Agricultural extension in Ethiopia: the case of 
participatory demonstration and training extension 
system 
KASSA BELAY1 
ABSTRACT 
This paper examines the participation of farmers in the Participatory 
Demonstration and Training Extension System and looks 
into the principal barriers to the adoption of modern agricultural 
inputs. The paper is based on a review of the literature and an 
analysis of data collected from 1482 household heads selected from 
16 sites in four regional states, namely, the Amhara National 
Regional State, the Oromia National Regional State, the Southern 
Nations, Nationalities and People's Regional State and the Tigray 
National Regional State. The historical review reveals that 
extension programmes and policies have been formulated without 
due consideration to the farmers' opinion and traditional 
knowledge system. The various extension approaches have been 
biased against the livestock subsector and research and extension 
activities have been carried out by different organizations without 
proper co-ordination which, in most cases, has led to redundancy of 
effort and wastage of resources. Both the historical review and the 
survey results reveal that extension service coverage in the country 
has been very low, the research-extension linkage has been very 
poor and extension agents have been involved in different activities 
which are not related to their normal duties. The study also makes it 
clear that a host of factors, some of which are policy related, were 
1 Department of Economics, Alemaya University, Ethiopia 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 49
responsible for the low rate of adoption of modern agricultural 
inputs in the country. 
Introduction2 
Ethiopia, with a total area of 1.1 million km2, has an estimated population 
of 61.7 million (MEDaC 2000). Agriculture is the mainstay of the Ethiopian 
economy. It generates over 45 per cent of the GDP and 90 per cent of the 
total export earnings of the country. It is also estimated that agriculture 
provides employment for about 85 per cent of the labour force. The 
striking differences among regions in altitude, topography, soils, climate 
and fanning systems enable the country to produce a variety of field 
crops, fruit and vegetables and rear different species of livestock. 
However, Ethiopian agriculture is characterized by low productivity 
and over the last two decades it has been unable to produce sufficient 
quantities to feed the country's rapidly growing population. In fact, 
natural calamities and famine seem to have become the salient feature 
and permanent problem of the country. For instance, in 1973—1974 and 
1984-1985 the country experienced the worst droughts in living memory 
that claimed the lives of several ^thousands of people. 
One of the reasons for the existing structural food insecurity in the 
country is the low level of technological development, which acts as 
the principal barrier to the efficient utilization of the country's natural 
resources. Even though different extension approaches have been 
implemented, experiences over the past four decades have not made 
economists particularly sanguine with respect to bringing major impacts 
on the productivity of smallholders and the utilization of modern inputs. 
For instance, the adoption rate of modern agricultural technologies in 
the country is very low. On this point, MEDaC (1999) points out that the 
Ethiopian fanner continues to use low fertilizer rates which are estimated 
2 Paper presented at the National Workshop on Technological Progress in 
Ethiopian Agriculture, organized by the Department of Economics, 
Faculty of Business and Economics, Addis Ababa University and 
USAID/Ethiopia, held in Addis Ababa, Ethiopia, November 29-30,2001 
50 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
to be an average of 7 kg. of nutrients per hectare of arable land, compared 
with a sub Saharan average of about 9 kg. nutrients per hectare of 
arable land. The world average stood at 65 kg. per hectare. The physical 
application rates of fertilizer by most peasant farmers are well below 
those recommended by the extension programme (100 kg. DAP 
[diammonium phostphate] and 50 kg. urea per hectare) and could in 
some cases be as low as 20-30 kg. per hectare. Daniel et al. (1997) 
reported that only very few Ethiopian farmers use improved seeds: 
about 5-10% of total seed used for maize and even less for sorghum, 
barley and teff (Eragrostis Abyssinica). Similarly, MOA (1998a) points 
out that, of the 3.5 million hectares of the potentially irrigable land only 
161,010 hectares (4.6%) are currently under irrigation. 
A closer look at the different extension approaches reveals that they 
have been planned and implemented without the participation of the 
very people for whom they have been designed. Apart from being biased 
against the livestock subsector, these approaches have captured only 
fanners located a few kilometres from both sides of all-weather roads 
(Belay 1998). With the exception of a few non-governmental 
organizations engaged in agricultural development activities and other 
governmental bodies such as the Coffee and Tea Development 
Authority, the defunct Ministry of Natural Resources and Environmental 
Protection and the Ministry of State Farms' Development, which had 
been doing some extension work, the Ministry of Agriculture has been 
the sole authority responsible for the national agricultural extension 
system. 
The objectives of this paper are: to review the different extension 
approaches used in the country over the past 50 years; to assess the 
participation of farmers in the Participatory Demonstration and Training 
Extension System and to identify the most important barriers to the 
adoption of modern agricultural inputs. The paper is organized in four 
sections. Section I reviews the history of extension work in Ethiopia 
and the different approach used so far. Section II provides a brief 
discussion on the method of data collection and the study areas. Section 
III presents the results of the study. The final section summarizes the 
main empirical findings and draws appropriate conclusions. 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 51
I History of extension work in Ethiopia 
Agricultural extension work in Ethiopia began in 1931 with the 
establishment of the Ambo Agricultural School which is one of the 
oldest agricultural institutions in Ethiopia and the first agricultural high 
school offering general education with a major emphasis on agriculture. 
Apart from training students and demonstrating the potential effects of 
improved varieties and agricultural practices to the surrounding farmers, 
the school did not do extension work in the sense of the term that we 
understand today. It was with the creation of the Ministry of Agriculture 
in 1943 that the country witnessed the commencement of limited 
extension activities in different areas.3 Even then, as there was no 
separate division in the Ministry responsible for extension work, different 
services to farmers were made available by the various divisions of the 
Ministry. According to Haile Selassie (1959), the services rendered were 
more of a regulatory nature and included providing advice in soil 
conservation through the grow-more-trees campaign; better variety of 
seeds and seedlings; cleaning and seed selection; the protection of 
game fish; the preservation of hides and skins and so on. 
However, real agricultural extension work began in the early 1950s 
following the establishment of the Imperial Ethiopian College of 
Agriculture and Mechanical Arts (IECAMA, now Alemaya University) 
with the assistance of the United States of America under the Point 
Four Programme. The academic programme of the College was modelled 
on the Land Grant College system with three fundamental but related 
responsibilities; training high level manpower; promoting agricultural 
research and disseminating appropriate technologies. 
The role played by the IECAMA in developing the agricultural 
extension system is considerable. In fact, when the College was founded 
it was given the mandate to develop and deliver a national programme 
in agricultural extension. To this end, on October 1 1954, it employed 
two Ethiopians who had graduated from the Ambo Agricultural School 
3 Up to 1943 the responsibility for agricultural matters in Ethiopia was 
vested in the Ministry of Commerce, Industry and Agriculture. By a law 
issued early in 1943, a Council of Ministers and 12 Ministries were set 
up. One was the Ministry of Agriculture 
52 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
as extension agents. They were stationed at Assela (Arsi) and fitche 
(Shoa) to establish demonstrations with co-operative fanners. During 
the next few years the number of extension agents increased 
considerably and they were stationed at posts all around the country. 
According to Huffnagel (1961), from the beginning of the extension 
service efforts were made to obtain men who had, at least, a basic 
knowledge of Ethiopia's agriculture. An eighth grade education was 
the minimum requirement for the first selected groups of agents and 
trainees. From later candidates a twelfth grade education—either at 
Jimma or Ambo [agricultural high schools]—was demanded. 
Prior to their deployment, agents were trained in various fields 
(such as modern techniques of crop production, modern animal 
husbandry practices, the operation and care of small, improved farm 
implements) and extension methodologies. Moreover, they took part 
in in-service training on a regular basis. The major extension activities 
were concentrated in areas where the college had experimental stations. 
These included the main campus at Alemaya, the central experiment 
station at Debre Zeit and the Jimma Agricultural and Technical School. 
