Impact Assessment of the ACDI/VOCA Livestock Markets in 
Pastoralist Areas of Ethiopia 
PLI Policy Project 
Tufts University 
November 2010 
This report was made possible by the support of the American People through the 
United States Agency for International Development (USAID.) The contents of the 
report are the sole responsibility of Tufts University and do not necessarily reflect the 
views of USAID or the United States Government. 
Contents 
Summary ..................................................................................................................................i
1. Introduction ................................................................................................................... 1
2. Design and Methodology............................................................................................ 1
2.1 Markets visited ............................................................................................. 2
2.2 Key informants ............................................................................................. 2
2.3 Data issues..................................................................................................... 3
3. Findings .......................................................................................................................... 4
3.1 Local market management.......................................................................... 4
3.1.1 Progress towards local market management ........................................... 4
3.3.2 Outstanding local management issues...................................................... 5
3.2 Trends and market performance................................................................ 7
3.2.1 Local traders and bush markets ................................................................. 7
3.2.2 Local conflicts................................................................................................ 8
3.3 Changes in livestock supplies and prices in 2010.................................... 9
3.3.1 Camels............................................................................................................ 9
3.3.2 Cattle ............................................................................................................ 10
3.3.3 Small ruminants.......................................................................................... 12
3.3.4 Summary of market supply and price trends......................................... 12
3.4 Transaction costs ........................................................................................ 12
3.5 Use and under‐use of market facilities.................................................... 13
3.6 Service provision around the new markets ............................................ 17
3.7 Unexpected impacts................................................................................... 17
4. Lessons and Recommendations ............................................................................... 18
4.1 Project design issues: pastoralist marketing behaviour and  
market access............................................................................................... 18
4.2 Project design issues: hard versus soft inputs ........................................ 20
4.3 Project design issues: community participation .................................... 22
4.4 Project design issues: a framework for analysis..................................... 23
Annex 1: Question checklist for the impact assessment ................................................. 25
i 
Summary 
In 2005 USAID Ethiopia provided a grant to ACDI/VOCA, a US-based private non￾profit organization (hereafter called ‘the implementer’) to strengthen livestock 
marketing in Afar, Somali and Oromia regions, with a view to increasing market off￾take and enhancing pastoralist livelihoods. The project budget was approximately 
US$ 8 million. From October 2005, project activities included the construction of new 
livestock markets. A cost extension phase from 2008 focused on the introduction of 
local management arrangements for the markets. 
In June 2010 and at the request of USAID Ethiopia, Tufts University conducted an 
impact assessment of the market construction and management components of the 
project. The assessment was based on visits to 18 livestock markets constructed by the 
project in the three regions, from the 25 markets which were constructed in total. The 
impact assessment design included attention to attribution and specifically, analysis of 
trends in the absence of project inputs (counterfactual analysis). Ideally, an impact 
assessment of new market infrastructure would measure the additional volume and 
value (if any) of market transactions before and after the construction of the markets. 
The assessment would then determine whether any measurable changes were 
attributable to the project, or, whether these changes would have occurred even the 
project had not been implemented. These kinds of measurements help to overcome the 
weaknesses of ad hoc, direct observation of market activity and the reporting of 
market transactions with no reference to a baseline. The lack of available baseline 
data on market activity before the project hindered this kind of analysis. 
In part, this approach involved examination of two bush markets, which evolved 
independently of the project. Although the assessment used some participatory 
methods, the overall design of the assessment was not participatory. Participatory 
impact assessment requires specific target communities and geographical areas to be 
defined for a project, and this did not occur in the project. 
Main findings 
The main findings of the assessment were as follows: 
• Drivers of market activity - of the 18 markets visited, nine markets were active 
and some of these were an important source of livestock for export. The drivers of 
increased trade included increasing demands for meat in the Middle East 
combined with higher-level Ethiopian government commitment to supporting 
meat exports, and to a lesser extent, live animal exports. Both trends were evident 
before the project started, and therefore, could not be attributed to the project. The 
main beneficiaries of the export trade include the export companies, 
intermediaries, and wealthy and middle-wealth group pastoralists. 
• Markets as a cause of conflict - in pastoralist areas the careful positioning of new 
services and structures is needed to ‘do no harm’ and avoid conflicts between 
neighbouring groups, or conflicts between new structures and with the owners of 
existing services. In the project three markets were closed indefinitely because of 
conflicts (in Yallo, Chereti and Kersadula), and three other markets were 
adversely affected when one ethnic group refused to use the market because of 
ii 
tensions with another group (in Negele, Moyale and Hudet); in other words, 
nearly 25% of the new markets that were constructed were affected by conflict 
issues. This finding points to an initial weak analysis when positioning the 
markets. The emergence of new bush markets was partly a response to these 
conflicts, as the bush markets were positioned so as to avoid confrontation 
between ethnic groups. 
• Systematic institutional issues – the project supported Livestock Marketing 
Management Boards (LMBBs), and these had been established in Somali and 
Oromia regions. However, a working system for the appropriate collection of 
market revenue and use of revenue for market maintenance and running costs was 
in place, partially, in only two markets (Haraboke and Finchewa) and was not 
evident in the remaining 16 markets. In the more active markets (nine in total), in 
theory the volume of livestock trade should have been sufficient to generate 
revenue for the long-term market maintenance and running costs. However, 
various deep-rooted institutional weaknesses hindered progress, including the 
informal collection of taxes and fees. Given that market infrastructure was already 
deteriorating, the likely scenario was that the markets would fall further into 
disrepair unless further external assistance was provided. The assessment did not 
examine the written agreements between the project and local government 
partners, which were reported to describe the government roles and 
responsibilities for maintaining the markets. However, stakeholder interviews 
indicated that in most cases, these agreements had not been met. Analysis of local 
organizational capacities and institutional arrangements before, or at the onset, of 
the project should have informed an assessment of future risks for the project, 
such as over-reliance on local government capacities. In part, problems with 
livestock market management by the public sector reflect wider governance 
challenges in pastoral areas. 
• Dormant markets - of the 15 new markets which were initially assumed to be 
active, six were more or less dormant. The low volume of trade in these markets - 
less than 50 goats and few cattle at most per market day – was predicted in 2005, 
indicating that these markets should not have been constructed. It is very unlikely 
that these markets can raise enough revenue to even hire market attendants. 
• Use of specific market facilities - the design of the new markets included various 
facilities such as water and feed troughs, partition pens, livestock scales, crushes, 
bulletin boards and public toilets. In the 15 open markets which were visited, only 
27% of these facilities were in use. This finding supports the conclusion that most 
of the market features which the implementer initially deemed necessary were not 
actually needed. The most commonly used facilities included veterinary posts, and 
livestock compartments, and in some markets, the loading ramps. 
• Marketing guidelines - draft livestock marketing guidelines were prepared for 
the three regions by the project. The guidelines proposed an allocation of 30% of 
market revenue for the maintenance of the facilities. The guidelines were not 
being implemented at the time of the assessment. 
• Veterinary posts - at some of the new markets, veterinary and other services were 
available. In the case of veterinary care, analysis of impact requires baseline data 
iii 
on the accessibility, availability, affordability, acceptance and quality of service. 
This information was not available. However, the provision of clinical veterinary 
care by local government at market locations could be seen to contradict policies 
which support delivery of clinical veterinary services by the private sector. Further 
assessment is needed to determine the impact of these new veterinary facilities. 
The assessment should include the extent to which these government services 
complemented or undermined pre-existing private veterinary service providers in 
the project areas. There are also various technical questions to consider, such as 
the risks of positioning veterinary posts at livestock markets and the possible 
contact between sick and healthy livestock. 
• Growth of bush markets - the new bush markets observed during the assessment 
reflected the dynamic nature of livestock marketing in pastoralist areas, and the 
efficiency of ‘appropriate technology’ in responding to trends in market demands. 
In these markets locally-based traders were sourcing animals for direct movement 
to export companies. These markets were easier to access for producers, and 
enabled producers to sell to traders who were more trusted. One implication is that 
the bush markets will further reduce the viability of the more formal, new markets 
in urban centres and on main roads. 
• Women’s groups - the project supported service provision in the new markets 
e.g. by helping to organize women’s groups and by aiming to provide these groups 
with seed money. At the timing of the assessment, it was too early to measure the 
impact of this activity. For example, not all groups had received the seed money. 
 
