Promoting farm/non-farm linkages for rural development 121

Chapter 5
Rural farm/non-farm income linkages in
northern Ethiopia
Tassew Woldehanna

INTRODUCTION
Rural development policies often neglect the role of rural non-farm activities
and their link with agriculture. This might be because the role of the rural non-
farm sector is the least understood component of the rural economy; its role in
the broader development process is not well known (Lanjouw and Lanjouw,
1997). This knowledge gap has been reflected in the policies of developing
countries such as Ethiopia where there is no development policy that identifies
and includes the rural non-farm sector. Agricultural ministries have instead
focused on farming, and industry ministries have focused on manufacturing.
There is a social cost to the failure to recognize the importance of the rural non-
farm sector in decreasing rural-urban migration and its potential role in absorbing
the growing rural labour force, thus contributing to the national economy and
promoting a more equitable distribution of that income.
Despite the recent increase in the literature on farm/non-farm linkages, there
has been no significant systematic study conducted on marginal areas in Ethiopia,
particularly Tigray. Most studies concentrate on dynamic agriculture, where cash
crops are widely grown. The main purpose of this chapter is to inform government
agencies, NGOs and donors about the development and constraints of the rural
non-farm sector, its link to the farm sector and its importance for rural
development policy. This chapter addresses the following questions.
1. What are the patterns and determinants of rural households’ non-farm income
and participation in non-farm activities?
2. What types of linkages exist between farm and non-farm activities? What is
the relative importance of the linkages? Are farm and non-farm income
substitutes or complements? What is the influence of non-farm income on
income inequality in rural communities?
3. What are the major policy, social, cultural and economic problems for the
development of non-farm microenterprises and small enterprises? What kind
122 Rural farm/non-farm income linkages in northern Ethiopia

of policies, institutional support and technological developments are

necessary to alleviate constraints to non-farm employment?
The rest of this chapter is organized as follows: the second section describes
the data collected for this analysis and the method. The third section provides an
overview of the study region, including the roles of the government and NGOs,
the development and constraints of microenterprises in the region and the
functioning of the labour market. The fourth section provides empirical evidence
on the nature and relative strength of production and consumption linkages in
the region, non-farm work participation and its impact, and household-level farm/
non-farm income linkages; the roles of farm and non-farm activities in income
inequality among rural households is explored. Policy and programme
implications of this research are discussed in the fifth section.

DATA USE AND APPROACH

In order to analyse farm/non-farm linkages in Tigray, a survey of rural households
was conducted and there were interviews with labour-market participants and
major employers. Secondary data from national and regional statistics offices
was also used.
The survey data were collected in two districts, or woredas: Enderta and
Adigudom in the southern zone of Tigray. The survey consisted of 201 heads of
122households chosen randomly from a stratified sample area. The data include
rural
detailed information on the allocation of labour to home, farm and off-farm,
income sources, purchase of farm inputs TABLE 1
including hired labour, sale of farm outputs, Description of the household-level
consumption expenditures, credit and data set (n=402)
household composition. The data was Variables Mean
collected for the years 1996 and 1997 for
Family size 5.58
a total of 402 observations.
Number of dependents 3.26
Data characterizing the surveyed Land cultivated (tsimidi) 7.06
households are given in Table 1. Family Number of plots cultivated 3.65
size ranges from 1 to 11 people. On Land owned (tsimidi) 5.88
average, family size is 5.6 members, which Number of plots owned 3.06
is slightly above the regional average of Age of the household head 48.00
4.8 and the national average of 5.2. The Note: A tsimidi is a local area-measurement
average dependency ratio is 58 percent. unit; 1 ha = 4 tsimidis.
Most adults are illiterate; only 35 percent Source: Author’s calculations from household
survey.
of household heads can read and write.
Promoting farm/non-farm linkages for rural development 123

Rural households participate in a variety of farm, off-farm and self-

employment activities. Farming activities include crop production, accounting
for 20 percent of rural households, livestock husbandry, accounting for 6 percent,
and mixed farming of crops and livestock, accounting for 69 percent. Traditional
farming technology – simple hand tools, ox-driven implements and labour – is
the dominant farm input and comes mainly from the family. Only 5.7 percent of
rural household use irrigation. Off-farm activities involve wage employment
and self-employment. Wage employment includes paid community development
work, often called food for work, farm work and manual work in construction,
masonry and carpentry. Self-employment includes small trading, transporting
goods by pack animal, selling fuelwood, making charcoal, selling fruit, making
pottery and handicrafts and stone mining.

BACKGROUND ON THE TIGRAY REGION

Description of the study area
The study area is the Tigray region of northern Ethiopia, which is part of the
African dryland zone often called the Sudano-Sahelian region (REST/
NORAGRIC, 1995). The region has an area of 80 000 km2 and a population of
3.1 million people in 598 004 households; 85 percent of the population reside
in purely rural areas; the remaining 15 percent live either in the regional capital,
district centres or rural centres.
Tigray ranges from flat lowland to rugged mountainous plateaux where
altitudes range from 500 m in the eastern part, Erob, to 3 900 m in the southern
zone near Kisad Kudo. Natural resources are under extreme stress because of
the increasing population (REST/NORAGRIC, 1995). Many of the steep slopes
have lost their protective cover and are seriously overused by cultivation and
livestock grazing. Soil runoff from slopes has caused severe erosion (BPED,
1998). The natural forest of the region has been destroyed, mainly by
encroachment of subsistence cultivation; only localized patches of woodland
around churches and in remote places remain uncut. The crop production and
animal-husbandry potential of the region has declined as a result of degradation
of natural resources.
The region does not have well developed infrastructure (BPED, 1998); most
areas of Tigray are difficult to reach by mechanized transport. Until 1997, towns
in the region did not have a 24-hour supply of electricity; telephone lines and
postal services are inadequate and of low quality. Rural radio call telephone
124 Rural farm/non-farm income linkages in northern Ethiopia

systems are found in remote towns in the southern, eastern, western and central
zones. There are only seven postal branches in the region.
Table 2 presents data on the level and growth of regional GDP disaggregated
by economic activity. At 2 817.66 million Birr, the GDP of Tigray constituted
22 percent of the national GDP in 1994/95. Agriculture is the dominant sector
at national and regional level. Based on 1995–1996 estimates, agriculture, forestry
and fishing account for 64 percent of the regional GDP and 90 percent of
employment; industry accounts for 23 percent of regional GDP, services account
for 4 percent and other services account for 9 percent. In 1995/96, the overall
regional GDP grew by 7.3 percent in real terms. The service sector is the sector
with the highest growth in the region at 16.2 percent, followed by the industrial
sector at 6.9 percent. The growth of the industrial sector is largely driven by the
124 percent growth in large and medium-scale industry.
Agriculture in Tigray consists of crop husbandry, livestock husbandry and
mixed farming, which is the dominant type of farming system. In 1996, for
example, the proportion of farm households that were engaged in crop husbandry