By 1963,77 extension posts had been established with a total of 132 
nationals servicing the various areas. These agents were- actively 
engaged in demonstrating and helping fanners use new techniques 
in tools and machinery, insect and disease control and improved 
practices in the production of livestock and crops; paying regular 
visits to individual farmers; organizing and holding adult educational 
meetings and field days and encouraging the formation of agricultural 
youth clubs. 
In August 1963, the imperial government transferred the mandate 
for agricultural extension from the College to the Ministry of 
Agriculture, with the suggestion that the IECAMA concentrate its 
outreach efforts to help fanners in the vicinity of the College. Since 
this time the Ministry of Agriculture has been responsible for national 
extension activities. 
Extension approaches 
A review of extension activities during the last 50 years reveals that 
a range of extension approaches has been used. The approaches 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 53
tended differ with each successive political regime. In what follows the 
extension approaches used by the different political regimes are 
presented separately. 
Under the imperial regime 
Following the transfer of the responsibility for national extension 
administration to the Ministry of Agriculture, extension service became 
one of the departments in the Ministry. When the Ministry decentralized 
its departments by establishing provincial offices, extension provincial 
supervisors were appointed in all the 13 provinces of the country (with 
the exception of Eritrea). Each supervisor was in charge of between six 
and eight extension agents. The latter were located along the main 
roads in the country and covered an area up to 25 or 30 km from their 
headquarters. Stommes and Sisaye (1979) state the situation of 
agricultural extension in the 1960s succinctly: 
Since more than 60 per cent of the peasant population live at least 
a half-day's walk from all-weather roads and since the few extension 
agents had been assigned along all major highways, there was 
relatively little contact between extension agents and farmers. 
The feudal nature of the social structure in rural areas had also 
limited contacts and advice of the extension agents to big landlords 
and influential farmers. 
Up until the middle of the 1960s, policymakers paid little attention to 
the development of peasant agriculture. For instance, during the First 
Five-Year (1957-1961) and the Second Five-Year (1963-1967) 
development plans, despite its importance to the national economy, 
agriculture received only 13.7 per cent and 21.3 per cent of the total 
investment, respectively. Even worse, almost all the investment allotted 
to the agricultural sector was channelled to the expansion of large-scale 
commercial farms engaged in the production of cash crops for export 
and raw materials for local industries. 
Following the increased realization of the continued stagnation of 
agriculture and pressure from international aid donors, it was only in its 
Third Five-Year development plan (1968—1973) that the government 
gave formal recognition to the peasant sector and made attempts to 
54 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
modernize it. However, considering the fact that the country's trained 
manpower, material and Financial resources were insufficient to 
modernize peasant agriculture in all areas of the country simultaneously, 
the government opted for the comprehensive package approach. This 
involved the removal of barriers to production by concentrating efforts 
in strategically-selected areas in which results could easily be seen. 
In the Ethiopian context, the comprehensive package approach 
involved the co-ordinated application of different but fundamentally 
related strategies, such as improving the existing infrastructure, 
dispensing better and well organized social service and providing 
effective transportation, marketing and credit services, as well as 
popularizing appropriate, well-tested and locally-adapted improved 
agricultural technologies. The rationale for the comprehensive package 
approach was that progress made in selected sites would have multiplier 
effects on the surrounding areas by way of demonstration and as a 
result of social interaction. 
The first comprehensive package project, the Chillalo Agricultural 
Development Unit (CADU) was established as an autonomous entity 
in the Arsi region south of Addis Ababa in September 1967 and was 
financially backed by the Swedish International Development Authority 
(SIDA). CADU aimed at a general socioeconomic development. Towards 
this end it integrated planning, credit and marketing facilities, price 
stabilization, mechanization, research into inputs and intermediate 
technologies and training local project employees. The method CADU 
adopted in reaching the peasants was basically that of demonstration. 
The project region was divided into extension areas where agricultural 
extension agents and model farmers demonstrated the effects of new 
agricultural techniques. The extension agents cultivated demonstration 
plots. The model farmers, selected by the peasants in the neighbourhood, 
were provided with fertilizers, improved seeds and improved farm 
implements and were instructed by the agents. Field days were held 
frequently on the agents' demonstration plots and on fields of the 
model farmers so that the rest of the peasants could then compare the 
yield from their own traditional methods with the yield resulting from 
the new techniques applied by the project. 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 55
Based on the experience gained from CADU, in the following years, 
other autonomous comprehensive package projects with varying 
objectives and approaches were initiated with the financial assistance 
obtained from different countries. These included the Welamo 
Agricultural Development Unit; the Ada District Development Project; 
the Tach Adiabq and Hedekti Agricultural Development Unit in the 
northwest of Tigray; the Southern Region Agricultural Development 
Project in the vicinity of Awassa town; and the Humera Agricultural 
Development. However, it was only CADU that was fully operational 
until it was phased out in 1986. 
It was soon realized that the comprehensive package projects failed 
to serve the very people for whom they were destined, the tenants and 
small-scale farmers. Most importantly, the principal beneficiaries were 
landlords and commercial farmers who reaped almost all the services 
rendered. In evaluating the experience from CADU, Schulz (1981), 
underlined the-fact that the distribution of CADU loans between tenants 
and landowners has always been biased in favour of owners and so, 
proportionately, there have been roughly only half as many tenants on 
the credit list as there are in the target population. Other authors have 
shown that, by encouraging the process of mechanization in larger 
commercial farms, the package projects accelerated the eviction of 
tenants (Betru 1975, EPID 1970, Mengisteab 1990, Task Force on 
Agricultural Extension 1994a,Tesfai 1975). 
It also became apparent that the comprehensive package projects 
were too expensive, both financially and in terms of trained manpower 
requirements, to warrant replication in other areas of the country. As a 
result, in 1971 the government, in co-operation with SID A, designed an 
alternative strategy envisaged to be compatible with the availability of 
resources called the Minimum Package Project I (MPP-I). MPP-I was 
prepared for the 1971 -1974 period and was designed to provide smallscale 
fanners with services considered to be the minimum essential 
elements for agricultural development (Mengisteab 1990, Schulz 1981). 
These included agricultural credit, marketing and extension advice, 
including the dissemination of innovations such as fertilizers and highyielding 
hybrid seeds. 
56 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL t8 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
It was also in 1971 that the government established the Extension and 
Project Implementation Department (EPID) in the Ministry of 
Agriculture. EPID was commissioned to administer the minimum package 
projects and supervise the activities of comprehensive package projects. 
The MPP-I was supposed to reach a large number of farmers by making 
use of the technologies generated and tested by the comprehensive 
package projects. As to its method of technology transfer, it employed 
an individual farmer extension approach, where both model farmers and 
extension agents demonstrated the importance of improved techniques 
of production. An extension agent under MPP-I was expected to cover 
an extension area of about 10 to 15 km along an all-weather road and 
about 3 to 5 km (but sometimes up to 10 km) on both sides of the road. 
Each extension area had a marketing centre and a crop demonstration 
plot and agents sold fertilizer and seeds on credit. Five extension areas 
constitute a full-fledged MPP area. Each MPP area, which extends over 
75 km, was designed to serve about 10,000 farm families. 
Though EPID was able to provide agricultural services in 280 of the 
580 districts and some improvements were made in terms of the adoption 
of improved inputs, MPP-I failed to have a significant impact on the 
agricultural sector because the government was reluctant to put in place 
the necessary reform measures in the areas of land tenure, tenantlandlord 
relationships and the organizational and administrative systems 
of the different institutions entrusted with agricultural development of 
the country (Harbeson 1990, Mengisteab 1990, Schulz 1981). For 
instance, the quality and extent of research work aimed at developing 
technological packages adapted to the different ecological zones of the 
country fell below expectations. Moreover, as extension activities were 
concentrated in areas where mixed farming system prevails, MPP-I made 
a very marginal contribution to those farmers in the lowland areas 
engaged in animal production. Tesfai (1975) points out that, just as in 
the case of comprehensive package projects, the principal beneficiaries 
of the MPP-I were wealthy farmers who had access to modern inputs. 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 57
Under the military regime 
Following the 1974 revolution, the new military regime enforced land 
reform on 4 March 1975. The land reform proclamation banned the private 
ownership of rural lands and declared that land would be distributed to 
the tillers without compensation to former owners. It also limited the 
size of land to be allotted to any single family to a maximum of 10 
hectares. Moreover, it prohibited the transfer of land by sale, exchange, 
succession, mortgage, lease or other means. Chapter 3 of the 
proclamation contains provisions for the establishment of peasant 
associations, the basic instrument for implementing the land reform. 