• Construction issues – as stated above under use of specific market facilities, there 
was limited use of many of the facilities. In addition to some facilities not being 
required, was evidence that many facilities had fallen into a state of disrepair 
either due to weak maintenance, and/or, problems with the quality of the initial 
construction. 
Recommendations
The overall conclusion of the assessment was that USAID should not provide further 
support to livestock marketing infrastructure in pastoralist areas. Attempts to improve 
market management need to be considered against the question of whether elaborate 
market infrastructure is really needed for pastoral livestock marketing. In general, 
simple bush markets work well. The assessment also concluded that a main driver of 
livestock price increases and related trade from pastoral areas, especially in Borana, 
was government commitment to support meat exports. The key challenges relate not 
to market construction, but to various outstanding policy and institutional constraints. 
The assessment team did not include construction experts, and the pre-agreed 
questions for the assessment did not include an examination of the quality of market 
construction. However, given the poor physical condition of market facilities in some 
locations, at a relatively short time after construction, USAID should consider a 
separate independent review which looks specifically at construction issues and 
supervision of construction subcontracts. 
iv 
Four specific recommendations are made, focusing on the need for better pre-project 
analysis and opportunities to strengthen government and implementer capacities in 
such analysis. 
Recommendation 1
Attempts to intervene in pastoral livestock marketing systems should document local 
marketing behaviours by wealth group and carefully examine any proposed causal 
pathway that associates more market infrastructure with more sales. Such pathways 
are unlikely in pastoralist areas. Policy engagement and economic analysis with 
government partners is needed to raise understanding of these issues at federal and 
regional levels. 
Recommendation 2
Livestock markets in pastoralist areas require basic facilities, such as loading ramps 
and water troughs. Investment is more elaborate infrastructure is only warranted if a 
positive pre-project benefit-cost analysis shows an appropriate return on investment. 
In addition, a pre-project assessment of local government management and 
administrative capacities is needed, including analysis of the actual capacity of 
government to deliver versus their proposed/stated capacity e.g. as expressed in 
project agreements. In many cases, the key constraints to pastoral livestock markets 
will be policy and institutional (e.g. inappropriate taxation; restrictive livestock 
movement regulations), rather than hardware. Support to government to improve 
understanding of these issues and build capacity for benefit-cost analysis and 
organizational capacity analysis is warranted. 
Recommendation 3
In addition to a capacity to conduct economic analyses and understand local 
marketing behaviours, implementers also need to understand local cultural, social and 
conflict dynamics, and recognize the need to consult a range of local actors and 
stakeholders. In conflict-affected or conflict-prone areas, ‘Do No Harm’ approaches 
are needed, as already recommended and used by other USAID implementers in 
pastoralist areas. When selecting implementers, USAID needs to consider past 
organizational experience specifically in pastoralist areas, and understanding and 
capacity to conduct conflict analysis and implement conflict-sensitive programming. 
Recommendation 4
USAID supports ‘scaling-up’ of new approaches based on evidence. In the case of 
livestock markets in pastoralist areas, a substantial body of evidence already exists. 
Future needs include building the capacity of government and implementers to 
conduct appropriate, comprehensive pre-project analysis using mainly existing 
information, but supported by rapid applied assessments as needed.
1
1. Introduction 
Under the Pastoralist Livelihoods Initiative (PLI) project in Ethiopia, USAID 
provided a grant to a US-based private non-profit organization (hereafter called ‘the 
implementer’) to strengthen livestock marketing in Afar, Somali and Oromia regions. 
The aim was to increase market off-take and enhance pastoralist livelihoods. The 
project budget was approximately US$8 million. From October 2005, project 
activities included the construction of new livestock markets. In August 2008, an 
impact assessment was conducted on 10 of these markets, which were judged by the 
implementer be functional at that time. Some of the key findings of the earlier 
assessment in 2008 were as follows: 
• in the new markets only 20% of the facilities were being used; 
• livestock marketing activity was taking place outside seven (out of 10) of the new 
markets which were assessed; 
• there was no evidence to show that livestock sales had increased as a result of the 
construction of the new markets; 
• taxation had increased in the locations with the 10 new markets; on average these 
increases were 10% for cattle and 33% for camels, sheep and goats. 
In 2008 USAID awarded a cost-extension to the implementer with the aim of 
resolving some of the management issues in the new markets. 
In June 2010 USAID requested Tufts University to conduct a second impact 
assessment of the project. The assessment was based on visits to 20 markets as 
follows: 15 out of 18 new livestock markets which the implementer reported as open 
at the time; three out of seven markets which were new but judged to be non￾functional (note –the project constructed 25 markets in total); and two local bush 
markets which were not part of the project. In agreement with USAID and the 
implementer, the assessment focused on the following issues: 
• progress made towards sustained use of markets; 
• changes in livestock supply, sales volume and prices, if any, and attributable 
factors; 
• transaction modalities and perceptions of market actors and service providers on 
issues of importance; 
• intended and unintended impacts of the new markets on livestock trade and 
transaction modalities in general; 
• status, and usage level of installed facilities; 
• lessons to be drawn from the exclusive use of the Shinile market by a livestock 
exporter. 
2. Design and Methodology 
The same team who conducted the 2008 impact assessment also conducted the 
assessment in 2010, and 10 markets initially covered in the earlier assessment were 
also included in the 2010 assessment. This approach allowed the assessment team to 
track progress over a two-year period in 10 markets. 
 
2 
2.1 Markets visited 
The markets visited during the assessment are listed in Table 1. 
Table 1. Location and status of markets assessed in 2008 and 2010 
Location and type of livestock market Impact assessment 
August 2008 
Impact assessment 
June 2010 
Somali Region 
New, open 
New, non-operational1
Afar Region 
New, open 
New, non-operational 
Oromia Region 
New, open
New, bush market2
Jijiga 
Hartisheikh 
Lefe Issa 
Moyale 
Hudet 
Shinile 
Eliwoha 
Chifra 
Asayita 
Logia 
Gewane 
Andido 
Teltele 
Negele 
Dubuluk 
Bokoloboma 
Finchewa 
Harobake 
Surupa 
El Waya 
+ 
+ 
- 
+ 
- 
- 
- 
- 
+ 
- 
- 
- 
+ 
+ 
+ 
+ 
+ 
+ 
- 
- 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
+ 
1 Of the non-operational markets, Shinile market was leased to live animal exporters and the intention was for 
Andido market to be leased to a traders association; visits were made to these sites to see how they are operating. 
2
 Two bush markets were included in the assessment to appraise their performance compared to the newly 
constructed markets. The bush markets had evolved independently of the project. 
2.2 Key informants 
The assessment was carried out in close collaboration with project coordinators. All 
sites were visited on market days. Key informants are listed below. 
• Pastoralists, traders and brokers and people providing cattle holding and 
loading services to local traders in bush markets. 
• Truck owners, tax collectors, guards and women and children selling tea, food 
and cold water at market sites. 
• Local councils and administration officials and various associations linked to 
livestock marketing activities. 
Inevitably, when interviews involved sensitive questions around tax collection, use of 
revenues, conflicts and other issues, most informants preferred to remain anonymous. 
Research instruments were designed in consultation with the implementer. The 
question checklist for the assessment is provided in Annex 1. 
3 
Specific information provided by informants during interviews was triangulated with 
information provided by other informants, and with secondary data including: 
• An audit of livestock marketing in Ethiopia, Kenya and Sudan conducted in 
2002, providing data from before the ACDI/VOCA project1
; 
• Field assessment of the performance of livestock marketing groups in Borana 
in 2004, providing market trend and price information for the period before the 
ACDI/VOCA project2
; 
• Various papers arising from the USAID-funded GL-CRSP project that 
conducted research on livestock marketing in northern Kenya and southern 
Ethiopia, mostly before the ACDI/VOCA project3
; 
• Minutes of the quarterly PLI Technical Coordination Meetings for Oromia and 
Somali regions since October 2005; 
• Information from the Livelihoods Information Unit, Ethiopia, on livestock 
market prices. 
• A study on livestock marketing constraints in Ethiopia and Kenya conducted 
as part of the USAID-funded regional Pastoral Areas Coordination Analysis 
and Policy Support project in 20084
; 
• Field research in Borana and Somali areas conducted as part of an IGAD-FAO 
study on livestock marketing in pastoralist areas of the Horn of Africa in 
20095
; 
• A policy paper on the potential for pastoralist livestock to support national 
economic growth in Ethiopia, commissioned by DFID and presented to the 
Prime Minister’s Office in late 20106
. 
In summary, these and other reports enable analysis of market trends in Borana and 
Somali areas over about ten years. 
The assessment team was not provided with copies of the implementer’s monitoring 
reports, and therefore, did not use project M&E data to triangulate the findings of the 
assessment. 
2.3 Data issues 
Ideally, an impact assessment of new market infrastructure would measure the 
additional volume and value of market transactions before and after the construction 
of the markets. The assessment would then determine whether any measurable 
changes were attributable to the project, or, whether these changes would have 
 