TABLE 2
Regional (Tigray) gross domestic product by economic activity at constant factor
cost in 1994/95 and 1995/96 (million Birr except per capita GDP)
Economic activity Real gross value Growth rate (%)
124 1994/95 1995/96 1994/95–95/96
Agriculture, forestry and fishing 1 797.6 1 917.5 6.7
Industry 648.7 693.7 6.9
Mining and quarrying 155.0 175.7 13.4
Manufacturing 73.1 92.3 26.3
large and medium-scale manufacturing 13.2 29.6 124.2
small-scale industry and handicraft 59.9 62.7 4.7
Electricity and water 23.9 25.8 7.7
Construction 396.7 399.8 0.8
Services 371.3 412.5 11.0
Trade, hotel and restaurant 88.0 102.1 16.0
Transport and communication 35.4 41.4 16.7
Other services 247.9 269.1 8.5
Total regional GDP 2 817.7 3 023.7 7.3
Population (million) 3.114 3.193 2.5
Regional per capita GDP (Birr) 904.7 946.9 4.7

Source: Regional Bureau of Planning and Economic Development of Tigray Region.

Promoting farm/non-farm linkages for rural development 125

was 19 percent, in livestock husbandry 3 percent and in mixed farming 78 percent.

The growing population has decreased the average farm size in the region to
only 0.97 ha; 70 percent of farm households own less than 1 ha. Agricultural
output is mainly for domestic consumption, although some production is for
export, including sesame oil, horse beans, field peas and animal skins and hides.
Farming systems in Tigray are characterized by traditional technology based
entirely on animal traction and rainfed land. A variety of crops such as cereals,
pulses and oil crops grow in the region. Cereals are the dominant crop; pulses
are of secondary importance. The major crops are sorghum, teff, barley and
wheat. Vegetables are cultivated in limited amounts in areas where irrigation
water is permanent or semi-permanent, and in areas closer to rural centres and
urban areas. In Enderta and Adigudom districts, for example, 6 percent of farm
households grow vegetables, but 78 percent of them live near towns. Scarcity of
arable land leads to extremely intensive land use. Farmlands are owned and run
by small farms that are divided into minor plots scattered over large areas.
Production is family based, with little hired labour. Livestock except for plough
oxen play an important but secondary role. The livestock component increases
in the lowland areas, especially in the southern and western zones (REST/
NORAGRIC, 1995).
The Tigray region is marginal compared with the southern and central part
of the country. Agricultural production in the region is below the national average.
In a good year such as 1996, average yield is 1 167 kg/ha nationally but only
1 096 kg/ha in Tigray; in bad years of drought, the yield gap is even greater. The
most basic constraints for agricultural development, especially crop production,
are unreliable rainfall, lack of oxen, low soil fertility and outbreaks of crop
pests. In the central zone, for example, unreliable rainfall is seen by farm
households to be the most significant problem, followed by crop pests and lack
of oxen (REST/NORAGRIC, 1995). Lack of pasture and fodder are the main
constraints in animal production. Scarcity of veterinary clinics is a significant
constraint in livestock development. The revival of livestock farming after a
drought is very difficult, because a great number of cattle die during droughts.
It will, in short, be very difficult to increase employment in agriculture.
Agricultural productivity is very low because of low soil fertility and unreliable
rainfall. Livestock husbandry in Tigray is constrained by a shortage of grazing
land. Animal dung is used as fuel for cooking, not for enriching the soil; expansion
into marginal and steeper slopes is a widespread practice. The result has been
widespread degradation of highland soils as a result of erosion. Labour absorption
in agriculture is only possible through intensification of agricultural production
126 Rural farm/non-farm income linkages in northern Ethiopia

and the use of irrigation, which is unlikely in the near future. The non-farm
sector has not yet sufficiently developed to absorb the growing population; its
contribution to the overall employment and income generation remains low.
The role of government and NGOs
The 1974 revolution resulted in a series of policy measures aimed at expanding
collective and state-owned farm and non-farm enterprises and managing the
economy through central planning. The government restricted individuals to a
single type of occupation. Farmers were not allowed to engage in off-farm
activities, hire of labour was restricted and farmers were forced to be members
of producer and service cooperatives. These cooperatives were given priority
for most of the financial-assistance and extension services. Industrial products
were distributed through the service cooperatives; private traders in rural areas
did not officially exist.
Monopolized public institutions were given the responsibility for promoting
the non-farm sector. The Rural Technology Promotion Department (RTPD) was
entrusted with the task of developing and promoting improved farm and non-
farm technologies and food processing. The Handicraft and Small Industrial
Development Agency (HASIDA) was in charge of issuing licenses, organizing
cooperatives and assisting in the marketing of products. Ministry of Education
adult training centres attempted to teach various handicrafts, construction and
farming
126 skills in urban and rural areas. Their efforts were, however, constrained
by policy and institutional factors from the very beginning. All promotional
activities were directed towards cooperatives. Individuals trained in crafts were
unable to establish themselves, because they lacked credit, tools, raw materials
and business skills.
When the military government collapsed in 1991, a market-based economy
replaced the centrally planned economy. After the formulation of the Federal
Democratic Republic of Ethiopia in 1995, the government decided to liberalize
the economy and promote investment in the agricultural and industrial sectors.
Current government policy emphasizes both sectors, but is less precise with
regard to the rural non-farm sector. The focus of the recent economic reform is
on structural adjustment aimed at strengthening producers’ supply response,
developing the private sector, promoting growth of financial intermediation and
creating a market for privatization of financial assets.
The main objective of government agricultural policy is to ensure adequate
food security through increased agricultural production and employment. A broad
Promoting farm/non-farm linkages for rural development 127