The Peasant Association is a territorial organization encompassing 800 
hectares or more. The average Peasant Association membership is 250- 
270 families (households). 
It was planned that, at the end of the MPP-I period, MPP-II would be 
undertaken overthe 1975/76-1979/80 period. However, because of the 
political instability and major structural changes in the rural areas, 
including the formation of peasant associations and producers' cooperatives 
as well as the implementation of the land reform, it was not 
possible to carry out this plan. There was not much organized and coordinated 
extension work in the country, therefore, until the beginning 
of the 1980s and it was only in 1981 that MPP-II was started. 
MPP-II had the same objectives as MPP-I. However, MPP-II was 
envisaged to cover 440 of the total 580 districts and reach as many 
farmers as possible. One major difference between the two was the 
channel employed in the transfer of technology. Under MPP-II the 
peasant associations and co-operatives were used as the focal points 
through which improved inputs, techniques of production and advice 
were channelled to the member farmers. 
As EPID was dissolved following the reorganization of the Ministry 
of Agriculture in 1979, the extension service, formerly under one umbrella, 
was split up and its activities were taken over by the line departments of 
the Ministry. The principal extension activities carried out by the different 
departments of the Ministry during MPP-II were in soil and water 
conservation, crop production and protection, livestock and fisheries 
and forestry. Given the fact that the extension services had been 
58 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
disintegrated, different extension agents representing the interests of 
the different departments could approach one and the same fanner, 
leading to the duplication of efforts and at times misuse of the limited 
available resources. 
MPP-II was assisted by the World Bank, the International Fund for 
Agricultural Development (WAD) and, to a small extent, by SIDA. During 
its implementation (1981-1985), the MPP-II did not attain its stated 
objectives because the very limited number of extension agents available 
in the country were made to cover as wide an area as possible without 
adequate facilities and logistical support. The same agents were 
overloaded with different assignments, such as collecting taxes, 
promoting producers' co-operatives, collecting loan repayments and 
mobilizing labour and resources on the part of public authorities, which 
were, at times, not in their domain of responsibility (Task Force on 
Agricultural Extension 1994a). 
The poor research-extension linkage was another factor responsible 
for the mediocre performance of the extension service of MPP-II. Most 
importantly, the country did not have the capacity and resources to 
develop innovations suitable to its socially and ecologically varied 
regions. Moreover, as compared to the MPP-I, in the course of 
implementing MPP-II the Ministry was compelled to work under a very 
limited budget. 
The MPP-II was phased out in 1985 and replaced by another strategy 
called the Peasant Agriculture Development Extension Programme 
(PADEP). PADEP was designed to bring perceptible changes in peasant 
agriculture through concerted and co-ordinated efforts in the areas of 
agricultural research and extension. The strategy was based on a critical 
evaluation of past extension strategies and underscored the importance 
of stratifying the country into relatively homogeneous zones, 
decentralizing the planning and execution of agricultural development 
activities and empowering and giving considerable attention to zones 
which were to be the centres of development efforts. Accordingly, on 
the basis of resemblances in climatic conditions, cropping patterns, 
natural resource endowments and geographical proximity, the country 
was divided into eight agricultural development zones. 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 59
The programme had different objectives for the different agricultural 
development zones. However, the principal ones were: increasing food 
production at least to the level of self-sufficiency; developing the 
production of cash crops for export and raw materials for domestic 
industries; increasing rural sector employment opportunities; supporting 
and encouraging the development of rural co-operatives; preventing 
further soil depletion and introducing suitable farming system in erosionprone 
areas of the country. 
It was initially planned to concentrate the programme on high 
potential areas so as.to raise their production and productivity by 
channelling the limited resources and extension services towards them. 
To this end, 148 surplus-producing districts were selected out of the 
total 580. PADEP employed a modified Training and Visit (T & V) 
extension system. In the selected districts an extension agent was 
assigned to serve 1300 peasant households through contact farmers 
organized into groups (the conventional T & V system recommends 
one extension agent for 800 farmers) and 2500 farmers in all other areas 
(non-surplus producing areas). Moreover, extension agents were trained 
monthly instead of fortnightly and zonal subject matter specialists were 
trained quarterly instead of monthly, as proposed by the conventional 
T & V system. In each district there was one extension co-ordinator for 
10 extension agents and the co-ordinator visited the agents once a 
week. Each extension agent worked with 48 contact farmers. The agent 
made regular visits of four days a week and on each day six contact 
farmers, who had each 26 follower farmers, were visited for a period of 
30 minutes each. Each contact farmer was therefore visited twice a month. 
As the poor research-extension linkage was considered to be an 
essential factor affecting the efficiency of extension work, Research 
Extension Liaison Committees were formed in 1986 both at the national 
and zonal levels. The committees were established to serve as a formal 
linking mechanism between research and extension and were mandated 
to review and approve research proposals submitted by research 
institutes. They were also to serve as a forum where the views of 
extension workers were taken into account in identifying research 
problems for the formulation of research topics. 
60 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
This was thought to help ensure that both researchers and development 
agents address the real problems that fanners face. However, the 
committees did not live long enough to be of practical use for two 
reasons. On one hand, some of the newly-created agricultural zones 
had no research stations and, on the other, the committees had no 
budget and were not backed up by the public authorities concerned. 
The activities of the committees were interrupted in 1991 because of the 
change in government, which resulted in the dissolution of the zonal 
agricultural offices and the transfer of their roles to the new Regional 
Bureaux of Agriculture. 
Like many of its predecessors, PADEP was designed as a foreignaided 
project (the principal donors for the PADEP were the European 
Economic Commission, JFAD, Italy, African Development Bank, Sweden 
and the World Bank). Consequently, its implementation had to be 
postponed pending the government's compliance with the conditions 
laid down by donor organizations. More specifically, donor countries 
and organizations had been pressing the government to abandon its 
agricultural policy, which was biased in favour of state and collective 
farms, to liberalize agricultural marketing and to give considerable 
emphasis to small-scale farmers. 
It was only in 1989 that PADEP could effectively be started. Even 
then, only six out of the eight PADEP programmes secured funding 
from both donors and government and were in operation. Even in areas 
where extension activities were undertaken it was not possible to bring 
togejner fanners and extension workers. Extension messages were not 
entirely devoid of political objectives and agents were seen by the 
farmers as government spokesmen rather than development workers. 
On all counts the extension approach was defective, not only because 
it was not participatory, but also because of its inflexible and top-down 
nature (Befekadu and Berhanu 1999-2000). The principal factor 
responsible for the inefficiency of extension work during 1975-1991 
was the government's agricultural policy, which favoured the 
development of state and collective farms. Although the 1975 radical 
land reform put an end to the tumultuous tenant-landlord relationships, 
the collectivization and villagization policies pursued by the Marxist 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 61
government and its commitment to increasing public ownership 
contributed greatly to the low performance of the agricultural sector in 
the 1980s. 
A number of empirical studies on the Marxist government's 
agricultural development strategy concluded that the state and collective 
farms, which accounted for less than 10 per cent of the total cultivated 
area, received the lion's share of subsidized agricultural inputs 
(agricultural credit, fertilizers, improved seeds and so on), extension 
services, farmers' training and the government's investment in 
agriculture, to the detriment of the private farms, which accounted for 
more than 90 per cent of the total agricultural production (Clapham 
1987, Cohen and Isaksson 1988, Mengisteab 1990, Pickett 1991). 