1
 Akilu, Y. (2002). An Audit of the Livestock Marketing Status in Kenya, Ethiopia and Sudan. 
AU/IBAR and Tufts University, Nairobi. 
2
 Aklilu, Y. (2004). Pastoralist Livestock Marketing Groups in Southern Ethiopia: Some preliminary 
findings. Presentation to the CORDAID Access to Markets Workshop, November 2004, Nazreth. 3
 McPeak, J.G. and Little, P.D. (2006). Pastoral Livestock Marketing in Africa: Research and Policy 
Challenges, IT Publications, Rugby. 4
 Aklilu, Y. (2008). Livestock Marketing in Kenya and Ethiopia: A review of policy and practice. Tufts 
University, Addis Ababa. 
5
 Aklilu, Y. and Catley, A. (2009). Livestock Exports from Pastoralist Areas: An analysis of benefits by 
wealth group and policy implications. Report for the Intergovernmental Authority for Development 
and Food and Agriculture Organization. Tufts University, Addis Ababa. 
6
 Akilu, Y. (2009). Livestock and Animal Product Trade and their Importance for Pastoral Economic 
Growth in Ethiopia. Unpublished Concept Note for the Pastoral Economic Growth and Development 
Study, Addis Ababa, Ethiopia. 
4 
occurred even the project had not been implemented. These kinds of measurements 
help to overcome the weaknesses of ad hoc, direct observation of market activity and 
the reporting of market transactions with no reference to a baseline. Such observations 
and reports are of limited value in terms of assessing impact, because active markets 
might have been present before the project, or, would have occurred despite the 
project. The lack of available baseline data on market activity before the project 
hindered this kind of analysis. 
3. Findings 
3.1 Local market management 
The focus of the cost-extension phase of the project was to establish management 
systems for the new livestock markets and to introduce standardized systems, fair 
management practices, and ensure the sustained use of the installed market facilities. 
Progress towards these aims was based on the formation of Livestock Marketing 
Management Boards (LMMB), established in Oromia and Somali regions, but not 
Afar region. The LMMB members were representatives from the municipality, and 
finance, agriculture, health, water development, cooperative and security bureaux (in 
the case of Jijiga, brokers were also represented). The LMMBs had conducted at least 
one meeting in Oromia and Somali regions. 
Before the formation of the LMMBs in Oromia, new official receipts were issued for 
Harobake market (which were being used by the four kebeles), with some financial 
support from the project. The project had also negotiated for the relocation of women 
engaged in service provision to the vicinity of the Moyale livestock market. 
3.1.1 Progress towards local market management 
Outcomes of the LMMB meetings included the following: 
• A water pipeline had been extended to Moyale, Negele and Teltele markets; 
• Woreda Agriculture Offices opened or reopened veterinary posts and assigned 
personnel in all operational market sites, including the provision of drugs in some 
markets; 
• The implementer negotiated with the authorities in Negele for the use of a single 
receipt for taxation purposes by the Revenue Office; 
• The implementer facilitated the opening of bank accounts for the Harobake and 
Finchewa markets. At the time of the assessment mission, the Harobake bank 
account showed a deposit of EB 18,400 (~US$ 1,343) and Finchewa EB 10,000 
(~US$ 730). The savings are to be used for the maintenance of infrastructure. 
• The implementer organised women operating around the new market sites into 
cooperatives and donated EB 15,000 (~US$ 1,095) seed money per cooperative; 
by June 2010 this money had been provided to the cooperatives, other than those 
in Jijiga, Negele and Moyale markets. 
• Livestock marketing guidelines were drafted in cooperation with the three 
regional governments. 
5 
3.3.2 Outstanding local management issues 
Market-related income and its management 
The implementer made notable efforts with regional and local authorities to introduce 
a management system for the sustained use of the markets, on a self-financing basis. 
The LMMBs were established in two regions (not Afar Region) and agreements were 
reached, at least theoretically, to allocate 25 to 30% of the market-related income for 
maintenance purposes. The practicality of this arrangement was doubtful in all 
markets visited, except in Harobake and Finchewa where saving accounts were 
opened for maintenance purposes. Officials in Jijiga in particular stated that the 
maintenance of the market facilities was the responsibility of the regional government 
(as for other infrastructure in the region) and therefore, a separate bank account for the 
maintenance of the market was not needed. 
Taxing of unsold animals in Oromia markets was still evident despite the 
implementer’s efforts to get this repealed. The LMMB members stated that tax 
standardization and the allocation of money to the revenue office versus maintenance 
purposes was beyond their mandate, and should be dealt with at regional or federal 
level. 
Savings in Harobake and Finchewa markets did not reflect actual income levels, as 
agreed with the implementer. In Harobake, savings over a 50-week period were EB 
24,400, of which EB 6,000 was used for the printing of official receipts. The savings 
arrangement with the implementer was for the four kebeles to deposit EB 2,000 per 
market day for the first 35 weeks of the year, and EB 500 per market day for the last 
15 weeks of the year7
. Therefore, the total net saving for the 50 weeks should have 
amounted to about EB 71,500. In Finchewa too, the municipality head confirmed that 
deposits into the savings account was interrupted for undisclosed reasons. 
In Harobake there was one market day a week, which sold around 1,200 cattle, 300 
small ruminants and about 100 camels per market day (1,600 livestock in total). Five 
thousand new tax collection receipts were issued to market tax collectors on 17th
March 2010, to be used for all livestock species brought to this market. At roughly 
1600 animals sold per market day, these receipts should all have been issued within 
three weeks (i.e. three market days). However, when visiting the market on 27th June 
2010, the receipts were still in use. That was 10 weeks after the receipts should all 
have been issued. One explanation for this discrepancy was that receipts were not 
always issued when tax was collected. 
The problem of saving money for maintenance purposes is mostly explained by the 
low level of revenue raised from the majority of the markets. Markets with low 
transaction volumes, notably in Afar and Somali regions but also including 
Bokoloboma in Oromia Region, are unlikely to acquire meaningful sums for future 
maintenance because much of the limited revenue they raise is paid as salaries of 
market attendants. For example, in Afar tax income data collected by project staff 
from Chifra (the best performing market in the region) showed a tax income of 
 
7
 The arrangement was for each kebele was to collect tax on monthly basis and deposit the fixed 
amount for market maintenance into a bank account in Yabello. 
6 
between EB 1,000 to EB 1,500 per market day, equivalent to EB 4,000 to EB 5,000 
per month. However, the monthly salaries of three guards and three cleaners, needed 
to take care of the three market compartments amounted to EB 3,600 per month 
(using the EB 600 minimum government monthly salary rate for Afar region). Many 
of the other markets were raising less than half the taxes seen in Chifra. 
The finding above implied that in theory, the Borana markets (including Moyale) had 
the potential to raise money for maintenance and upgrading. Two of these markets, as 
mentioned above, had made a start although progress was irregular. However, the 
markets in Borana raised fees from sold and unsold animals, plus additional fees were 
collected from traders and producers by individuals in the market, with no receipts 
provided but with apparent authorization by the kebeles. Loading ramps and other 
facilities or services had become income-generating activities for people who had 
organised themselves into an informal association. Similar behaviour was also 
observed in Asayita. In summary, there were varied personal and group interests in 
acquiring extra, informal income from the markets. The effort of the project to 
formalize and streamline market management was undermined by this behaviour, 
even in markets which raised substantial revenues; this needs to be clearly understood. 
However, to some extent such opportunistic tendencies are predictable if local 
government capacities are considered along with the pre-existing system of taxation 
on livestock trade and the imposition of various ‘informal taxes’. The directive that 
regional governments should also raise as much local revenue as possible was also 
used as a pretext to redirect market income for private personal or group benefit. 
Leasing of markets for alternative uses 
The new Shinile market had never been operational and it was decided to lease it to 
private livestock exporters as a feedlot and holding ground. The first company which 
leased the facility installed water and feed troughs inside the market compartment8
 but 
then withdrew. Project staff indicated that a shortage of feed and the high price of 
feed may have been the reason for the withdrawal. The current lease arrangement was 
made between the Shinile Municipality and the REZ Agrovet Company, based in Dire 
Dawa. The lease agreement was EB 3,000 per month plus covering the salaries of 
three guards. The company was expecting to start using the facility in August 2010, 
although the lease agreement began in March 2010 and the company had paid rent for 
three months, including salaries for the guards (amounting to EB 10,800). 
Municipality officials refused to disclose how this money was being used, but agreed 
that it was not saved. 
Despite these arrangements, it was notable that both parties were still requesting the 
implementer to add missing market facilities or repair dilapidated facilities, instead of 
using the lease money. While the project initiative to lease the market for an 
alternative use was commendable, serious questions remain over the capacity of local 
government to manage the lease and the related income. 
 
 
8
 This is a practical arrangement compared to putting water troughs outside the market yard. 
7 
Marketing guidelines 
The draft guidelines developed in collaboration with the three regional governments 
had not yet been finalized. The implementation of these guidelines remains doubtful 
unless robust enforcement mechanisms are put in place to control the various personal 
and group interests at the local level. For example, the guidelines propose the use of 
official, printed receipts for all tax transactions but there is no mechanism to ensure 
that tax collectors actually issue a receipt for all transactions. 
3.2 Trends and market performance 
Two emerging trends were observed with potential implications on the performance 
of the new livestock markets covered in the assessment. The first trend involves a 
marketing strategy employed by local livestock traders, and the second was a result of 
inter-clan conflicts. Both trends were outside the control of the project. 
3.2.1 Local traders and bush markets 
Taking advantage of improved roads and access to mobile phones, locally-based 
livestock traders were increasingly sourcing livestock from bush markets, instead of 
markets on the main roads. Such traders seemed to be financially well-established and 
ventured into bush markets with their own trucks in a bid to out-compete non-resident 
traders, by purchasing animals at source. This trend prompted producers to set up 
basic loading ramps and holding grounds close to the bush markets, for rent. 
Producers were also selling hay to trade herds from enclosed kallos9
. 
The growing importance of bush markets in Borana was very evident in El Waya and 
Surupa. Compared to 2009, the average supply of camels in the Surupa market had 
increased by 40%, sales by 30% and price by 18% in 2010. In El Waya, the average 
supply of cattle had also increased by 17%, price by 30% while sales has gone down 
by about 1%. Similarly, the supply of small ruminants went up by 30%, sales by 10% 
and price by 20%. 
Pastoralists in Borana, including Moyale have become increasingly resistant to the use 
of scales when selling animals. This is due to the experience of being cheated by non￾local agents buying sheep and goats for the export abattoirs10. The refusal of these 
agents to use the scales provided by the implementer in the new markets reinforced 
the suspicions of producers. Although the owners of the export abattoirs seemed to be 
aware of such malpractices, they chose to remain silent in fear of losing their agents. 
These trends towards greater use of bush markets had enabled locally-based livestock 
traders to take the lead in livestock transactions in Borana, and in raising their status 
as the main suppliers of livestock to live animal exporters, feedlot operators and 
export abattoirs. As a result, these major actors had effectively exited from operating 
directly in Borana markets, except through such locally-based intermediaries. 
 