based Agricultural Development-led Industrialization (ADLI) strategy has been

formulated that concentrates on three priority areas: accelerating growth through
the supply of fertilizer, improved seeds and other inputs, expanding small-scale
industries to interact with agriculture and increasing exports to pay for the import
of capital goods. Under ADLI, a new system of agricultural extension called a
participatory demonstration and training extension system was launched in 1994–
1995. It provides agricultural inputs in package form with extension advice.
The reform process, particularly structural adjustment, has affected the earlier
institutions that were in charge of promoting non-farm activities. RTPD, for
instance, was brought under the regional Bureau of Agriculture but has been
limited by budget and manpower constraints. HASIDA offers technical and
managerial services to small-scale industry and handicrafts, but its operations
are financed through the revenue generated by charging for services; it is still
being reformed and its services cover only selected urban areas.
With the reform of state institutions, a number of international, national and
regional NGOs have become directly involved in farming and non-farming
communities. They provide farmers with a variety of services mainly focused
on agricultural development, reforestation, soil and water conservation and rural
water supply, and credit for income-generating activities, including small-scale
trade and handicrafts. The emphasis on rural non-farm activities is minimal.
The exceptions are the regional NGOs Relief Society of Tigray (REST) and the
Tigray Development Agency (TDA), which are actively engaged in a more
diversified set of activities than the international and national NGOs.
REST’s involvement in the rural non-farm sector is mainly through its rural
credit and saving programme, which operates through 12 branches and 103 sub-
branches across the region. It provides loans for cottage and small agro-industries,
artisans engaged in rural arts, crafts, horticulture, growing cash crops and rearing
pigs. The activities for which loans are provided are:
• crafts such as embroidery, pottery, basket making, spinning, weaving,
carpentry and metal work, especially for agricultural implements;
• small-scale trade such as buying and selling at open markets, retailing,
hairdressing, tailoring and preparing local food and drugs;
• agriculture such as livestock rearing, bee keeping, horticulture and cereal
production.
A loan is provided on a group basis at 12.5 percent interest. The maximum
loan is 5 000 Birr; the minimum is 50 Birr. The duration of the loan is one year,
128 Rural farm/non-farm income linkages in northern Ethiopia

depending on the repayment capacity of the borrower and the nature of the
activities. The REST credit programme has improved farmers’ access to the
financial market, but cannot completely satisfy farm households’ demand for
credit; loan duration is short and is not coupled with sufficient business advice.
TDA is primarily involved in providing basic education and technical training.
It was initially involved in an integrated rural development programme, but since
1996 its focus has been on urban and rural education. TDA runs four technical
training centres, two in the central zone in Shire and Axum and two in Mekelle.
School dropouts, ex-soldiers, farmers, women and the jobless are allowed to
join. Training is provided in basic construction skills such as masonry and
carpentry, metalwork, woodwork, electrical systems, motor mechanics and
handicrafts such as carpet making. Graduates are provided with the tools and
credit to start their own businesses, but their capacity is very limited because of
financial and accommodation problems.
In short, it has been and still is unclear which government organization is
responsible for the promotion of non-farm activities in rural areas. The minimal
promotion of non-farm activities coordinated by the Bureau of Agriculture
through RTPD is not suited to rural areas. HASIDA activities are not targeted to
rural non-farm activities in general or to rural non-farm activities carried out by
farm households. Substantial promotional work on farm and rural non-farm
activities has been done by regional NGOs, especially REST and TDA; they
128to be better than the Government at targeting the rural poor and rural non-
seem
farm activities, but their activities still require more coordination with government
organizations to improve efficiency and avoid duplication.
Micro and small-scale enterprises: developments and constraints
Statistics from 1998 published by the Tigray Regional State Bureau of Industry,
Trade and Transport (ITTB) show that small-scale manufacturing enterprises
have flourished during the last seven years. In 1991, small-scale industry barely
existed except for cottage industries, of which there were 206 in 1994, reflecting
remarkable growth. In 1997 they totalled 599, providing employment for
approximately five people per establishment. Average capital investment per
establishment is 153 000 Birr. Grain mills are the most common type of small-
scale industry. About 20 percent of small-scale manufacturing enterprises are
found in Mekelle, the capital of the region; the remaining 80 percent are found
in other zones and woreda towns (Table 3). Most of the raw materials used for
production are locally produced; imports constitute about 14 percent of total
raw materials.
Promoting farm/non-farm linkages for rural development 129

TABLE 3
Distribution of small-scale manufacturing enterprises in Tigray
Tigray Mekelle Southern East Central Western

Number 599 117 117 93 105 168

Investment (Birr) 91 651 331 34 252 604 15 550 533 20 872 695 9 369 499 11 983
Employment 2 957 842 765 521 373 558

Source: ITTB Statistical Bulletin No. 1, 1998.

The Central Statistics Authority (CSA) estimates that there are 25 012 cottage
and handicraft enterprises, of which 9 percent are found in Mekelle; the remaining
91 percent are in other towns of the region. Cottage industries are known to use
more locally produced raw material than small-scale manufacturing industries.
Cottage industry in the region covers a variety of industrial groups, including
the following major products: food and beverages, textiles and non-metallic
mineral goods, which constitute 90 percent of regional cottage industries. Average
initial capital invested per establishment for rural areas is 376 Birr, and 276 Birr
for urban areas. Most of the finance for initial investment – 44 percent – comes
from people’s own savings and from friends and relatives.
Trade, the most common non-farm activity, has grown rapidly over the last
seven years. In 1995, the growth rate of this sector was 16 percent. If unlicensed
trade by farmer households is included – it is often underestimated in GDP
calculations – the growth rate would have been much higher: 234 wholesale
businesses, 11 765 retail enterprises and 2 799 service providers have been
established in the region, about 16 percent of which are in Mekelle and the rest
in zone and woreda centres. The initial capital required for the retail trade is
lower than for wholesale and service businesses (Table 4). Average initial capital
per establishment is 31 301 Birr for wholesale, 4 326 Birr for retail and 14 992
Birr for services. Women own most of the service establishments such as bars
and beauty salons, where value-added per unit of investment is the lowest; most
wholesale and retail establishments are owned by men. The educational status
of the owners of wholesale and retail establishments is comparable; people with
elementary education are the owners of most of the trade establishments. The
dominant type of ownership is sole proprietorship. In Mekelle, for example, all
of the wholesale trade, 98 percent of the retail trade and 99.8 percent of the
service trade are under individual proprietorship. A possible reason for the
dominance of sole proprietorship is the fear of friction among partners and the
transaction costs associated with resolving a dispute.
130 Rural farm/non-farm income linkages in northern Ethiopia