Paradoxically, state and collective farms have proved disappointing in 
terms of productivity, employment creation and environmental protection 
(Cohen and Isaksson 1988, Mengisteab 1990). In general, in the 1980s 
extension activities were obstructed by the government's selective 
agricultural policy and 'the non-surplus producing regions' had not 
received enough attention. 
The current situation 
• Following the change in government in 1991, the T & V extension 
approach was adopted as a national extension system with major 
government financing until its replacement by the Participatory 
* With the change in government in 1991, the country was divided into nine 
administrative regions, a federal capital (Addis Ababa) and one special 
administrative division (Dire Dawa). At present, extension activities are 
the entire responsibility of regional agricultural bureaux. The extension 
division of the federal Ministry of Agriculture has the task of coordinating 
inter-regional extension work, providing policy advice on 
nationwide agricultural extension issues, advising regional bureaux of 
agriculture in the areas of extension management and administration, 
developing extension training materials and organizing training programmes 
in agricultural extension for regional extension personnel. The 
regions are given full autonomy in the planning, execution, monitoring and 
evaluation of extension programmes 
62 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Demonstration and Training Extension System in 1995.4 The latter was 
adopted from the Sasakawa Global 2000 (SG 2000) extension strategy, 
initiated in Ethiopia in 1993 by the Sasakawa Africa Association and 
Global 2000 of the Carter Centre. As clearly indicated by Quinones and 
Takele (1996), the SG 2000 extension programme had the following 
objectives: 
• to assist Ethiopia's efforts to increase agricultural production 
through an aggressive technology transfer programme that disseminated 
improved production technologies to small-scale fanners 
through the extension service of the Ministry of Agriculture; 
• to strengthen the capacity of extension services for the expedient 
dissemination of proven, research-led technologies to small-scale 
producers, particularly in food crops; 
• to invigorate the linkages between research and extension in order to 
streamline the process of technology generation and dissemination 
and to provide appropriate feedback to research for technological 
interventions when necessary and 
• to extend, through extension services, improved grain storage and 
preservation technologies as well as agro-processing techniques 
suitable for small-scale producers. 
According to Takele (1997), the centrepiece of the SG 2000 technology 
transfer method is the extension management training plot (EMTP). 
EMTPs are on-farm technology demonstration plots established and 
managed by participating farmers who are selected by the local extension 
workers and SG 2000 personnel. The extension agents play a facilitating 
role in the management of the plots. The agents also use the EMTPs to 
train both participating and neighbouring farmers so that they can put 
into practice the entire package of recommended practices. The size of 
each EMTP is usually half a hectare and adjacent fanners can pool their 
plots to form an EMTP if they cannot meet the half-hectare requirement 
individually. 
The SG 2000 extension activities started by assessing available 
agricultural technologies in the country with the support of the national 
research and extension bodies. On the basis of the availability of 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 63
improved varieties and recommendations of the research and extension 
experts, in 1993 technology packages for maize and wheat production 
were defined and demonstrated to 160 farmers residing in seven districts 
of the Oromia National Regional State and the Southern Nations, 
Nationalities and Peoples Regional State. In 1994 the SG 2000 extension 
programme expanded its extension activities both in terms of area 
coverage and technology packages. More specifically, sorghum and 
teff technology packages were included in the programme, the number 
of participating fanners rose to 1600 and the programme was expanded 
to some districts of the Amhara National Regional State and the Tigray 
National Regional State. 
In 1995 good weather conditions, coupled with the material and 
technical support that participating farmers received from SG 2000, 
resulted in substantial yield increments. The impressive yield increments 
obtained by- the participating farmers persuaded the Ethiopian 
government that self-sufficiency in food production could be achieved 
by adopting the SG 2000 extension approach. Consequently, in 1995 the 
government took the initiative to run the programme on its own and 
launched the participatory demonstration and training extension system 
(PADETES) as the national agricultural extension system. 
PADETES was developed after a critical evaluation of the past 
extension approaches and the experience of SG 2000. Its major objectives 
include increasing production and productivity of small-scale fanners 
through research-generated information and technologies; empowering 
farmers to participate actively in the development process; increasing 
the level of food self-sufficiency; increasing the supply of industrial 
and export crops and ensuring the rehabilitation and conservation of 
the natural resource base of the country (Task Force on Agricultural 
Extension, 1994b).5 
The system gives special consideration to the package approach to 
agricultural development Initially, PADETES promoted cereal production 
"• According to government officials, an important element of the 
PADETES approach is the promotion of the active participation of rural 
communities in problem identification, analysis, planning, implementation 
and evaluation. 
64 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
packages and the beneficiaries were mainly those farmers who live in 
high rainfall areas of the country. Over the years, however, the packages 
have been diversified to address the needs of farmers who live in different 
agro-ecological zones of the country. Currently, PADETES promotes 
packages on cereals, livestock (dairy, fattening and poultry), high 
economic value crops (oil crops, pulses, vegetables and spices), 
improved post-harvest technologies (handling, transport and storage), 
agro-forestry, soil and water conservation and beekeeping developed 
for different agro-ecological zones (highland mixed farming system, 
highland-degraded and low moisture, lowland agro-pastoralist and 
lowland pastoralist zones). 
The major elements of the extension package are fertilizer, improved 
seeds, pesticides and better cultural practices mainly for cereal crops 
(teff, wheat, maize, barley, sorghum and millet). PADETES uses EMTPs 
and a technology transfer model which, in principle, nurtures linkages 
between research, extension, input and credit distribution. Under 
PADETES the major tasks of extension agents include organizing 
demonstration trials, assisting farmers in obtaining agricultural inputs 
and channelling farmers' problems to the relevant organizations, 
particularly to the district agricultural office. 
The PADETES approach is meant to improve access to inputs by 
providing credit in kind. As farmers cannot borrow from banks due to 
collateral problems, extension credit is guaranteed by the regional 
governments and administered jointly by them and the two government 
banks (the Development Bank of Ethiopia and the Commercial Bank of 
Ethiopia). Loans are taken up by the regional governments and 
channelled into the district administration offices.6 Fanners participating 
6 In the new extension system, input supply and credit are dealt with in 
one transaction. The procedures involved in input loan disbursement are 
as follows. Tlie regional government borrows directly from the banks and 
relies on its administrative machinery and peasant organizations to 
disburse and collect the loan. Farmers have to apply via the service cooperatives, 
which submit applications for credit to the district agricultural 
office. The district finance office is also involved. The service cooperative 
collects a 25 per cent down payment of the (contd. over). 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 65
in PADETES then receive credit in kind, via the district agricultural and 
finance offices. Participants agree to allocate land for a demonstration 
plot and pay a 25 per cent down payment on the input package at the 
time of planting, with the balance due after harvest. The participants 
pay a 10.5 per cent interest rate on the input loan. In 1995—1996 the 
Ethiopian government sponsored the establishment of about 36,000 
half-hectare on-farm demonstrations. In the 1996-1997,1997-1998 and 
1998-1999 production years, the number of government-sponsored 
demonstration plots was 600,000,2.9 million and 3.8 million, respectively 
(MOA1997,1998b, 1999). The trend is for this number to keep growing. 
Likewise, the number of farmers participating in the new extension 
programme increased from 35,000 in 1995-1996 to 3.7 million in 1998- 
19997 
As to the number of extension personnel in the country, the author's 
discussion with a senior extension expert in the Ministry of Agriculture 
in September 2001 revealed that this is estimated at little more than 
14,000. Most hold certificates and diplomas but lack adequate and 
appropriate technical and communication skills. This figure is too small, 
even by the standards of sub Saharan Africa, when viewed in relation 
to the number of fanners the extension personnel have to serve. 