9
A kallo is an enclosed area to help the re-growth of pasture or hay, usually protected by temporary or 
permanent fences, and privately owned. 
10 During the assessment, the team observed an event where an agent weighed a goat and told the 
owners it weighed 20kg. The actual weight was 24kg. When we confronted the agent, he ran away with 
the scales. 
8 
The growth of bush markets was impacting the supply levels of livestock to the new 
market facilities in Borana, including the volume of sales (compared to 2009). This 
will result in the under-use of facilities and a corresponding reduction in the amount 
of fees and taxes to be raised. Similarly, the income of other service providers in and 
around the new markets will also decrease. 
The development of bush markets will likely bring mixed results. Vibrant bush 
markets will attract other types of businesses including the provision of commodities 
by traders. Therefore, producers will not have to travel long distances to sell animals 
and purchase essential commodities. Rural producers will benefit from alternative 
income sources through service provision (guarding, trekking, renting loading ramps 
and holding grounds, etc.). As commercialization intensifies, the production mode 
will gradually become market-oriented in response to the market demand. Value 
adding could take place at village levels by using commercial feed and pasture from 
enclosed kallos. Increased use of kallos, in turn, could promote agro-pastoralism with 
the resulting restrictions on seasonal movements and could become the source of 
tension and localized conflicts. 
3.2.2 Local conflicts 
The creation of new bush markets was also affected by inter-ethnic conflicts, and such 
markets tend to be positioned to avoid conflict (Table 2). 
Table 2. New livestock markets affected by inter and intra ethnic conflicts 
Location of new market Type and outcome of conflict 
Yallo, Afar Region Market not opened - the new market was never opened by the project. A 
clan which was gaining income from the old market, about 5km away, 
refused to allow the new market to open. 
Chereti, Somali Region Market not opened - for the same reason as above, the new market was not 
open by June 2010. 
Kersadula, Somali 
Region 
Market not opened - the new market was never opened because of a land 
issues between the Oromos and Somalis. The Oromos owned the famous 
old market in Medawalabu, an historic site where the Welabu clan 
originated. The new market site was about 5km away from the old 
Medawelabu market and was located on the Somali side. 
 
Negele, Oromia Region Reduced market supply - the Digodi (Somali) pastoralists stopped visiting 
Negele due to conflict with the Oromos. The Digodis are establishing a 
new bush market at Awregebeya, located about 50km from Negele. 
Moyale, Somali Region Reduced market supply - Boran pastoralists were not using this market, 
preferring to sell cattle in Kenya Moyale and small ruminants on the road 
side; this was due to deep-rooted conflict with the Somali Garis. 
Hudet, Somali Region Reduced market supply -the Borans have refused to use the Hudet market 
for the same reason as above. 
It appears that livestock market dynamics in southern and south-eastern Ethiopia are 
influenced by three major factors: locally-based traders are increasingly promoting 
9 
bush markets, assisted by improved roads and mobile phones; ethnic conflicts and 
conflicts of interest are leading to the boycotting of the new markets or the 
establishment of bush markets; and in a few cases, the location of the new markets 
have become a cause for conflict. These trends were particularly evident in Borana, 
but also in Somali Region and in one specific case in Afar Region. 
The trends suggest that livestock markets will become more ethnically-orientated as 
bush markets expand and grow in some areas, and with corresponding reductions in 
the supply of livestock to markets in major settlements or close to main roads. 
Reduced income from market fees and taxes, in turn, will affect the long-term 
viability of the new markets, despite possible improvements to market management. 
Crucially, the fact that bush markets operate efficiently without modern market 
infrastructure and facilities raises the question as to whether such amenities were ever 
really needed in the first place. 
Producers confirmed to the assessment team that market information (as regards 
prices of animals) is not a problem to them because they can get this information 
easily from brokers, who are clan members. They added that the expansion of mobile 
telephone networks and wireless telephone service has made it possible to get 
information on livestock prices without the need to come to market sites. 
3.3 Changes in livestock supplies and prices in 2010 
3.3.1 Camels 
Camels were on sale in seven markets (Jijiga, Moyale, Negele, Harobake, Dubuluk 
Chifra and Aysayta) during the field assessment (Table 3). The supply was low in all 
Borena markets11 (except Negele) compared to Afar and Somali regions12. Substantial 
increases were observed in camel prices relative to 2009. 
Table 3. Camel market transactions in new markets and bush markets, June 2009 and June 2010 
Market location Supply trenda
 Sales trenda
 Average price (EB) 
June 
2009 
June 
2010 
 June 
2009 
June 
2010 
 June 
2009 
June 
2010 
New markets: 
Jijiga 
Moyaled
Negele 
Dubuluk 
Harobake 
Chifra 
Aysayta 
38 
38 
28 
63 
61 
32 
31 
62 
62 
72 
38 
39 
68 
69 
 
16 
13 
21 
54 
52 
11 
17 
37 
41 
62 
13 
24 
41 
32 
 
5,070 
3,995 
3,900 
ND 
7,125 
ND 
4,500 
8,950 
10,200 
7,875 
ND 
11,750 
ND 
7,000 
Bush marketsb 29 71 13 55 43% increasec
a
 Data derived from proportional piling. For the supply trend informants were given a pile of 100 stones 
to depict all animals supplied to the market in June 2009 and June 2010. They were then asked to 
divide the stones to illustrate the relative supplies in June 2009 and June 2010. These two piles were 
 
11 Excluding the bush markets. 
12 For example, there were only six camels brought for sale at Dubluk on the day of the assessment 
team’s visit. 
10 
then further sub-divided to show the proportion of animals sold e.g. in Jijiga in 2009 the supply figure 
of 38 was reduced to a sales figure of 16. 
b
 Averaged data. 
c Data derived from proportional piling. 
d Camels sold in Moyale included those sourced in Kenya and Somalia; camels from Somalia were 
known for their large size.
ND = No data. 
In terms of attributing increases in camel supply, sales or prices to the construction of 
new markets, similar trends were also evident in the two bush markets in Oromia 
Region. Also, traders explained that the price of camels has gone up due to increased 
export demand. Subsequently, supplies of camels increased in Afar and Somali 
markets in response to the demand, whereas the volume of supply decreased in 
markets along the Yabello-Moyale route. Overall, marked increases were observed in 
camel prices and supplies in 2010. The average price of camels rose by around 86% 
relative to 2009 from EB 4,500 in 2009 to EB 9,155 in 2010. 
The increase in camel prices was more pronounced in Moyale and Harobake markets. 
Of note, most of the camels brought to Moyale market were sourced from Kenya and 
southern Somalia. Meanwhile, the camel supply at Harobake was affected by the 
growing importance of Surupa bush market. However, the substantial increase in 
camel prices in both markets (amounting to EB 6,000 in Moyale and close to EB 
5,000 in Harobake) could imply that the quality of ‘exportable camel types’ in these 
markets was far better relative to other markets. The price increase in Jijiga, Negele 
and Aysayta markets was also substantial in 2010 (at between EB 3,000 to 4,000) but 
lower than Moyale and Harobake markets because these markets responded to the 
export demand by increasing supplies. 
 
Of the seven new markets where camels were sold, the proportion of camels sold in 
2010 was higher in Negele, Moyale, Chifra and Jijiga. The exceptions were: Harobake 
because of the increasing dominance of Surupa, as a major camel market; Dubluk, 
because of the importance of Surupa and Moyale markets for camels; and Aysayta, 
because exporters were not operating in this site. Given the good rains in 2010 
(gu/genna season), the increase in supplies indicates a positive response to demand. 
The rise in the price of camels is partially attributed to the good rains as pastoralists 
have the option to take back unsold camels if they did not agree on the price. 
3.3.2 Cattle 
Cattle were on sale mainly in Jijiga, Chifra, Aysayta, Negele, Moyale, Dubuluk, 
Teltele, Harobake and Finchewa markets during the assessment (Table 4). Cattle sales 
were very low in Lefe Isa, Hudet and Bokoloboma markets. 
A substantial reduction was observed in the proportion of cattle supplied and sold in 
the markets studied in 2010 compared to 2009, but prices were far higher in 2010 
(Table 4). At the nine new markets the average price increase was from EB 2,141 Birr 
in 2009, to EB 3,105 in 2010, equivalent to a 34% increase. 
11 
Table 4. Cattle market transactions in new markets and bush markets, June 2009 and June 201013
Market location Supply trenda
 Sales trenda
 Average price (EB) 
June 
2009 
June 
2010 
 June 
2009 
June 
2010 
June 
2009 
June 
2010 
New markets: 
Jijiga 
Moyale 
Negele 
Dubuluk 
Teltele 
Harobake 
Finchewa 
Chifra 
Aysayta 
59 
70 
39 
45 
59 
36 
42 
68 
78 
41 
30 
61 
55 
41 
64 
58 
32 
22 
37 
60 
30 
21 
37 
21 
22 
43 
45 
12 
24 
42 
39 
29 
41 
30 
13 
10 
 