TABLE 4
Characteristics of commerce in Tigray
Type of trade Initial capital per Female Owner Owners
establishment (Birr) owners (%) illiterate (%) grade 1–6 (%)

Wholesale 31 301 11 16 56
Retail 4 326 25 22 61
Service rendering 14 922 71 34 42

Source: Calculated from data provided by ITTB.

TABLE 5
Value-added and employment potential of non-farm activities in Tigray
Type of Initial capital investment Value added per Value added
non-farm activity (Birr) per unit of unit of investment per person
employment provided

Cottage industry 269 2.21 595

Small-scale industry 3 508 1.42 4 966
Total commerce 202 075 0.004 804
Wholesale 156 941 0.04 6 023
Retail 18 000 – –
Services 426 406 0.004 1 828
130
Source: Calculated from CSA Statistical Bulletin No. 182.

Cottage industry and small-scale manufacturing industry have a greater role

than commerce in providing employment and generating income (Table 5). In
Tigray, cottage industries require 269 Birr of capital investment to employ one
person; the figure for small-scale manufacturing industry is 3 509 Birr. Value-
added per unit of investment is lower for commerce, generating 0.004 Birr per
unit of initial investment. Cottage industry generates 2.21 Birr per unit of initial
investment; the figure for small-scale industry is 1.42 Birr.
Constraints to the development of small enterprises and microenterprises
fall into two categories: insufficient infrastructure and firm-specific limitations.
The infrastructure problem relates to the low quality and insufficient supply of
roads, electrical power and telephone lines; the main road connecting other
regions and the Government is not well maintained. There was no supply of
electricity that could provide power to a manufacturing industry until May 1998
in most urban areas. Since May 1998, most towns have received electricity, but
Promoting farm/non-farm linkages for rural development 131

it can still take several months to get electrical power connected because of
shortages of electrical equipment. The capacity of the government office
responsible for the service is limited. Business people have to spend an enormous
amount of time ordering and receiving raw materials and other commodities.
CSA statistical abstracts have documented the firm-specific problems in
cottage and small-scale manufacturing industries and trade; these are summarized
in Table 6. In cottage handicrafts and small-scale manufacturing enterprises, the
major problem is lack of sufficient initial capital, which affects 48 percent of
cottage industries 36 percent of small-scale manufacturers. Other problems are
lack of adequate start-up skills in cottage industries and lack of raw materials
and premises in small-scale enterprises. A few small-scale and cottage industries
are not working at full capacity, for which the main reasons are absence of market
demand for products, shortages of raw materials and lack of working capital:

TABLE 6
Problems faced by small and microenterprises in Tigray
Business type Problems in starting business Operational difficulties
(% of responses) (% of responses)

Cottage/handicraft Lack of sufficient initial capital 36 Lack of sufficient initial capital 48

enterprises Lack of continuous supply of 15 Lack of adequate skill 11
raw materials Absence of market demand 42
Lack of working premises 12
Small-scale Lack of sufficient initial capital 36 Lack of working capital 16
enterprises Lack of continuous supply of Shortage of supply of raw
raw materials 15 materials 11
Lack of working premises 12
Wholesale Lack of sufficient own capital 21 Limited market 29
Lack of working premises 19 Shortage of working capital 19
No problem 45 Lack of workplace 9
No problem 16
Retail Lack of sufficient own capital 36 Shortage of working capital 38
Lack of working premises 17 Limited market 31
Government regulations 6 Lack of workplace 6
Service trade Lack of working premises 17 Lack of working premises 37
Lack of sufficient own capital 9 Shortage of working capital 24
Access to raw materials 7 State harassment 9
Government regulations 4

Source: CSA Statistical Bulletin Nos. 172, 179 and 182.

132 Rural farm/non-farm income linkages in northern Ethiopia

the main problems in trade enterprises are similar. About 6 percent of retail
establishments and 4.2 percent of service providers reported that government
regulations are a problem in starting businesses; 9 percent of service providers
in Tigray report government harassment while operating a business. These
problems appear less acute for wholesale traders: about 46 percent of
establishments reported no problems in starting up.
Labour markets
Detailed analysis of the labour market is beyond the scope of this paper. This
section summarizes the results of an analysis of the labour market conducted
using the data collected for the present study (Woldehanna, 1998).
Farm households are not entirely self-sufficient and are partially integrated
into the labour market as employers and labourers. The demand for hired labour
among smallholders occurs in peak agricultural seasons such as harvesting and
weeding and in slack seasons, implying that some farm households are labour
constrained even in slack seasons. This shows that public programmes scheduled
for slack seasons are not without opportunity costs. The non-farm wage rate for
farm households is influenced by location, type of wage employment and year; the
rate varies with location, implying that there is a lack of labour mobility, which
requires further investigation. Age and education affect the wage rate received by
the main woman in the household, but not others. This may be due to the fact that
most
132jobs do not require education (Rosenzweig 1978, 1984 and 1988). The wage
rate varies with season and TABLE 7
activity, implying that wage Forward and backward production linkages in Enderta
rates reflect demand and and Adigudom districts (household-level data)
supply of labour and the Linkages Birr/household
amount of effort required to Backward production linkages 64.27
carry out a job. The Expenditure on fertilizer 61.78
relevance of subsistence or Expenditure on insecticide 0.62
nutritional determination of Expenditure on herbicide 0.01
wage is thus very low. Expenditure on veterinary medicine 1.87
Wage determination is Labour market linkages
better explained by the Expenditure on hired farm labour 89.85
forces of supply and Forward production linkages
demand and effort required. Sale of crop output 252.64
Farm households in the Sale of livestock product sold 72.94
farm-labour market and Sale of livestock 176.99
other employers in the non-
farm labour market rely on Source: Author’s calculations from household survey data.
Promoting farm/non-farm linkages for rural development 133

relatives and friends to search for labourers to hire. Most workers rely on relatives
and friends to get information about jobs. Labour-market limitations have
considerably increased transaction costs associated with hiring labour and searching
for jobs. Most of the people working as masons and carpenters acquire their skill
after a long apprenticeship, which is slow and unproductive. This has led to a
shortage of skilled masons and carpenters for construction and other investment
activities.