6 (contd.) input prices. An agreement is signed between the finance office 
and the co-operative. The down payment and signing result in a delivery 
order by the finance office, which the co-operatives use to collect their 
stock from the designated supplier 
7 Quinones and Takele (1996) and MOA (1997, 1998, 1999) reported that 
the average yields of different crops obtained from EMTPs are much 
higher than the traditional averages. For instance, according to MOA 
(1999), in the 1998-1999 production year, the average maize, sorghum, 
teff, wheat and barley yields of EMTPs fanners in the Oromia National 
Regional State were 247%, 67%, 100%, 225% and 129% above the 
traditional averages respectively. The corresponding figures for the 
Southern Nations, Nationalities and Peoples Regional State were 205%, 
71%, 64%, 207% and 141% respectively 
66 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
11 Methodology 
The data used in this paper were obtained from the fifth Ethiopian Rural 
Household Survey conducted by the Economics Department of Addis 
Ababa University in collaboration with US AID. Four successive surveys 
had been conducted prior to this one and all were aimed at generating 
time-series data on the Ethiopian rural households. The fifth survey 
was carried out in 1999 and the collected data refer to the 1998-1999 
cropping seasons. 
This paper is based on data collected from a total of 1482 household 
heads randomly selected from four regional states of the country. More 
specifically, data were collected from 16 sites located in the Amhara 
National Regional State (ANRS), the Oromia National Regional State 
(ONRS), the Southern Nations, Nationalities and Peoples Regional State 
(SNNPRS) and the Tigray National Regional State (TNRS).8 According 
to MEDaC (2000), these four regions account for about 86 per cent of 
the country's population (61.7 million). More specifically, in 1999, the 
total population of the ANRS, ONRS, SNNPRS and TNRS was 16 million, 
21.7 million, 12.1 million and3.6 million, respectively. The same source 
shows that the ANRS, ONRS, SNNPRS and TNRS account for 12.7%, 
28.2%, 9% and 4% of the total land area of the country. 
Seven of the 16 study sites, namely Adele Tiqe, Bako Tibe, Debre 
Zeit-Sherba Goditi, Eteya, Jimma—Somodo, Koro Degaga-Sodere and 
Shashemene are located in the Oromia National Regional State. Four of 
the sites, Debre Berhan, Dinki-Ankober, Shumsha-Lalibela and Yetmen, 
are located in the Amhara National Regional State. Three sites, namely 
Domma, Durame—Azedebos and Gara Godo-Areka, are located in the 
8 This study covered only four of the nine regional states of the country. 
These regions constitute the densely settled, mixed-farming regions of the 
temperate and central highlands of the country. About 90 per cent of the 
rural population of the country is found in these regions (CSA, 1999). 
More specifically, about 27%, 36%, 21 % and 6% of the rural population 
of the country live in the ANRS, ONRS, SNNPRS and TNRS 
respectively. These four regions account also for almost all the nation's 
agricultural production 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 67
Table I: regional distribution of the respondents 
Region Number of Percentage 
respondents 
31.4 
42.2 
16.3 
10.1 
100 1482 
Amhara 466 
Oromia 625 
Southern Nations, Nationalities and Peoples 241 
Tigray 150 
Total 
Southern Nations, Nationalities and Peoples Regional State. The 
remaining two, Haressaw and Geblen, are located in the Tigray National 
Regional State. The regional distribution of the respondents is presented 
in Table I. 
The survey employed a structured questionnaire. The questionnaire 
was prepared to collect information on various aspects of agriculture 
such as landholding, crop and livestock production, informal and formal 
credit services, demography, social phenomena, natural occurrences, 
environmental degradation and conservation attempts, off-farm 
activities, extension services and the supply use of commercial inputs. 
niResiiHs and discussion 
The mean age of the respondents at the time of the survey was 49 years 
but there were extreme variations, the oldest being 95 and the youngest 
18 years old. About 75% and 24% of the household heads were men 
and women, respectively. One hundred and twelve respondents (7.6% 
of the total respondents) reported that they had served as contact 
fanners during the military regime. A closer look at the regional 
distribution of the respondents who had served as contact farmers 
during the military regime shows that 21,66,13 and 12 were from the 
ANRS, ONRS, SNNPRS and TNRS, respectively. 
One surprising outcome of this study is that the majority of the 
former contact fanners had not participated either in PADETES or the 
SG 2(XX) extensionpro^amme at the time of the survey. More specifically, 
only 51 (45.5%^) and 8 (7.1 %\) of the former contact fanners reported that 
they had participated in PADETES and SG 2000 extension programme at 
68 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 5003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Table II: Number of sample respondents who served as 
region. 1994/95-1998/99 c r o p p i n g seasons 
contact farmers In the SG 2000 extension programme by 
TOTAL SNNPRS TNRS ONRS ANRS 
• 
Cropping 
Season 
14 17 1994/95 
14 19 1995/96 
20 1996/97 28 
27 16 1997/98 
2
2
3
4
4 22 
1
2
4
5
5 
0
1
1
2
4 1998/99 35 
the time of the survey, respectively.9 It is quite obvious that, under the 
military regime, contact farmers had been working very closely with 
extension agents and shared their experience and skills with a number 
of follower farmers. Therefore, the fact that the majority of the former 
contact farmers had not participated either in PADETES or SG 2000 
extension programme at the time of the survey, must be considered as a 
very important point of concern in that it reflects lack of continuity in 
extension work. 
The survey results reveal that the number of sample farmers who 
served as contact farmers in the SG 2000 extension programme was very 
limited (Table IT). 
Table II shows clearly that only a very small proportion of the sample 
respondents had been participating in the SG 2000 extension programme 
over the 1994—1999 period. However, the number of sample respondents 
who participated in the new extension programme had been growing 
over the same period (Table HI). 
9 The regional distribution of the 51 former contact farmers who reported 
that they had been participating in PADETES shows that 7, 31, 11 and 2 
were from the ANRS, ONRS, SNNPRS and TNRS respectively. Six of 
the eight former contact farmers who had taken part in the SG 2000 
extension programme were from the ONRS, one was from the ANRS and 
the remaining farmer was from the TNRS 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 69
Table III: Number of sample respondents who participated in 
PADETES by region, 1994/95-1998/99 cropping seasons* 
TNRS Cropping SNNPRS ONRS ANRS TOTAL 
Season 
9 (3.7) 88(14.1) 1994/95 3 (0.6) 8 (5.3) 108 (7.3) 
12(5.0) 1995/96 9 (6.0) 133 (9.0) 7(1.5) 105(16.8) 
36(14.9) 1996/97 13(8.7) 179(12.1) 5(1.1) 125(20.0) 
86(35.7) 10(2.1) 132(21.1) 14(9.3) 242(16.3) 1997/98 
94(39.0) 12(8.0) 259(17.5) 20 (4.3) 133(21.3) 1998/99 
* Figures in parentheses refer to percentages 
Table HI shows that the proportion of sample farmers who took part in 
the new extension programme had been growing steadily in the SNNPRS. 
The proportion had been more or less stable in the ONRS and TNRS 
between 1996/97 and 1998/99. In the ANRS only a very small fraction of 
the sample farmers (less than 5 per cent) had participated in the 
programme over the 1994-1999 period. 
Tables II and IE show that the vast majority of the sample respondents 
in all regions had not participated either in PADETES or the SG 2000 
extension programme over the period under study. The sample 
respondents who had not participated either in PADETES or the SG 
2000 extension programme at the time of the survey were asked to point 
out the reasons for their non-participation (Table IV). 
Table IV shows that lack or shortage of working capital for down 
payments, shortage of land and the limited geographical coverage of 
the programme are the principal factors responsible for the nonparticipation 
of the sample respondents in either PADETES or the SG 
2000 extension programme. Table IV shows also that about 41 per cent 
of the respondents cited the shortage or lack of working capital as an 
important barrier to the adoption of modern agricultural inputs. This is 
partly explained by the substantial increase in the prices of these inputs. 