918 
1,965 
1,963 
2,775 
2,125 
4,075 
3,000 
1,800 
650 
2,060 
2,895 
3,592 
3,650 
3,350 
4,650 
4,000 
3,000 
750 
Bush marketsb
 46 54 32 37 82% increasec
a
 Data derived from proportional piling – see Table 3 for methodology. 
b
 Averaged data. 
c Data derived from proportional piling. 
Table 5. Small ruminant market transactions in new markets and bush markets, June 2009 and June 
2010 
Market location Supply trenda
 Sales trenda
 Average price (EB) 
June 
2009 
June 
2010 
 June 
2009 
June 
2010 
 June 
2009 
June 
2010 
New markets: 
Jijiga 
Hartishekh 
Moyale 
Negele 
Dubuluk 
Teltele 
Harobake 
Finchewa 
Bokoloboma 
Chifra 
Aysayta 
Eluha 
Logia 
69 
56 
88 
48 
62 
61 
64 
58 
67 
28 
45 
41 
69 
31 
44 
12 
52 
38 
39 
36 
42 
33 
72 
55 
59 
31 
 
48 
39 
71 
26 
56 
40 
46 
43 
55 
15 
11 
34 
55 
11 
15 
11 
34 
29 
35 
30 
37 
29 
59 
41 
66 
12 
281 
258 
ND 
ND 
400 
250 
ND 
ND 
483 
400 
475 
ND 
ND 
454 
382 
ND 
ND 
625 
700 
ND 
ND 
345 
450 
418 
ND 
ND 
Bush markets 35 66 19 51 43% increasec
a
 Data derived from proportional piling. 
b
 Averaged data. 
c Data derived from proportional piling. 
ND = No data. 
 
 
13 The main cattle buyers in Jijiga and Aysayta markets were highland farmers and local butchers. 
Highland farmers buy young bulls from these markets to raise them as plough oxen. Demand for young 
bulls in these markets is, therefore, seasonal and drops low during the farming season – the main reason 
why cattle prices are lower in these markets relative to others. 
12 
3.3.3 Small ruminants 
Goats and sheep were on sale in 13 new markets during the assessment (Table 5). 
Data on prices collected from pastoralists, traders and brokers showed on average, a 
32% price increase for small ruminants from EB 364 in 2009, to EB 482 in 2010. 
For exportable sheep and goats, the price per live-weight unit increased on average by 
26%, from EB 8.8 per kg in 2009, to EB 11.2 in 2010. A summary of prices for small 
ruminants, obtained from seven new markets, is presented in Table 6. 
In contrast to rising prices, the supply of small ruminants decreased by an average of 
16% (from 58% in 2009 to 42% in 2010). No notable differences were observed in 
sales volume between 2009 and 2010 in the markets assessed.
Table 6. Changes in the prices of exportable shoats (per kilogram body weight) in 2009 and 2010 
Market location Price (per kilogram bodyweight) 
2009 2010 
Eluha 
Negele 
Dubuluk 
Teltele 
Harobake 
Finchewa 
Bokoloboma 
9 
8 
10 
8.5 
10 
9 
9 
9.5 
11 
12 
11.5 
12 
11.5 
11 
3.3.4 Summary of market supply and price trends 
Overall, three factors explained the increase in the price of camels, cattle and small 
ruminants in 2010. These were inflation, increased export demand (particularly for 
camels, sheep and goats), and the effects of the good rains in the first half of 2010. 
Pastoralists responded to the current market dynamics in different ways. The 
exceptional increases in camel prices resulted in increased supplies of camels to 
Somali and Afar markets, and also to Negele and Surupa markets in Borana. Given 
that the off-take volume of camels has always been low relative to cattle and small 
ruminants, it seemed that the unusually high price of camels was an opportunity not to 
be missed. 
However, the response was different for cattle and small ruminants. The good rains 
meant there was enough food around for pastoralists to withhold supplies to markets 
in order to re-build herds. It also meant better bargaining power for producers, who 
raised prices above the level of inflation. This was reflected in a reduction in the 
proportion of cattle sold while the proportion of shoats sold remained the same in 
2010 relative to 2009, despite price increases of well above 30%. This analysis was 
reinforced by the fact that meat exporters were compelled to leave Afar and Somali 
markets because of the high prices of small ruminants. 
3.4 Transaction costs
Transaction costs were assessed for Dubluk, Finchewa and Harobake markets in 
Borana. These costs included: 
• cattle and small ruminant loading services; 
• youth associations trekking cattle to temporary holding grounds; 
13 
• hoof tax collected at check points (kela); 
• other government taxes; 
• transport cost, including charges by lorry brokers. 
Specific details are provided in Table 7. 
There were slight increases in taxation rates relative to 2008 in some of the Somali 
and Afar markets, but there were no major complaints about the new rates. In Afar 
and Somali regions, buyers were responsible for paying tax. The practice of taxing 
unsold animals in Borana, however, began after the new markets were constructed 
and the practice was continuing despite the efforts of the implementer to halt it. 
Table 7. Minimum transactions costs in three Borana markets 
Type of transaction cost Cost (EB) per head of cattle 
Dubuluk Harobake Finchewa 
Truck rental 
Accompanier 
Loaders 
Truck brokers 
Hoof tax 
Government tax 
Total 
Average cattle price at time of assessment 
Total transaction costs as a proportion of 
price
326 
35 
10 
10 
2 
10 
393 
3,650 
10.8% 
326 
31 
10 
10 
2 
10 
389 
4,650 
8.4% 
326 
25 
10 
10 
2 
12 
385 
4,000 
9.6% 
Note – in these markets, local loaders and cattle accompaniers had stopped traders from using bush￾based traditional loading facilities and from bringing their own accompaniers from highland areas, 
because they are organized into legal associations by the government. 
3.5 Use and under-use of market facilities 
Market construction in the project was based on the design of a ‘model market’ with 
various facilities including: compartments; crushes; shaded areas; offices for 
veterinary and tax officials; detention pens; inspection points; tax check points; water 
supply; water troughs; feed troughs; loading ramps; public toilets; and stands for 
weighing scales. 
Detailed lists of the facilities constructed and the facilities which were actually in use 
in the 15 markets are provided in Tables 8 to 10. Overall, 27% of the market facilities 
were in use in the three regions. 
Some findings regarding the use of the market facilities were as follows: 
• Jijiga - all user groups were insisting to return to the old market site. 
• Aysayta - pastoralists were refusing to use the small ruminant compartments even 
though the municipality had deployed the police force to try to persuade people to 
use the facilities. 
• Eliwuha - the small ruminant market operated only for a few hours in the morning, 
due to lack of shade; the cattle and camel compartments in this market were used 
as storage areas for food aid and animal feed. 
• Logia - small ruminants were sold directly to lorry drivers and hotels, and so 
sellers did not see the need to use the new livestock market; plus, the new market 
14 
was located at some distance away and people preferred the old market; part of the 
infrastructure in Logia market had also been destroyed. 
Table 8. Construction and use of new market infrastructure in Afar region (number of open markets 
assessed = 4) 
Type of infrastructure Installed In use Utilized (%) 
Cattle compartment 
Camel compartment 
Sheep/goat compartment 
Cattle health check point 
Camel health check point 
Sheep/goat health check point 
Cattle tax check point 
Camel tax check point 
Sheep/goat tax check point 
Record office 
Veterinary post 
Sheep/goat water trough 
Cattle/camel water trough 
Sheep/goat feed trough 
Cattle/camel trough 
Detention pen 
Crush for cattle 
Loading and unloading ramp 
Public toilet (men and women) 
Water tanker 
Information board 
Market shade 
Total 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
4 
8 
8 
4 
4 
6 
98 
2 
2 
3 
0 
0 
0 
2 
1 
0 
0 
2 
0 
0 
0 
0 
0 
0 
4 
0 
0 
0 
5 
21 
50 
50 
75 
0 
0 
0 
50 
25 
0 
0 
50 
0 
0 
0 
0 
0 
0 
50 
0 
0 
0 
83.3 
21.4 
The open markets visited were Eluha, Chifra, Aysayta and Logia. 
• Bokoloboma - the volume of livestock transactions was far too low for the size 
and type of market which was constructed; the market handles only about 30 to 50 
small ruminants and up to 10 cattle per market day. 
• Finchewa - the camel compartment was abandoned due to a preference for the 
nearby bush market called Surupa (that was developing into a major camel 
market); the cattle compartment was used mainly for tax collection purposes as 
cattle sales actually took place about four kilometres away from the market 
compound, well before the militias start driving animals into the yard at noon. 
• Harobake - cattle, camel and equines were all sold outside the market compound 
before the militias start driving animals into the yard at around 11: 30 am; 
purchase of small ruminant took place early in the morning and the biggest 
complaint from small ruminant buyers was that the tax collectors do not show up 
before 11:30 am. 
• Lefa Issa - there are no camel supplies to the market; the supply of both cattle and 
small ruminants is too low to make use of the new facilities. 
• Hartisheikh – few animals were offered for sale on the day of the assessment visit 
and producers were trying to sell these animals in town rather than at the new 
market. Tax was not collected at the market site in Harisheikh and the assessment 
team was told that tax is collected in the town later from buyers. 
 