FARM/NON-FARM INCOME LINKAGES, SURVIVAL STRATEGIES AND INCOME

INEQUALITY

Production linkages
Backward and forward production linkages in the region are limited (see Table 7).
Farmers purchase few farm inputs such as fertilizer and pesticides. The average
value of fertilizer used per household in the southern part of the region is 62
Birr, which is a very small percentage of farm output. There is little use of
pesticides and veterinary medicine. Households consume most of their farm
produce; sales of crop and livestock output are still at a low level. A farm
household sells on average only 13 percent of its crop and 15 percent of its
production from animal husbandry. Agriculture is in general unable to support
major processing industries; there is only one food-processing industry and one
tannery in the region.
Table 8 shows the correlation between non-farm activities and a number of
variables. The results indicate that district-level non-farm activities are positively
related to population density, but its correlation with farm income is very weak.
This is because agriculture has limited backward and forward production linkages.
The correlation of wholesale and retail trade with agricultural income is much
higher than with service trades and small manufacturing industries, which
indicates that consumption linkages are higher than production linkages, and
that farmers are significant users of wholesale and retail trades. Service trades,
small enterprises and microenterprises are negatively correlated with farm output,
probably because farmers are forced to participate in non-farm activities when
agriculture is unable to support the growing population. This supports the residual
sector hypothesis that non-farm activities absorb workers who cannot be readily
absorbed into agriculture. Rural centres are an important stimulus for the
performance of microenterprises and small-scale enterprises. Districts closer to
Mekelle have access to some of the services needed to run small-scale
manufacturing industries, such as roads, electricity and telephone lines; the further
134 Rural farm/non-farm income linkages in northern Ethiopia

TABLE 8
District level correlation between farm income, population density and capital
invested in non-farm income in Tigray
Non-farm Retail Wholesale Service Small manuf.
activities trade trade trade industry

Population density 0.4695 0.5660 0.2614 0.6150 0.2626

Actual farm output (100 kg) 0.0164 0.0904 0.2711 -0.1012 -0.1292
Farm output per capita 0.0430 0.0848 0.3115 -0.0852 -0.1057
Potential farm production 0.0930 0.1345 0.3618 -0.0445 -0.0784
Distance from Mekelle -0.0931 0.0301 0.1152 -0.1578 -0.1954
Distance from zonal town -0.1880 -0.1873 -0.0460 -0.2332 -0.1570

Source: Author’s calculations.

districts are from Mekelle and rural towns, the lower the amount of capital
invested in service trades and small-scale manufacturing.
Consumption linkages
Consumption linkages are the strongest type of linkage in Tigray. Demand for
consumption goods increases as agricultural income increases, but the commodity
composition of that demand will change as some commodities and services
increase in importance while others diminish. Household consumption demands
134
are more complex, with varying income elasticity of demand for individual
commodities. Analysis of consumption demand of the farming population
therefore deserves special attention.
To analyse the relative importance of different commodity groups in demand
linkages, marginal budget shares and expenditure elasticity are derived from
Engel functions, estimated using ordinary least squares, which have a nonlinear
relationship between consumption and income.
Table 9 summarizes the expenditure behaviour of the average farm household.
The results are obtained by evaluating average budget share, marginal budget
shares and expenditure elasticity at the sample mean. The commodities consumed
are categorized into food and non-food items and into locational groups. The
food items include cereals, pulses, oil crops, vegetables and animal products
such as milk, butter and cheese, and sugar, tea and salt. The non-food items are
grouped into social expenses such as services and ceremonial expenditure,
contributions for local organizations and taxes and industrial products such as
household durables, clothing and footwear.
Promoting farm/non-farm linkages for rural development 135

TABLE 9
Demand linkages in Enderta and Adigudom districts, southern zone, Tigray
Average budget Marginal budget Expenditure
share share elasticity

Total food expenditure 0.79 0.69 0.87

Cereals 0.49 0.20 0.41
Pulses 0.05 0.06 1.20
Oil crops 0.004 0.01 2.79
Vegetables 0.001 0.001 1.00
Animal products 0.11 0.25 2.27
Coffee, sugar, tea, salt, spices 0.14 0.17 1.21
Total non-food 0.21 0.31 1.48
Service, ceremonial and other
social expenses 0.05 0.11 2.20
Industrial products 0.16 0.20 1.25
Household goods 0.01 0.02 2.00
Clothes, shoes and cosmetics 0.15 0.19 1.27
Locational group
Own produced food 0.52 0.49 0.94
Purchased food, local 0.13 0.02 0.15
Purchased food, non-local 0.14 0.17 1.21
Industrial products, non-food
(not locally produced) 0.16 0.20 1.25
Purchased locally, non-food 0.05 0.11 2.20

Source: Author’s calculations from household survey data.