Given the fact that most of these inputs (especially fertilizer and agrochemicals) 
are imported and the national currency has been losing its 
70 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Table IV: Reasons for non-participation In PADETES or SG2000 
Extension Programme, by region 
Reasons 
SNNPRS A N R S ONRS 
30.0 19.6 
29.8 42.9 
2.8 2.1 
Shortage of land 
Shortage/lack of working 
capital for down payments 
Did not repay previous loans 
Not interested in the 
2.4 3.0 programme 
Insufficient knowledge of the programme's 
3.3 4.4 potential impact 
1.5 1.2 Shortage/lack of labour 
Participants are selected by 
2.6 Peasant Association Officials 1.6 
4.6 0.7 Not invited to participate 
The programme is not more profitable 
1.1 2.5 than traditional methods 
High interest payable on input 
0.9 loans 0.0 
0.2 0.2 
0.7 0.2 
Shortage of draft power 
High price of inputs 
The programme regulations 
0.2 0.5 are too strict 
0.2 6.2 Sharecropped/rented out land 
0.0 ] Do not want to use fertilizers 0.5 
0.7 0.5 
1.7 0.2 
Request turned down 
Fear of taking loans 
No information on the 
programme 16.7 10.6 
0.7 3.2 Other responses (not specified) 
100 Total 100 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 71 
Percentage of non-participant 
sample respondents 
TOTAL TNRS 
25.6 25.2 30.2 
40.8 51.9 59.7 
2.2 2.2 0.7 
2.3 2.2 0.0 
3.3 3.0 0.0 
1.6 0.0 5.0 
2.3 5.9 0.0 
2.2 1.5 0.0 
1.4 0.0 0.0 
0.7 0.7 2.2 
0.2 0.0 0.0 
0.4 0.0 0.7 
0.3 0.0 0.0 
5.2 3.1 0.7 
0.2 0.0 0.0 
0.4 0.0 0.0 
0.8 0.0 0.0 
10.6 0.0 0.7 
1.7 2.2 0.0 
100 100 100
value over the past ten years, their prices have been increasing every 
year. In addition, up to 1997 fertilizer prices had been subsidized and 
farmers paid relatively lower prices even after the national currency was 
devalued in 1992. When the government subsidy was lifted in January 
1997, the free market prices became so exorbitant that they put fertilizer 
beyond the reach of many fanners. There was even a widespread fear 
that fertilizer consumption would plummet. However, fertilizer 
consumption has been increasing nationally as a result of the new 
extension policy which promotes fertilizer credit as one of the essential 
elements of the technology package.10 
Input loans are an integral part of the new extension programme on 
the grounds that they play a pivotal role in facilitating increased access 
by poor fanners to modern agricultural inputs. However, the input loan 
scheme is associated with two ineluctable problems: on one hand, as it 
is not based on a realistic assessment of the borrowers' financial needs, 
it is difficult to ascertain whether or not the credit has reached the very 
people who need it and, on the other hand, it is awkward to administer 
and the risks of non-repayment loom large. 
In an effort to minimize the default rate and the loss of their yearly 
budgets which they use as collateral for the bank loans, regional 
governments have frequently called on local police to enforce repayment 
This has resulted in some confiscation or the forced liquidation of farm 
assets (such as animals and equipment) to meet repayments (Howard et 
al. 1999). As the PADETES system is intimately tied up with the delivery 
of credit in kind, the coercive loan collection system seems to dissuade 
many farmers from participating in PADETES.n Similarly, the fact that 
input loans have to be repaid right after harvest, when the prices of 
agricultural produces are extremely low, may discourage many fanners 
from participating in PADETES. 
10 According to the National Fertilizer Industry Agency of Ethiopia, the 
total fertilizer consumption in 1994,1997 and 1999 was 190,000, 
220,431 and 285,823 tons respectively 
'' Dercon (2000) argues that the nature of credit arrangements reduces the 
attractiveness of input uptake. To be eligible, a farmer must have repaid 
all previous loans. A service co-operative may not receive new credit if 
7Z JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
1 
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Table IV shows also that about 26 per cent of the respondents identified 
shortage of land as being the principal reason for their non-participation 
either in PADETES or in SG 2000 extension programme. The survey 
results show that the average landholding of the sample respondents 
at the time of the survey was 1.34 hectares and about 29 per cent of the 
respondents had less than 0.75 hectares of land. As land is state-owned 
and farmers are not allowed to buy and sell it, an increase in the rural 
population results in the further fragmentation of holdings. 
Another interesting outcome of this study is the respondents' view 
on the non-participatory nature of PADETES and the SG 2000 extension 
programme. As indicated in Table IV, 2.3 per cent of those respondents 
who had not participated either in PADETES or the SG 2000 extension 
programme reported that the participants were selected by officials. 
Similarly, 2.2 per cent indicated that they had not been invited 
(approached) to take part in these programmes. These results suggest 
the new extension system is not as participatory as its name implies. 
The results are also in complete agreement with MOA (1998b), which 
reported that in all regions 75-S0 per cent of the fanners who participated 
in PADETES were selected by the local extension agents. The rest were 
either selected by the officials of the Peasant Association or participated 
in the programme on their own initiative. These results show that 
extension agents tend to work with those 'model' farmers who show an 
interest in the extension packages.12 Similarly, in his study on the 
constraints to agricultural extension work in Ethiopia, Belay (2002) points 
' (contd.) one member defaults on the loan. This measure is applied even 
though no local community participatory measures exist to screen 
borrowers. Input credits are expected to be paid regardless of the harvest 
and there are no provisions to help even those requesting the postponement 
of repayment to the next season 
12 One of the criteria used to evaluate the performance of extension agents is 
the number of farmers adopting the technology packages in their area of 
work. In other words, quotas (the minimum number of farmers who 
should take up the technology packages) are imposed on extension agents. 
As a result, the extension agents use whatever means available to 
persuade farmers to adopt the packages and thereby meet their quotas 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 73
Table V: Respondents who used fertilizer and improved seed 
in the 1998-1999 main cropping season, by region 
Percentage of respondents 
ANRS ONRS SNNPRS TNRS TOTAL 
Fertilizer 45^3 75^4 73^4 19^3 59.9 
Improved seed 5.4 21.9 45.6 8.7 19.2 
out that fanners make a very marginal contribution in designing and 
formulating extension activities. He also notes that neither the fanners 
nor the frontline extension agents are consulted in the course of policy 
formulation. 
When requested to indicate whether or not they had been using 
modern agricultural inputs, about 60 per cent reported that they used 
fertilizer in the 1998-1999 main cropping season. The proportion of 
sample respondents who used fertilizer was the highest in the ONRS 
and the lowest in the TNRS. Similarly, 19 per cent of the respondents 
reported that they used improved seed in the same period. At regional 
level, the percentage of respondents who used improved seed was the 
highest in the SNNPRS and the lowest in the ANRS (Table V). 
One important factor in the adoption of improved inputs is the 
availability of useful and reliable extension services. In this connection, 
the survey results show that only 13.3 per cent of the sample respondents 
had been visited by extension agents in the 1998-1999 main cropping 
season. This could be explained by the fact that the number of extension 
agents in the country is extremely small in relation to the number of 
people whom these agents have to serve. In fact, the proportion of 
sample farmers visited by extension agents varied from one region to 
another. More specifically, it was 4.3%, 15.5%, 29.5% and 6%, in the 
ANRS, ONRS, SNNPRS and TNRS, respectively. It is also interesting to 
note that 56.3%, 33.5% and 7.7% of the sample respondents who 
participated either in PADETES or SG 2000 extension programme qualified 
the assistance and advice they had received from extension agents, 
while adopting improved inputs, as very useful, useful and not useful, 
respectively, while 2.6 per cent did not comment on the nature of services 
rendered by extension agents. 
74 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Similarly, participating farmers reported that they were visited, on the 
average, 4.4. times by extension agents in the 1998-19999 main cropping 
season. This average masks differences which exist among regions. 
Participating sample farmers reported that they were visited, on the 
average, 3.3 times in the ANRS, 5.6 times in the ONRS, 2.6 times in the 
SNNPRS and 3.2 times in the TNRS. 