15 
Table 9. Construction and use of new market infrastructure in Oromia Region (number of open markets 
assessed = 6) 
Type of infrastructure Installed In use Utilized (%) 
Cattle compartment 
Camel compartment 
Sheep/goat compartment 
Cattle health check point 
Camel health check point 
Sheep/goat health check point 
Cattle tax check point 
Camel tax check point 
Sheep/goat tax check point 
Record office 
Veterinary post 
Cattle/camel weighing scale 
Sheep/goat weighing scale hanging post 
Sheep/goat water trough 
Cattle/camel water trough 
Sheep/goat feed trough 
Cattle/camel feed trough 
Detention pen 
Crush for cattle 
Loading and unloading ramp 
Public toilet for men and women 
Water tanker 
Information board 
Market shade 
Total 
6 
6 
6 
6 
6 
6 
6 
6 
6 
6 
6 
3 
21 
6 
6 
6 
6 
6 
6 
12 
12 
6 
4 
3 
163 
4 
3 
6 
0 
0 
0 
6 
3 
6 
0 
5 
0 
11 
0 
0 
0 
0 
0 
0 
4 
4 
3 
0 
3 
58 
66.7 
50 
100 
0 
0 
0 
100 
50 
100 
0 
83.3 
0 
52.4 
0 
0 
0 
0 
0 
0 
33.3 
33.3 
50 
0 
100 
35.6 
The open markets visited were Negele, Dubuluk, Teltele, Harobake, Finchewa and Bokoloboma. 
• Hudet - there were only three cattle in the yard on the date of the assessment visit; 
the woreda veterinary office had moved its office to the veterinary post at the 
market yard. 
• Dubuluk - the record office was used for tax collection. 
In the previous impact assessment in August 2008, some modifications were 
suggested for the more important secondary markets including: 
• relocating water and feed troughs inside the market yard; 
• partition for purchased animals within the market compartments; 
• construction of collection and guiding crush pens for the loading and unloading 
ramps; 
• extension of shades in the yards for protection against sun and rain. 
None of the above recommendations were implemented. 
The 2008 impact assessment also reported that marketing activity took place outside 
seven out of the 10 new markets which were assessed. The same assessment also 
noted the use of force to move people and livestock inside the markets, largely for the 
purpose of tax collection.
 
16 
Table 10. Construction and use of new market infrastructure in Somali Region (number of open 
markets assessed = 5) 
Type of infrastructure Installed In use Utilized (%) 
Cattle compartment 
Camel compartment 
Sheep/goat compartment 
Cattle health check point 
Camel health check point 
Sheep/goat health check point 
Cattle tax check point 
Camel tax check point 
Sheep/goat tax check point 
Record office 
Veterinary post 
Cattle/camel weighing scale 
Sheep/goat weighing scale hanging post 
Sheep/goat water trough 
Cattle/camel water trough 
Sheep/goat feed trough 
Cattle/camel feed trough 
Detention pen 
Crush for cattle 
Loading and unloading ramp 
Public toilet for men and women 
Water tanker 
Information board 
Market shade 
Total 
5 
4 
5 
5 
4 
5 
5 
4 
5 
5 
5 
3 
6 
5 
5 
5 
5 
5 
5 
10 
10 
5 
5 
6 
127 
2 
2 
5 
0 
0 
0 
2 
2 
2 
0 
3 
0 
4 
0 
0 
0 
0 
0 
0 
0 
3 
1 
0 
4 
30 
40 
50 
100 
0 
0 
0 
40 
50 
40 
0 
60 
0 
66.7 
0 
0 
0 
0 
0 
0 
0 
30 
20 
0 
66.7 
23.6 
The open markets visited were Jijiga, Hartisheikh, Lefa Issa, Hudet and Moyale. 
In June 2010 some trading activity took place outside six out of nine of the more 
active markets; details are provided in Table 11. 
Table 11. Livestock sales inside and outside the more active livestock markets 
Market 
location 
Livestock sold inside the new markets Livestock sold outside the new markets 
Finchewa 
Harobake 
Dubuluk 
Aysayta 
Moyale 
Negele 
Chifra 
Teltele 
Cattle, equines, small ruminants 
Camel, cattle, equines, small ruminants 
Camel1
, cattle, small ruminants 
Cattle, camels, small ruminants 
Cattle, camels, small ruminants2
Cattle, camels, small ruminants 
Cattle, camels, small ruminants 
Cattle, small ruminants 
Cattle, equines3
Camel, cattle, equines3
Cattle, camels 
Small ruminants4
 
Small ruminants, cattle, camels5
None 
None 
Small ruminants 
1 Only six camels were brought to the market on the day of the visit. 
2
 Small ruminants were not on sale on two consecutive visiting days in 2010. 
3 For equines the tax was collected outside the market. 4
 For these sales the animals do not go into the market at all and tax is collected only from the traders 
who truck at loading time. 
5 Sheep and goats were sold at around 3 km from the market on the Oromia side of the town; cattle and 
camels were sold outside the market when the market was closed. 
This finding, together with the increase in bush markets, reinforces the notion that 
market infrastructure is not necessary for livestock trade to take place. This is an 
17 
important point in terms of impact assessment and explains the flaw in the ‘causal 
pathway’ for the project – there is no cause and effect in terms of new market 
infrastructure resulting in more livestock trade in pastoralist areas. 
As in August 2008, in June 2010 the use of force was needed to move people into 
markets for the payment of taxes. 
3.6 Service provision around the new markets 
Services provided in and around the markets were assessed and included: 
• sales of tea, soft drinks and food by women and girls in all markets; sales of 
alcoholic drinks, particularly in Harobake, Finchewa and Teltele markets in 
Borana; 
• sales of cold water by children, particularly in the Aysayta market in Afar; sales of 
soft drinks, cigarettes and rope in all other markets by children; 
• looking after trade herds and gesso/mora14 renting service provided by boys in 
Eliwoha and Harobake markets. 
Due to the issues described in the previous sections of the report, it was difficult to 
determine the specific impact of the market construction on these groups. For 
example, in sites where there had previously been a market (such as Moyale and 
Jijiga), there were already various local service providers before the project started. 
Also, the service providers in and around a given market will be affected by the level 
of livestock trade taking place, with less trade leading to fewer potential customers. 
The situation is dynamic and varies by market. In some markets the trade is vibrant 
and in others, declining. In both cases, trade activity is not attributable to the physical 
presence of new market infrastructure but to other, non-project factors. 
Despite these caveats there were examples of project activity which seemed to benefit 
women service providers. For example, in Moyale the project assisted women to 
operate closer to the market yard and sell food. The income of these women increased 
by around 36% between 2009 and 2010, though it was difficult to determine what 
proportion of this increase was directly attributable to the project. The project also 
organized women into cooperatives and was to provide EB 15,000 (US$ 1,050) seed 
money to each one. The use of the seed money was not assessed, but as explained 
above, the profits from service provision around the markets would partly depend on 
the levels of livestock trade activity and the number of buyers and sellers. 
3.7 Unexpected impacts 
Some of the unexpected impacts of constructing the markets were as follows: 
• Taxation in bush markets – taxation has now been introduced to bush markets in 
Borana, including for unsold animals, whereas previously there was no taxation in 
these markets. The kebeles also collected about EB 10 to 15 from each service 
provider selling food, drinks and other consumables, and no receipts were issued. 
• Human sanitation – most of the new markets visited were in a poor state of 
hygiene. Human excreta was seen everywhere. In Hartisheikh, water was 
 
14 Geeso (Afar) and mora (Oromiffa) are fences constructed by local people, especially young boys, 
who then rent the areas as yards for purchased cattle and small ruminants. 
18 
stagnating in the troughs and would act as breeding sites for mosquitoes. The issue 
was raised to the woreda cabinet by the assessment team and they stated that this 
particular problem was the result of the troughs having no drainage holes. 
• Market construction by regional governments - encouraged by higher authorities 
during the national Pastoral Day celebration in Jijiga, regional governments were 
constructing additional livestock market infrastructure in areas where such 
facilities do not exist; four markets had already been completed in Afar Region 
and there were similar plans in Somali and Oromia regions. These facilities are 
unlikely to provide benefits to market actors, given the minimal usage and the 
destruction of some market facilities already installed by the project. The sites 
where such facilities are being constructed may not have the required volume of 
livestock supplies to make use of the facilities. It is also likely that some of the 
design faults which were evident in the project markets would be replicated in the 
new sites. As with the project markets, the local capacity to manage new markets 
remains questionable. 
4. Lessons and Recommendations 
4.1 Project design issues: pastoralist marketing behaviour and market access 
The design of the ACDI/VOCA project assumed that pastoralists were not market￾orientated, but would sell more animals if new market infrastructure was provided. An 
evaluation of the ACDI/VOCA project in March 2008 concluded that the underlying 
logic of the project was flawed: 
‘… the project design was founded on assumptions about how pastoralists 
marketed their animals. These assumptions are mostly several decades out of 
date, having been revised by development and research evidence about 
Ethiopian and other African pastoralists’15.
This statement by the evaluators drew on USAID-funded research in east Africa 
which demonstrated that the livestock marketing behaviour of pastoralists varies by 
wealth group, and that an understanding of this behaviour is central to the design of 
pastoral development initiatives. The same body of research also challenged persistent 
myths on pastoral livestock marketing. For example: 
• In Afar, Borana and Somali pastoralist areas, household economies are 
characterized by high, direct consumption of livestock milk and the sale of 
livestock to acquire cereals16. This means that almost every pastoralist household 
has access to traders, who enable the livestock-cereal exchange. In other words, all 
pastoralists already have market access in one way or another. 
• Livestock marketing behaviour varies by household wealth. The main strategy for 
economic growth pursued by poorer households is to build financial capital and 
this means building livestock herds. In the absence of alternative economic 
opportunities this is a highly rational strategy. It also means that poorer 
households only sell livestock when they need to acquire cereals or have other 
 