Food accounts for 79 percent of total expenditures, leaving a small share of

the budget for non-food items. The marginal budget share of food items is
65 percent, which is less than the average budget share. Expenditure elasticity is
0.87, implying that the budget share of food items will decline when total income
rises. The result is comparable to the 0.88 reported by Hazel and Hojjati (1995)
for Zambia and the 0.81 reported by Hazel and Roell (1983) for the Gusau
region of Northern Nigeria.
Among food items, cereals account for 0.49 percent of the budget, but their
importance declines as income rises. Expenditure elasticity is 0.41, and the
marginal budget share is 0.20, less than half the average budget share. For higher-
value food items including pulses, oil crops, vegetables, animal products, and
coffee, sugar, tea, salt and spices, expenditure elasticities are very high, which
136 Rural farm/non-farm income linkages in northern Ethiopia

implies that their budget share will increase if household income increases; their
average budget shares are currently very low.
All non-food items have high expenditure elasticities, implying that their
importance in the budget share will increase as farm-household income rises.
The relative increase will be greatest for expenditures in service, ceremonial
and other social expenses, and expenditure on clothes and footwear. This clearly
shows that agriculture has the potential to strengthen local demand for non-food
items in Tigray.
The locational linkage results in Table 9 show that about 86 percent of total
expenditure is on regionally produced food and non-food items; the remaining
14 percent are regionally imported non-food items. Expenditure elasticities of
imported items, however, are higher than those of local items, which indicates
that there are strong household demand linkages to the local economy that are
predominantly benefiting the agricultural sector in the short term but that these
will diminish when farm-household incomes rise. The average budget share of
local products of the non-farm service sector is 0.05; expenditure elasticity is
2.2. Household demand linkages that go to the local non-farm sector such as
services and ceremonial expenditures are currently very low, but will increase
dramatically when farm-household income rises. Food items imported from
outside the region are also potentially important.
Off-farm
136 work participation and its impact on farm income
Rural households in Tigray participate in various off-farm activities such as
wage employment and non-farm own business. A summary of rural households’
participation in off-farm activities for two districts is shown in Table 10;
81 percent participate in some off-farm activities. Most off-farm work is
temporary and does not require skilled labour except for masonry and carpentry.
The proportion of households that participate in wage employment is 72 percent;
for own business activities the figure is 28 percent. The participation rate in
non-farm wage employment is 22 percent. When food for work is excluded, the
off-farm work participation rate of farm households is high at 43 percent.
Rural household income can be divided by activity source. On average, farm
production accounts for 57 percent of total income; livestock accounts for
16 percent and crop production for 41 percent. Off-farm labour income accounts
for 35 percent and non-labour income accounts for 8 percent of total income.
The amount of non-labour transfer income for households is very small compared
to their farm and off-farm income. The types of transfer income are remittances,
Promoting farm/non-farm linkages for rural development 137

TABLE 10
Farm household participation in off-farm activities in Enderta and Adigudom
districts
Type of off-farm activities Participation rate (%)

Own off-farm business 27.9

Total wage employment 71.5
Non-farm wage employment 21.6
Manual non-farm wage employment 19.2
Masonry and carpentry 3.5
Food for work 57.7
Off-farm work participation excluding food for work 43.0
Overall off-farm work participation 81.0

Source: Author’s calculations from household survey data.

47 percent, food aid, 20 percent and gifts and inheritances from relatives,
19 percent.
An economic analysis of the impact of off-farm activities on agricultural
productivity and output was conducted in an earlier paper (Woldehanna, 1998).
Results showed that income diversification results in higher agricultural output
per unit of land. On average, when off-farm income increases by 1 percent,
agricultural productivity increases by 0.34 percent. This could be because
households learn managerial skills through experience in various activities, reduce
soil mining and initiate better farming practices.
Apart from the diversification effect of off-farm activities, off-farm work
brings additional income to households that can be used to purchase farm inputs.
Results indicate that demand for hired farm labour is highly influenced by farm
and off-farm income controlling for other factors such as family size, number of
dependents, soil quality, year and location dummies. When farm output increases
by 1 percent, expenditure on hired farm labour increases by 0.66 percent; the
figure in relation to off-farm output is 0.55 percent. The demand for other variable
inputs is also highly influenced by farm output and off-farm income. When farm
output increases by 1 percent, expenditure on the variable input increases by
0.94; the figure in relation to off-farm output is 0.43 percent. To sum up the
effect of off-farm income on farm productivity through purchase of farm inputs
and income diversification, the total elasticity of farm productivity with respect
to off-farm income is calculated to be 0.39.
138 Rural farm/non-farm income linkages in northern Ethiopia

Off-farm income, entry barriers and income inequality

In order to evaluate the relationship between off-farm income and inequality, a
Gini decomposition by income source was calculated (Table 11). Crop income
has the highest contribution to overall income inequality as measured by the
Gini coefficient, followed by wage employment and income from livestock.
Crop income, livestock income and wage income decrease income inequality.
The results are mixed, however, when wage income is broken down into
categories. Paid food-for-work programmes are the only type of off-farm income
that decreases income inequality. Non-farm wage and self-employment income
increase inequality, as does income from unskilled manual and skilled manual
non-farm work. Non-labour income such as gifts, remittances and property rents
also increase income inequality.
A possible reason why non-farm income has an unequalizing effect is that
there is an entry barrier for the poor. Non-farm wage employment and self-
employment require skill and capital to start. In the absence of a perfect credit
market, only wealthier households can afford to enter into self-employment.
This implies that if rural non-farm activity programmes do not particularly target
the poor, wealthy farm households will dominate the most lucrative non-farm
activities such as masonry, carpentry and trade. The inequality effect of non-

TABLE 11
138
Gini decomposition by income sources
Household income Mean Sk Rk Gk Gk*Rk Sk*Rk* (Sk*Rk %
components Gk *Gk )/G elasticity
of Gini index

Off-farm self-employment 262 0.068 0.598 0.836 0.500 0.034 0.103 3.5
Off-farm wage 858 0.280 0.489 0.628 0.308 0.086 0.261 -1.9
Food for work 437 0.174 0.183 0.664 0.122 0.021 0.064 -11.0
Manual non-farm wage 284 0.085 0.406 0.883 0.358 0.030 0.092 0.7
Skilled non-farm wage 136 0.022 0.794 0.978 0.777 0.017 0.053 3.0
Non-labour income 194 0.039 0.707 0.951 0.672 0.026 0.080 4.1
Net farm crop income 1 339 0.448 0.698 0.442 0.308 0.138 0.419 -2.9
Livestock income 497 0.164 0.425 0.643 0.273 0.045 0.136 -2.8
Total household income 3 152 0.330

Note: Sk is the average share of income source k in total income; Gk is the Gini index of inequality for
income source k; Rk is the Gini correlation with total income ranking; G is the Gini index of total
income inequality; (Sk*Rk*Gk )/G is the percentage contribution of income source k to the Gini index of
total income inequality.
Source: Author’s calculations from household survey data.
Promoting farm/non-farm linkages for rural development 139

farm income has serious policy implications. If the objective of policy makers is
to reduce income inequality, poverty-focused rural non-farm investment has to
focus on the type of non-farm activities in which the poor can participate. If this
is not possible, the underlying factors that hinder rural households’ participation
in non-farm activities must be addressed. This requires the establishment of
training centres and provision of credit that focus on the rural poor.

