
T￾Investment 735,713 913,782 588,638 1,055,160 1,983,000 768,999 483,441 6,528,732 937,093 816,393 801,525 2,555,010 472,746 1,490,000 1,962,746 11,046,489
CASH IN
(Revenue) 1,521,463 1,453,409 344,415 245,151 1,800,045 223,468 483,636 6,071,586 956,855 275,443 614,000 1,846,298 402,760 437,480 840,240 8,758,124
CASHOUT
(expense) 1,389,358 1,309,140 341,937 292,151 669,100 137,874 466,543 4,606,104 620,715 212,848 351,931 1,185,494 713,254 579,718 1,292,972 7,084,570
NET CASH 132,104 144,269 2,477 (47,000) 1,130,945 85,594 17,093 1,465,482 336,140 62,595 262,069 660,803 (310,494) (142,238) (452,732) 1,673,553
2018
ROI 18% 16% 0% -4% 57% 11% 3.5% 22% 36% 8% 33% 26% -66% -10% -23% 15%
New
investment 117,000 880 117,880 - - 117,880
T￾Investment 2,100,000 484,321 2,584,321 1,490,000 1,490,000 4,074,321
CASH IN
(Revenue) 1,526,550 320,403 1,846,953 284,000 284,000 2,130,953
CASHOUT
(expense) 722,700 258,838 981,538 343,487 343,487 1,325,024
NET CASH 803,850 61,566 865,416 (59,487) (59,487) 805,929
ROI 38% 13% 33% -4% -4% 20%
Sum
Total
Investment 735,713 913,782 588,638 1,055,160 2,100,000 981,665 484,321 6,859,279 937,093 816,393 801,525 2,555,010 472,746 1,490,000 1,962,746 11,377,035
92
CASH IN
(Revenue) 2,545,499 2,265,764 460,104 285,251 3,326,595 446,936 1,126,787 10,456,935 1,511,734 451,843 906,666 2,870,242 741,760 721,480 1,463,240 14,790,418
CASHOUT 
(expense) 2,319,775 2,038,699 627,568 399,946 1,391,800 265,042 1,031,187 8,074,017 986,315 352,510 542,588 1,881,413 1,003,594 923,204 1,926,798 11,882,228
NET CASH 225,723 227,065 (167,465) (114,695) 1,934,795 181,895 95,600 2,382,918 525,419 99,333 364,078 988,829 (261,834) (201,724) (463,558) 2,908,189
ROI 31% 25% -28% -11% 92% 19% 20% 35% 56% 12% 45% 39% -55% -14% -24% 26%
93
Annex 9) List of Key Informants Consulted
Amhara Region
94
95
Oromia Region
96
97
98
SNNP Region
99
100
Tigray Region
101
102
103
Annex 10)List of Farmers that Participated in Group Discussions
S.N. Full Name Sex Kebele Woreda Region
Amhara Region
104
105
Oromia Region
106
107
SNNP Region
108
109
110
Tigray Region
111
112
113
Annex 11)List of beneficiaries on Whom Case Stories were Built
Amhara Region
Oromia Region
SNNP Region
Tigray Region
114
Annex 12) List of Project Woredas and Unions Visited for the Final Evaluation
Region Zone Woreda
Tigray Central Tigray Adwa
Temben
Eastern Tigray Sase Tseda Emba
KeliteAwlalo
Western Tigray Kafta Humera
NW Tigray T Keraro
Southern Tigray Raya Azebo
Ofla
Amhara Awi Dangila
Guagusa
East Gojam Gozamin
Mechakal
South Gonder Farta
Libokemkem
W Gojam Denbecha
Jabitehenan
Oromia West Shewa Dendi
S/W/Shewa Becho
Woliso
Arsi Lode Hetosa
Doddota
Tiyo
N/Shewa Kuyu
Jidda
SNNPR Gamo gofa Arba Minch
Chencha
Gurage Sodo
Hadiya Misha
Kembata Kedida Gamela
Sidama Wondo
Silti Dalocha
Wwolayita Damot Gale
115
Annex 13) List of FEED I Unions Visited/Consulted
Region Name of Union FEED
Project that 
supported
the union
Zone Woreda Town
Tigray 1. Geter Adwa
Multipurpose Coop. 
Union
Feed-II Central Adwa; Laelay
Michew
Adwa
2. Debre Asa
Multipurpose Coop. 
Union
Feed-II Central Kola-temben; 
Dega Temben
Abyi-adi
3. Bokra FCU Feed-I Southern Raya Azebo;
Ofla
-
Amhara 4. Admas Farmers
Coop. Union
Feed-II Awi Zone Guagussa;
Dangela
Kosober
5. Gozamin Farmers
Coop Union
Feed-II Eeast
Gojjam
Machacle; 
Gozamin
D/Markos
6. Merkeb FCU-I Feed-I West
Gojam
Fogera -
7. Megenagna Farmers
Coop. Union Feed- I S .Gonder Libo; Farta D/Tabor
Oromia 8. Liben Farmers' Coop.
Union
Feed-II South 
West 
Shoa
Becho; woliso Woliso
9. Galema FCU FEED II Arsi Tiyo & Hetosa Bokoji
SNNP 10. Ambericho Farmers
Coop. Union
Feed-II Durame Damboya; 
Kedida Gamela
Kembata
T.
11. Sidama Elto FCU Feed-I Sidama Wondo Genet -
12. Melik Farmers' Coop.
Union
Feed-II Siliti Worabe Selti; 
Dalocha