15 Behnke, R., Kerven, C. and Teshome, A. (2008). Evaluation of USAID Pastoral Development 
Projects in Ethiopia. USAID, Addis Ababa. 
16 E.g. see livelihoods baselines of LIU. 
19 
acute needs for cash. Such households have had market access for decades and the 
creation of new markets in urban centres will not affect the marketing behaviour 
of such households. They will continue to sell a minimum number of animals in 
order to maximise herd growth, and by so doing, develop their financial assets17. 
• Middle wealth and higher wealth group households depend more on livestock 
sales to meet annual cash and cereal demands, and sell relatively more animals. 
These groups will respond to increased market demands (e.g. growing export 
markets), but simple bush markets can adapt to handle any increased trade. As the 
export trade grows, the main beneficiaries are the various traders and service 
providers around the business, and relatively wealthy and middle-wealth group 
pastoralists.18 
 
Despite the large volume of research and evaluations to demonstrate the above points, 
there is still a common misperception about pastoralists - that they will sell more 
livestock with improved market infrastructure and access to market information. This 
perception has led to investments in infrastructure since the early 1970s by bilateral 
and multilateral agencies with no effective outcomes. 
More recently there has been a focus on the promotion of livestock exports, based on 
a belief that increased demand for livestock provides an opportunity for pastoralists to 
benefit by selling more livestock and at a better price. The flaw in this assumption is 
the notion that all pastoralists own surplus livestock available for sale at an opportune 
time. This assumption ignores two fundamental facts. The pastoral system is 
dependent on continuous ‘re-building of herd sizes’ in normal times in preparation for 
lean years. As such, what is considered as surplus by outsiders is regarded as survival 
stock by pastoralists. Regardless of the prevailing market demand, the system, 
therefore, limits the number and species of livestock each pastoralist can sell in a 
given year. Secondly, the number of animals pastoralists can sell in a given year is 
determined primarily by their wealth status (in terms of stock ownership) followed by 
the level of dependence on livestock for annual household income. The two factors 
are directly co-related. 
Better-off and middle income groups depend more on livestock for annual food and 
cash requirements. They therefore have the need and the capacity to sell more animals 
and even to respond positively to increased market demand, such as exports. 
Conversely, poor and very poor households to a varying degree depend on livestock 
and non livestock sources to meet their annual household needs. These two groups 
can only sell limited number of animals in a given year regardless of the market 
demand because they own fewer animals to begin with. The fact is that so many 
pastoralists are in poorer wealth categories limits the extent to which increased export 
demands, market infrastructure or market information will reduce poverty. In southern 
Ethiopia in 2008, the middle and better-off pastoral wealth groups sold respectively 
six and 18 times more small ruminants than the very poor. Similarly, in Guji areas, the 
 
17 These issues are well described in research studies funded by USAID e.g. McPeak, J.G. and Little, 
P.D. (2006). Pastoral Livestock Marketing in Africa: Research and Policy Challenges. IT Publications, 
Rugby 
18 Aklilu, Y. and Catley, A., (2009). Livestock Exports from Pastoralist Areas: An analysis of benefits 
by wealth group and policy implications. Report for the Intergovernmental Authority for Development 
and the Food and Agriculture Organization. Tufts University, Addis Ababa. 
https://wikis.uit.tufts.edu/confluence/display/FIC/Livestock+Exports+from+the+Horn+of+Africa  
20 
middle and better off-income groups sold respectively six and twelve times more 
small ruminants than the very poor (Table 12). 
While in both cases, middle income and better-off groups raised all their cash income 
from livestock sales, the poor and the very poor generated their annual cash income 
by complementing livestock sales with charcoal production, labor and safety net 
programmes to varying degrees. 
Table 12.  Annual household income and number of livestock sold by wealth group, southern
Ethiopia
Area Pastoral wealth group income
Very poor Poor Middle
income
Better‐off
Teltele, Dilo and Dier
Equivalent sheep or goats sold
US$114
5
US$202
8.5
US$714
31
US$2,100
92
Guji – Borana
Equivalent sheep or goats sold
US$132
5.5
US$231
10
US$768
34
US$1,500
66
Source: Livelihoods Information Unit, Ethiopia, (2008).
Table 12 implies that the real problem in pastoral areas is the widening wealth gap 
between the better-off and the poor groups. In other words, while wealthy and middle￾wealth groups seem to retain their livestock, poorer groups are characterized by 
decreasing livestock holdings and as human population grows, more households fall 
into the poorer categories. These trends were either described or predicted in pastoral 
areas of Ethiopia from the 1970s, and were explained by a combination of human 
population growth, reduced access to grazing land, commercialization of livestock 
production and marketing, and a result of these factors, the increasing negative impact 
of dry seasons and droughts on livelihoods. 
Recommendation 1
Attempts to intervene in pastoral livestock marketing systems should document local 
marketing behaviours by wealth group and carefully examine any proposed causal 
pathway that associates more market infrastructure with more sales. Such pathways 
are unlikely in pastoralist areas. Policy engagement and economic analysis with 
government partners is needed to raise understanding of these issues at federal and 
regional levels. 
4.2 Project design issues: hard versus soft inputs 
The findings in this report on new bush markets in Oromia are consistent with 
experiences from various countries dating back to the 1970s. For example, when 
describing how Somalia’s pastoral economy was responding to profitable export 
opportunities in the 1970s, in terms of bush markets it was noted that: 
‘The basic requirements, so far as market facilities are concerned, appear to 
be ample space for transactions and the provision of drinking water for 
market users. Little justification for investment in fencing, pens, scales or 
21 
auction rings can be established since the system appears to work well in its 
present simple, highly flexible form’
19. 
Similarly, when describing pastoralist livestock marketing more generally in Africa in 
1982: 
‘Existing traditional livestock marketing systems are perceived to suffer from 
inefficiencies, abuses in market conduct and technical imperfections the relief 
of which would benefit the pastoral system. But, studies of traditional 
marketing systems show that, in spite of being complex and traditionally 
based, they generally perform well their function of distributing livestock and 
meat products at reasonably low costs. 
Efforts to develop pastoral production systems via marketing therefore have to 
take into account 
‐ that the structure and performance of existing traditional marketing is 
generally satisfactory; 
‐ that the existing system does not appear to discourage production and 
supply of livestock from the dry areas.’20
In 2008, policy makers from the AU, COMESA, IGAD and member states visited 
Garissa in Kenya, one of the largest and most active livestock markets in east Africa, 
and situated in a pastoralist area. This is an efficient market, but has no perimeter 
wall, no compartments, and very limited shade. The only infrastructure of note is a 
series of loading ramps, and a few concrete water troughs. The market is surrounded 
by small shops selling tea and other items, and private suppliers provide livestock 
fodder. Between 1989 and 2007, the number of cattle sold at this market increased 
from 24,395 to 105,667, with no substantial investment in market infrastructure. A 
presentation by the local government was titled “Garissa Livestock Market – A 
Treasure for Meeting the Millenium Development Goals in a Pastoral based 
Millenium Municipality”. 
 