As already noted, the shortage of extension professionals is one of 
the most serious problems of agricultural extension work in Ethiopia, 
contributing to the limited extension service coverage and poor adoption 
of agricultural technologies. Although there has been an increase in 
the number of extension agents throughout he country over the last 
five years, each extension agent is still expected to reach a large number 
of farmers. The extension agent to fanner ratio varies from one region to 
another. It was reported to be 1:866 in the ANRS, 1:880 in the ONRS, 
1:570 in the SNNPRS and 1:1500 in the TNRS (MOA1998b, 1999). 
All the sample respondents were asked to indicate whether or not 
they planned to participate (or continue participating) in the new 
extension programme in the event that this happens to be the sole 
means of getting extension services in the future. In this respect, 966 
respondents (67.5 per cent) reported that they would participate (or 
continue participating) in the programme, 448 respondents (30.2 per 
cent) stated that they had no intention of participating in the programme 
and the remaining 38 (2.6 per cent) did not reveal their intentions. Of 
course, the proportion of sample farmers who reported that they would 
participate (or continue participating) in the new extension programme 
varied from one region to another. It constituted 54.7%, 81.9%, 54.8% 
and 64.7% of the sample respondents in the ANRS, ORNS, SNNPRS 
and TNRS, respectively. The sample respondents who reported that 
they intended to participate (or continue participating) in the new 
extension programme were also asked to list down the major reasons 
influencing their decision to do so in the future (Table VI). 
Table VI makes it clear that the principal motives for the decision to 
take part in the new extension programme in the future include the need 
to get fertilizer at the right time, the possibility of procuring improved 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 75
Table VI: Reasons for the willingness to participate (or 
continue participating) in the new extension programme in 
the future 
Reasons Percentage of respondents* 
TOTAL TNRS ONRSSNNPRS 
79.4 72.0 80.0 69.0 
45.5 
70.9 
65.3 
67.4 
23.7 
85.5 
45.7 
71.6 
ANRS 
To get fertilizer at the right time 71.3 
To purchase fertilizer at a 
lower price 49.0 
To get improved seed 63.3 
To learn about new techniques 
of production 43.4 
Others (not specified) 1.6 
52.3 
0.9 
59.2 
0.8 
55.0 
0.8 
34.7
- 
The percentages do not add up to 100 per cent because ot multiple responsesi 
seed, the desire to learn about new techniques of production and the 
possibility of purchasing fertilizer at lower price, in that order. 
Those respondents who reported that they had no intention of 
participating in a new extension programme were also asked to list down 
the major reasons which influenced their decision. Their responses are 
summari2ed in Table VII. 
Table VII reveals that the vast majority of the respondents who 
reported that they had no intention of participating in the new extension 
programme in the future cited the lack or shortage of working capital for 
down payments, shortage of land, the feeling that the packages are not 
profitable and the lack of suitable land as the principal reasons which 
would dissuade them from participating in a new extension programme. 
One interesting outcome of this study is the respondents' view on 
the profitability of improved inputs. About 30 per cent the respondents 
who reported that they had no intention of participating in the new 
extension programme said that modern inputs were not more profitable 
than the traditional ones.13 This could be partly explained by the fact 
11 Profitability analyses of maize and wheat production made by Quinones 
and Takcie (1996) and Chimdessa (1998) reveal that the EMTPs have 
been more profitable than the traditional plots 
76 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Bertay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Table VII: Reasons for the non-willingness to participate in 
the new extension programme in the future 
Reasons Percentage of respondents* 
ANRS ONRSSNNPRS TNRS TOTAL 
Shortage/lack of working 
capital for down payments 65.2 64.7 81.6 78.6 70.8 
The feeling that the packages 
are not profitable 32.3 41.2 20.4 26.2 30.5 
Lack of suitable land lathe 
package 19.5 35.3 41.7 33.3 30.2 
Shortage of land 
for the package 27.4 24.7 39.8 52.4 32.7 
Ottersfrtot specified) 4.3 1.2 2.9 - 2.8 
(* THE PERCENTAGES DO NOT ADD UP TO 100 PER CENT BECAUSE OF MULTIPLE RESPONSES) 
that in many parts of the country extension agents promote technologies 
as "blanket recommendations". In other words, these technologies are 
developed outside the users' system and are extended to farmers without 
prior adaptability trials. Available evidence shows that such a practice 
imperils the whole effort of extension work in the country. According to 
Sandford(1997): 
Ethiopia is a country of enormous ecological diversity ... such 
diversity greatly complicates the task of developing and testing 
improved agricultural technology. The truth is that the new 
technologies currently being advocated by the agricultural 
extension services were developed and tested in one set of 
environments (usually the most favourable) and are now being 
applied in others (often very marginal ones), for which they were 
not designed or even tested and where they may actual ly produce 
less than the old technologies they are intended to replace. 
The current reorganization of the national agricultural research system 
is expected to do away with these problems. But, given the country's 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 77
ecological diversity, it will take many years before technologies suitable 
to the different locations of the country are developed. 
As already noted, extension agents play a leading role in the 
popularization of improved inputs and cultural practices in rural areas. 
In the new extension programme, agents are under pressure to work 
with as many fanners as possible. In many parts of the country one 
agent may be required to supervise the demonstration plots of more 
than 300 fanners. Moreover, extension agents are quite often overloaded 
with different assignments, which are, in most cases, not related to their 
normal duties. Table VIII presents the respondents' view about the 
activities in which the extension agents are engaged. 
One can deduce from Table VHI that extension agents are not free to 
concentrate entirely on promoting improved technologies, identifying 
farmers' production problems and eventually bringing them to the 
attention of researchers. Over the years, the involvement of extension 
agents in non-extension activities has played against their reputation 
as development workers. Many people in rural areas continue looking 
at the extension agents as government spokesmen rather than facilitators 
in the rural development endeavour. If agents are to spearhead rural 
transformation in the country, their duties should be clearly defined 
ANRS 
Introducing modern inputs 
Popularizing new cultural 
practices 
Organizing farmers for credit 
access 
Mobilizing farmers for pubfc 
work 
Collecting loans repayments 
69.8 
Table VIII: Principal activities of extension agents as 
perceived by sample respondents, by region. 
Percentage of respondents* 
ONRSSNNPRS TNRS TOTAL 
71.5 
53.3 
22.1 
10.9 
13.8 
7.8 
Activity 
Collecting taxes 
(* The percentages do not add up to 100 percent because of muKpferesponSOS) 
78 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003 
74.2 72.6 66.9 
55.4 50.0 62.8 40.0 
23.6 21.1 30.0 20.2 
8.2 24.6 7.3 13.3 
16.7 3.1 17.3 10.5 
3.9 14.2 3.1 5.8
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
and they should not be made to handle other responsibilities that will 
compromise their real professional integrity. 
In this respect, Howard et al. (1999) noted that many of the extension 
agents interviewed expressed concern about their heavy involvement 
in credit administration—not only because it kept them from their 
technical responsibilities (advising farmers on use of improved 
technologies), but also because their role as credit collection agents 
had a negative impact on their personal relationships with farmers. 
It is, however, worth noting that some of the activities, such as the 
distribution of inputs and credit, could well be handled by the agents, 
provided that complicated administrative red tape is not involved and 
they do not take up too much of the agents' time. Even then, this needs 
to be done with maximum care as it may affect the extension agents' 
relationship with farmers. 
IV. Conclusion 
This paper focused on assessing the participation of farmers in the new 
extension system and identifying the principal barriers to the adoption 
of modern agricultural inputs. It is believed that this study, although 
limited in coverage and scope, provides information to all concerned in 
agricultural development so that they can make informed decisions. 
The historical review reveals that extension work in the country has 
not been participatory in its nature: Extension programmes and policies 
have been formulated without due consideration to the farmers' opinions 
and their traditional knowledge system. Moreover, both the historical 
review and the survey results confirm that only a very small percentage 
of the country's rural population has access to agricultural extension 
services. One interesting outcome of this study is that the number of 
sample fanners who used fertilizer and improved seed had been growing 
steadily over me 1994-1999period. 