DISCUSSION OF POLICY AND PROGRAMME IMPLICATIONS

The need for alternative employment opportunities
It is becoming very difficult to increase regional employment in agriculture. A
growing population has decreased farm size, leading to expansion into marginal
and steeply sloping land; the result has been widespread degradation of the
highlands. Crop residue and animal dung are used as fuel for cooking, not for
enriching the soil. Scarcity of land and malaria in lowland areas such as the
western zone limit the amount of land under cultivation in the region. Livestock
production is not promising either, because forage supplies come from
unimproved and overgrazed pasture and crop residue. Poverty is pushing farmers
to search for alternative sources of income, particularly wage employment. To
reduce the pressure on land, rural non-farm activities have to be expanded.
Waiting for non-farm activities to expand until poverty pushes people off the
land will further degrade natural resources and eventually increase the cost of
rehabilitation. Employing rural people through rural non-farm activities has two
advantages: it keeps farmers in the rural areas and reduces rural urban migration,
and it provides farmers with additional income and reduces the pressure on
land, hence reducing land degradation.
Development of complementary policies and organized promotional
activities
Off-farm income is important for the rural economy in Tigray. Rural households
with diversified sources of income have higher agricultural productivity.
Expenditure on farm inputs is dependent on off-farm income, which helps to
finance farming activities, as well as agricultural production. Farmers employed
in higher-wage activities such as masonry and carpentry have a greater capacity
to hire farm labour. The positive link between farm and off-farm income implies
that increasing agricultural output and raising agricultural productivity cannot
be done in isolation. Narrowly focused sectoral approaches with the sole target
of raising agricultural output and productivity are less likely to achieve significant
140 Rural farm/non-farm income linkages in northern Ethiopia

advances unless considerable attention is given to the importance of non-farm

income in the rural economy. Current agricultural extension programmes should
include farm and non-farm activities, encourage growth of small-scale business
and create non-farm employment opportunities in rural areas. Complementary
policies and programmes must be developed to strengthen the link between farm
and non-farm activities.
There are attempts in the region to promote rural non-farm activities in order
to provide farm households with alternative income sources. Public employment
schemes such as food for work, for example, have increased farm households’
access to off-farm income to about 35 percent. Efforts to promote off-farm
activities are disorganized and insufficient, however, and the links between farm
and non-farm activities are not fully recognized. Most government organizations
and NGOs have focused exclusively on agriculture, because the majority of the
population is engaged in it. Non-farm activities should not be left to the industry
and trade ministries; the agricultural ministries should be able to give special
focus to rural non-farm activities in order to ensure sustainable farming.
Institutional support might be necessary to create an enabling environment
for rural non-farm enterprises. It is not at present clear which government
organization is responsible for the promotion of non-farm activities in rural areas,
despite efforts by a few NGOs such as REST and TDA. The Bureau of Agriculture
concentrates on farming activities and the Industry and Commerce Bureau focuses
on140
non-farm activities in urban centres. A government organization must therefore
be established to coordinate promotion of rural non-farm activities; it should be
responsible for formulating, upgrading, coordinating and implementing measures
such as economic and financial policies to create an enabling environment, and
should run assistance programmes to promote rural non-farm activities. The
new institution could also lobby for policies that favour rural non-farm activities
and development of assistance programmes, because rural non-farm enterprise
owners do not have the capacity to organize themselves.
The fact that consumption linkages dominate production linkages signifies
that commerce is the main non-farm activity in the short term and should be the
focus of government policy. To exploit this potential, government regulations
that hinder expansion of business must be avoided. Infrastructure such as roads,
electricity and telephone connections must be improved; measures to improve
the efficiency of the economy, such as improving the bureaucratic and judiciary
systems, would also help. Improving the efficiency of commerce means creating
favourable markets for industrial products, especially those of small-scale and
cottage manufacturing industries.
Promoting farm/non-farm linkages for rural development 141

Rural towns as a focal point in rural development

Rural towns act as a focal point in the development of the rural non-farm economy.
It is essential to ensure adequate economic and social infrastructure to support
emerging rural non-farm activities and to renovate and develop traditional ones.
Physical infrastructure will undoubtedly play a significant part in strengthening
farm/non-farm linkages. Road access to rural towns is essential to stimulate
agricultural consumption linkages and provide farm inputs to farm households;
the roads will be relatively easy to construct and maintain. Efficient rural credit
and smooth labour markets are particularly important to promote rural non-farm
activities; investment in human capital is essential to ensure that rural non-farm
activities are reliable and able to cope with new technological developments.
Most of these service institutions must be located in rural centres, because it
would be difficult to establish them in widely scattered settlements.
Targeting of the vulnerable group
Care must be taken in planning programmes to combat rural poverty, because
promoting rural non-farm activities will not necessary target the poor. Wealthy
farm households dominate the most lucrative non-farm activities, particularly
masonry, carpentry and non-farm self-employment such as trade. Poverty-focused
rural non-farm investment will need to target non-farm activities in which the
poor can participate, or address the underlying factors that prevent poor rural
households from participating. This calls for establishment of training centres,
provision of credit for the poor, business extension, creation of favourable
conditions and improvement of infrastructures.
Women participate to a considerable extent in non-farm activities. They are
engaged in activities with lower value-added per unit of investment and low-
wage non-farm activities, however, such as public work programmes and manual
work in construction sites. Women need to be given training and provided with
credit to start their own businesses in order to allow them to participate in well
paid rural non-farm activities. To facilitate women’s contribution to rural non-
farm activities, assistance agencies must be involved and the Government must
explicitly recognize the importance of women’s roles.
The need to review and update existing policies and institutions
Government policies affect the magnitude of agricultural growth and the ability
of rural non-farm enterprises to response to agriculturally induced increases in
demand. If rural non-farm enterprises are to achieve their full potential for income
generation, policy makers need to review their agricultural, investment and
142 Rural farm/non-farm income linkages in northern Ethiopia