Regarding basic maintenance of most of the markets which were visited during this 
assessment, it seems unlikely that sufficient funds will become available. In the more 
active markets income has not been reserved for maintenance needs whereas in the 
less active markets, the volume of trade is unlikely to yield enough revenue for 
running costs. It also seems likely that more bush markets will open, and this trend 
would further reduce the revenue generated from the current markets. After the cost 
extension to the project, we see a general scenario in which rehabilitation work will 
not be carried out, and, the draft marketing guidelines will not be implemented. 
Regardless, and more importantly, livestock transactions will carry on with or without 
the new market facilities. 
Future support to infrastructure development in pastoralist areas could consider the 
wider context, in which infrastructure inputs are seen as solutions to a range of 
problems but with varied impacts. Some general experiences include a focus on the 
 
19 Reusse, E. (1982), Somalia’s Nomadic Livestock Economy: Its response to profitable export 
opportunity. World Animal Review 43, 2-11. 
20 Jahnke, H.E. (1982), Livestock Production Systems and Livestock Development in Tropical Africa. 
Kieler Wissenschaftsverlag Vauk, Kiel, p90. 
22 
“hard inputs” such as buildings (e.g. health, education, veterinary) with less attention 
to the long-term management of facilities, and unclear capacities and commitments of 
government to cover staff and operational costs. Partly for these reasons, alternative 
systems of service delivery are being tested in pastoralist areas. To some extent, the 
simple bush markets that are set up and operated by local traders fit the philosophy of 
alternative but locally-acceptable approaches. Weaknesses in government 
management and administrative capacities in pastoralist areas are widespread, cross￾sectoral and not specific to livestock markets. When a facility has potential to 
generate income, there will be tendencies for individuals to develop informal systems 
to appropriate some or all of this income for private benefit; accountability 
mechanisms are weak (e.g. veterinary clinics, health posts, markets). 
The assessment team recognised the effort made by the implementer to introduce 
local management systems for the new markets, with a view to enhancing use of the 
facilities and sustaining the inputs. However, there has been limited success. Of the 15 
markets visited, progress towards local management was limited to only two markets, 
in Harobake and Finchewa. Even in these two markets, there were still a number of 
outstanding management issues to be resolved. This level of progress can also be 
viewed from the perspective of the start date of the project, in October 2005 and the 
construction of a total of 25 markets. 
The assessment team concluded that the issues surrounding the management of the 
new markets are in most cases, too complex to resolve. Many of the facilities were 
inactive and the non-operational markets are vandalized beyond repair. This implies a 
lack of interest by local officials and communities in the facilities. In the more active 
markets (relatively speaking) there appeared to be more interest in what the markets 
generated for the respective revenue offices and in a number of cases, for private 
groups or individuals who monopolized certain services. The crumbling condition of 
some of the facilities in the nine more active markets also underlines that it will not be 
an easy task to set up proper management systems for the maintenance of the 
facilities. 
Recommendation 2
Livestock markets in pastoralist areas require basic facilities, such as loading ramps 
and water troughs. Investment is more elaborate infrastructure is only warranted if a 
positive pre-project benefit-cost analysis shows an appropriate return on investment. 
In addition, a pre-project assessment of local government management and 
administrative capacities is needed, including analysis of the actual capacity of 
government to deliver versus their stated capacity e.g. as expressed in project 
agreements. In many cases, the key constraints to pastoral livestock markets will be 
policy and institutional (e.g. inappropriate taxation; restrictive livestock movement 
regulations), rather than hardware. Support to government to improve understanding 
of these issues and build capacity for benefit-cost analysis and organizational capacity 
analysis is warranted. 
4.3 Project design issues: community participation 
The emergence of bush markets in southern Ethiopia indicates that local actors can set 
up and run markets in the absence of external support, other than roads and mobile 
23 
phone networks. In contrast, the mis-positioning of new government markets was a 
common cause of conflict in pastoralist areas. 
The experiences from the ACDI/VOCA project show that local analysis with a limited 
range of actors can lead to misdiagnosis of problems, and inappropriate advice on 
what types of markets and infrastructure are needed, and where. Implementers need to 
be aware of the incentives behind certain requests, including political incentives and 
opportunities for individuals to profit from construction contracts. This type of 
awareness requires experience, and a willingness and capacity to engage local 
communities in meaningful participatory dialogue and analysis. Part of the dialogue 
should include ways to manage systems in the long term, and appropriate locations for 
new infrastructure. In general, NGOs with long-term presence in pastoralist areas are 
far better able conduct this kind of analysis with communities and other stakeholders, 
relative to new implementers. 
Recommendation 3
In addition to a capacity to conduct economic analyses and understand local 
marketing behaviours, implementers also need to understand local cultural, social and 
conflict dynamics, and recognize the need to consult a range of local actors and 
stakeholders. In conflict-affected or conflict-prone areas, ‘Do No Harm’ approaches 
are needed, as already recommended and used by other USAID implementers in 
pastoralist areas. When selecting implementers, USAID needs to consider past 
experience specifically in pastoralist areas, and understanding and capacity to conduct 
conflict analysis and implement conflict-sensitive programming. 
4.4 Project design issues: a framework for analysis 
Under the PLI project, a set of key information needs have been identified if new 
approaches are to be scaled-up (Table 13), and these generic questions apply to 
livestock markets. Further analytical tools are available, such as the Emergency 
Market Mapping and Analysis Toolkit21. 
Recommendation 4
USAID supports ‘scaling-up’ of new approaches based on evidence. In the case of 
livestock markets in pastoralist areas, a substantial body of evidence already exists. 
Future needs include building the capacity of government and implementers to 
conduct appropriate, comprehensive pre-project analysis using mainly existing 
information, but supported by rapid applied assessments as needed. 
 
 
21 Albu, M. (2010) Emergency Market Mapping and Analysis Toolkit. Practical Action Publishing, 
Rugby. 
24 
Table 13: Information needed for scaling-up development approaches in pastoralist areas 
Type of information and 
analysis 
Key questions 
Local impact on 
livelihoods 
Quantitative and 
qualitative 
Has the approach achieved its expected impact in terms of livelihoods 
benefits? For example, did a community-based health project achieve health 
outcomes (improved human capital)? Did a livestock marketing group 
increase the financial capital of group members? Other questions include 
the sustainability of the benefits, especially if the main external financial or 
technical support is withdrawn. 
Cost-benefit analysis 
Quantitative 
Community-based approaches often require considerable technical or 
financial support from NGOs during the pilot phase. Cost-benefit analysis 
helps to examine economic efficiency and the feasibility and economic 
rationale for scaling-up. A challenge here is that benefits related to social 
capital can be difficult to quantify or value in economic terms. 
Policy and institutional 
issues 
Qualitative 
To what extent does the approach complement or contradict government 
policy or legislation, whether formal or informal? Is the approach legal? 
Would formal procedures or laws need to be adapted for scaling-up, and if 
so, are such changes realistic? What is the risk of conflict undermining the 
approach? 
Area-based economic 
potential 
Quantitative 
For approaches which depend mainly on private sector activity, what 
volume of economic activity is likely to be supported in a given area? For 
example, how many private community-based workers can an area support? 
How many livestock marketing groups are needed? These types of 
questions assume that not everyone can be a small trader, run a teashop, or 
sell hides and skins, and that, in pastoralist areas, the key financial asset is 
livestock, not cash. 
Technical capacities 
Quantitative and 
qualitative 
Good community-based systems and services often depend on skilled and 
experienced development workers who are familiar with participatory 
approaches. Scaling-up requires these capacities to be institutionalized, with 
associated incentives. Are such changes feasible, and if so, what are the 
costs and level of technical assistance needed, and for how long? 
 
 
 
25 
Annex 1: Question checklist for the impact assessment 
1. Infrastructure status 
 Infrastructure still in use (type and number) 
 Dilapidated infrastructure that needs repair 
 Repaired infrastructure and /or repair plans 
 Additional infrastructure since last survey (built by?) 
 2. Taxes and fees 
 Taxation rate for sold/unsold animals by species 
 Changes in taxation rate from last survey 
 Fees for use of facilities and services 
 Changes in fee rates from last survey 
 New taxes and service fees introduced since last survey 
3. List agencies, associations and individuals collecting taxes and fees (by type) 
4. Has a Marketing Management Board been established for the specific market? 
 Composition of members / structure of the Board 
 Main mandates of the Board including responsibilities 
 Is the specific market being managed by the Board? 
 Initiatives taken by the Board, so far 
a) In setting up a proper management system in place 
b) Resource collection and allocation 
c) Proportion of taxes and fees allotted for improvement / 
maintenance of infrastructure 
d) Additional amenities planned (e.g. water points, shops, tea rooms 
and restaurants, veterinary facilities, etc) 
e) How does the board intend to get resources for d above? 
6. Or, is the market still run by the local Council? 
7. Perceptions of pastoralists, traders and middlemen 
 Changes they see since the last survey 
In terms of service provision (list) 
 In the maintenance of infrastructure 
 In orderliness of transactions 
 In the use of facilities (scales, loading ramps..etc) 
 In the provision of veterinary services / inspection 
 In protection / security 
In additional amenities / facilities set up by the council or individual 
initiative 
 In taxation rates and fees 
 Other improvements, please list 
 Drawbacks they perceive in terms of importance (please list) 
 Suggestions for further improvements 
8. Perceptions of other service providers (shop owners, tea sellers, drug vendors..etc) 
 Positive attributes 
 Negative attributes 
 Suggestions for improvements 
9. Sales volume 
Changes in the number of species brought to the market - quantify and 
attribute 
 Changes in the number of species sold - quantify and attribute 
 Changes in livestock prices - attribute 
26 
8. Records 
 Does the Board or the Council keep records of transactions? 
 If yes, what types of records? 
 Reliability of record keeping 
9. Veterinary services 
 Is there any veterinary service provision close to the market center? 
 Who provides the services (public or private?) 
 In the case of private service providers, please state their professional level 
 What are the main types of services provided? 
 Rate of payment for services 
 Are pastoralists/traders satisfied with the services provided? 
10. Environmental impacts (list if any including proposed remedies) 
11. Does the new management system put in place ensure the sustainability of the 
specific market compared to the system during the last survey? 
 Please provide details. 
12. Comments on the Regional livestock marketing guideline 
13. Lessons 
 Important lessons drawn to be replicated elsewhere 
 Main drawbacks to be avoided elsewhere 
14. Specific to Shinile Market 
Assess the impacts brought about by the agreement entered between the 
project and livestock exporters for the exclusive use of the market facility in 
Shinile by the latter.