This study has also pointed out that the shortage or lack of working 
capital, the shortage of arable land, the shortage of extension personnel, 
the inadequacy of relevant technologies (as far as their adaptation to 
the different ecological, economic and sociocultural conditions of the 
country) and the involvement of extension agents in non-extension 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 79
activities were the most important barriers to the adoption of modern 
agricultural inputs in the country. 
In the light of these results it is imperative that policymakers pay 
utmost attention to the constraints that beset peasant agriculture. The 
issues which need immediate attention include, among others, 
undertaking adaptability trials before popularizing extension packages, 
giving special consideration to the farmers' indigenous and experiencebased 
knowledge system in research and extension work, changing the 
way the input credit arrangements are organized and enforced and 
encouraging resource-poor farmers to participate in the programme. 
Attempts must also be made to make agricultural research and 
extension activities complementary and participatory in nature. The 
fact that agricultural research and extension are carried out by two 
different bodies with very limited contact and working relationships, 
has made it difficult to develop and disseminate technologies adapted 
to the farmers' conditions. It is only when extension and research 
activities are complementary that fruitful results can be achieved. 
Consequently, concerted efforts need to be made to create a seamless 
web between the two. As it is impossible to expect success from rural 
development interventions formulated without the active participation 
of the target population, farmers should be empowered and encouraged 
to participate in the planning and execution of extension programmes. 
Similarly, the contact between extension agents and farmers must be 
strengthened by reducing the ratio of fanners to development agents 
and relieving the extension agents of activities which compromise their 
role as extension educators. 
Acknowledgements 
The author is grateful to the officials of the Economics Department 
of Addis Ababa University and USAID Ethiopia for their permission 
to use the data from the fifth round Ethiopian Rural Household 
Survey. The author would also like to thank Dr Mulat Demeke, 
Dean of the Faculty of Business and Economics at Addis Ababa 
University and Dr Birendra Kumar, Professor of Agricultural 
80 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2003
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Extension at Alemaya University for their invaluable comments and 
suggestions on an earlier version of this paper. 
References 
Befekadu, D. andBeriianu, N. (eds.) 1999-2000. Annual Report on the 
Ethiopian Economy, Vol I, Addis Ababa, The Ethiopian Economic 
Association 
Belay, K. 2002. Constraints to Agricultural Extension Work in Ethiopia: 
The Insiders' View, Alemaya University, Department of Agricultural 
Economics 
— 1998. Inappropriate policies and wrong priorities: invisible causes 
for the mediocre performance of Ethiopian agriculture. Agricultura 
Tropica et Subtropical! 201-210 
Betru, G. 1975. Integrated Development in Rural Ethiopia: An 
Evaluative Study of the Chillalo Agricultural Development Unit, 
Bloomington, International Development Research Center, Indiana 
University, Indiana 
Chimdessa, B. 1998. The Package Extension Approach and 
Smallholders' Farm Productivity in High Potential Areas of Ethiopia: 
The Case ofShashemene Area, Unpublished M.Sc. thesis, Alemaya 
University, Ethiopia 
Clapham, C. 1987. Revolutionary socialist development in Ethiopia. 
AfricanAffairs, 8634351-165 
Cohen, J and Isaksson. N. 1988. Food production strategy debates in 
revolutionary Ethiopia. World Development, 16 3 323-348 
CSA (Central Statistical Authority) 1999. Statistical Abstract: 1998, 
Addis Ababa, CSA. 
Daniel, M., Hagos, G., Jayne, T.S. and Shaffer, J. 1997. Designing 
Strategies to Support a Transformation of Agriculture in Ethiopia, 
Grain Marketing Research Project, Working Paper No. 4, Addis Ababa, 
MEDaC 
Dercon, S. 2000. Growth and Poverty in Ethiopia in the 1990s: An 
Economic Perspective, Centre for die Study of African Economies, 
Oxford University 
VOL 18 NO 1 JANUARY 20O3 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 81
EPBD 1970. A Master Plan for Extension and Project Implementation 
Department, Ministry of Agriculture Publication No. 1, Addis Ababa 
Harbeson, J.W. 1990. State and social transformation in modern Ethiopia. 
In: Qttaway, M. (ed.), The Political Economy of Ethiopia, New York, 
Praeger Publishers, 73-91 
Haile Selassie, B. 1959. An Analysis of Administrative Organization 
and Extension Services with Suggestions for Application in Ethiopia, 
Unpublished M.Sc. thesis, Cornell University 
Howard, J.A., Kelly, V., Stepanek, J., Crawford, E.W., Mulat, D. and 
Maredia, M. 1999. Green Revolution Technology Takes Root in Africa: 
The Promise and Challenges of the Ministry of Agriculture/ SG 2000 
Experiment with Improved Cereals Technology in Ethiopia, Michigan 
State University International Development Working Paper No 76, East 
Lansing, Michigan State University 
Huffhagel, H.P. 1961 .Agriculture in Ethiopia, Rome, FAO 
MEDaC (Ministry of Economic Development and Co-operation) 1999. 
Survey of the Ethiopian Economy: Review of Post-Reform Developments, 
1992/93-1997/98, Addis Ababa, MEDaC 
— 2000. The Ethiopian Population Profile: 1999, Addis Ababa, 
MEDaC 
Mengisteab, K. 1990. Ethiopia: Failure of Land Reform and 
Agricultural Crisis, New York, Greenwood Press 
MOA (Ministry of Agriculture) 1997. Implementation of the National 
Extension Intervention Programme in the 1996/97production year, 
(in Amharic), Addis Ababa 
— 1998a.Information on Ethiopian Agriculture (in Amharic), 45 Addis 
Ababa, MOA 
— 1998b. Implementation of the National Extension Intervention 
Programme in the 1997/98 production year, (in Amharic), Addis 
Ababa, MOA 
— 1999. Implementation of the National Extension Intervention 
Programme in the 1998/99 production year, (in Amharic), Addis 
Ababa, MOA 
Pickett, J. 1991. Economic Development in Ethiopia: Agriculture, the 
Market and the State, Paris, OECD 
82 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA VOL 18 NO 1 JANUARY 2O03
Belay/AGRICULTURAL EXTENSION IN ETHIOPIA 
Quinones, M. and Takele, G. 1996. An overview of the Sasakawa-Global 
2000 project in Ethiopia. In: Breth, S.A. (ed.), Achieving Greater Impact 
from Research Investment in Africa, Mexico City, Sasakawa Africa 
Association; 16-27 
Sandford, S. 1997. The most important job in the world. Farm Africa 
Food and Agricultural Research Management News, Quarterly, 
Christmas, 4-7 
Schulz, M. 1981. Extension services in Ethiopia: Strategies before and 
after revolution. In: Crouch, B.R. and Chalama, S. (eds.), Extension 
Education and Rural Development, Vol. 2, New York, John Wiley and 
Sons, 189-212 
Stommes, E. and Sisaye, S. 1979. The administration of agricultural 
development programmes: a look at the Ethiopian approach: Part I. 
Agricultural Administration, 6 3 221-239 
Takele, G. 1997. Sasakawa Global 2000 in the Intensification of 
Agriculture in Ethiopia, Paper presented in the 13th International 
Conference of Ethiopian Studies, Koyoto, 12-17 December 
Task Force on Agricultural Extension 1994a. A Study on the Ethiopian 
Agricultural Extension System (in Amharic), Addis Ababa 
Task Force on Agricultural Extension 1994b. Ethiopian Agricultural 
Extension System (in Amharic), Addis Ababa 
Tesfai, T. 1975. The Evolution of Alternative Rural Development 
Strategies in Ethiopia: Implications for Employment and Income 
Distribution, African Rural Employment Paper No. 12, Michigan State 
University, Department of Economics 
VOL 18 NO 1 JANUARY 2003 JOURNAL OF SOCIAL DEVELOPMENT IN AFRICA 83