commercial and infrastructure development policies that work against small

farmers and small rural non-farm enterprises. A policy that needs reform is the
proclamation that provides investment incentives such as income-tax relief to
local investors with over 250 000 Birr capital, which does not encourage rural
non-farm activities that require smaller capital investment. Policies should be
formulated to improve access by small rural non-farm enterprises to formal
financial institutions such as commercial and development banks.
Searching for jobs and looking for employees are hindered by the fact that
most households rely on friends and relatives for information. The transaction
cost associated with hiring labour and searching for jobs is exceptionally high,
which will retard investment activities and technological innovations in the farm
and non-farm sectors. There should therefore be government assistance to help
dealers emerge in the labour market in the long term. One way to achieve this
would be to cancel the law that prohibits the establishment of dealers in the
labour market and to publish in recognized places labour-market information
such as wage rates, magnitude and type of labour demand and lists of job seekers
by skill until the market supports the emergence of dealers.
Because of the time required to develop masonry and carpentry skills, there
is a shortage of well qualified masons and carpenters for construction and other
investment activities. Hands-on training of workers in building and construction
sites should be organized to improve the performance of investment activities; it
142 probably be necessary to improve the capacity of government and NGO
would
training institutions and to establish additional training programmes. TDA
masonry and carpentry training for farmers should be developed to cover the
whole region; vocational schools or local master craftsmen could give the training
programme as well. The most important features of successful training
programmes are those linked with the labour market: unless training
establishments respond to changing labour-market conditions, their graduates
will encounter difficulties in finding employment and the investment in training
will be socially unproductive.
Choosing the means of intervention
The evidence suggests that programmes aiming to provide a complete package
of financial, technical and management assistance are generally less effective
than programmes that identify and provide a single missing ingredient such as a
small credit programme (Haggblade, Hazell and Brown, 1989). Assistance to
firms is usually costly, especially for developing countries, because the rural
non-farm enterprises are normally small and geographically dispersed. Direct
Promoting farm/non-farm linkages for rural development 143

assistance should focus on system-level opportunities and constraints that open

up opportunities for large number of firms (Haggblade, 1995). This kind of
highly leveraged intervention requires subsector analysis to identify locations
or enterprises supplying inputs or marketing outputs that can expand the income
potential for many small firms, upstream or downstream.
Continuous research is necessary to identify opportunities and constraints
that hinder the development of rural non-farm activities. National agricultural
research centres, universities, international research organizations and donor
agencies should include rural non-farm activities in their programmes in order
to enhance the effectiveness of the research and increase the familiarity of policy
makers and practitioners with the nature and development of rural non-farm
enterprises. Surveys by CSA and national and regional bureaux of industry, trade
and transport should include rural non-farm activities by large and small firms
in the formal and informal sectors.

REFERENCES
BPED. 1998. Atlas of Tigray. Mekelle, Tigray, Ethiopia.
CSA. 1997a. Report on cottage/handicraft manufacturing industries Survey. Statistical
Bulletin, No. 182.
CSA. 1997b. Report on distributive and service trade survey. Statistical Bulletin, No. 179.
CSA. 1997c. Report on small-scale manufacturing industries survey. Statistical Bulletin,
No. 172.
Evans, H.E. & Ngau, P. 1991. Rural/urban relations, household income diversification
and agricultural productivity. Development and Change, 22: 519–545.
Haggblade, S. 1995. Promoting rural industrial linkages within agrarian economies
for rural poverty alleviation. Vienna, UNIDO. (Draft report.)
Haggblade, S. & Hazell, P.B.R. 1989. Agricultural technology and farm/non-farm growth
linkages. Agricultural Economics, 3: 345–364.
Haggblade, S., Hazell, P.B.R. & Brown, J. 1989. Farm/non-farm linkage in rural sub-
Saharan Africa. World Development, 17(8): 1 173–1 201.
Hazel, P.B.R. & Hojjati, B. 1995. Farm/non-farm growth linkages in Zambia.
Washington DC, IFPRI/EPTD. (Discussion paper No. 8.)
Hazell, P.B.R. & Röell, A. 1983. Rural growth linkages: household expenditure patterns
in Malaysia and Nigeria. Washington DC, IFPRI. (Research Report No. 41.)
ITTB. 1998. Statistical Bulletin, No. 1, February 1998. Mekelle, Ethiopia.
144 Rural farm/non-farm income linkages in northern Ethiopia

Lanjouw, J.O. & Lanjouw, P. 1995. Rural non-farm employment: a survey. Washington
DC, World Bank. (Policy Research Working Paper No.1 463.)
REST/NORAGRIC. 1995. Farming systems, resource management and household
coping strategies in northern Ethiopia: report of a social and agro-ecological baseline
study in central Tigray. Aas, Norway.
Rosenzweig, M.R. 1978. Rural wages, labour supply and land reform: a theoretical and
empirical analysis. American Economic Review, 68(5): 847–861.
Rosenzweig, M.R. 1984. Determinants of wage rate and labour supply behaviour in the
rural sector of a developing country. In H.P. Binswanger & M.R. Rosenzweig, eds.,
Contractual arrangements, employment and wages in rural labour markets in Asia,
p. 211–241. New Haven, Conn., USA, Yale University Press.
Rosenzweig, M.R. 1988. Labour markets in low-income countries. In H. Chenery and
T.N. Srinivasan, eds., Handbook of development economics, Vol. 1. Amsterdam,
Elsevier Science Publisher BV.
Woldehanna, T. 1998. Farm/non-farm income linkage and the working of the labour
market in Tigray, northern Ethiopia. (Paper for the 23rd Congress of the International
Association of Agricultural Economists, Sacramento, Calif., August 1997.)

144