The economic progress Ethiopia has been scoring since the introduction of the Agricultural-led Industrialization Economy has taken the attention of the international media and organizations. In this regard, the World Bank analyzed the progress in order to identify the factors that contribute to the progress and poverty reduction. “The Progress driven by agricultural growth, investments in basic services and effective safety nets. The pace of poverty reduction in Ethiopia has been impressive, especially when compared with other African countries,” it states. In addition to this, the bank also analyzed the contributions of the non-farm enterprises that this article deals with.

In addition to being the primary sector of activity for 11-14 per cent of the population, a further 11 per cent of rural households earn about a quarter of their income from operating non-farm enterprises (NFEs) in the service sector. This income is earned largely during harvest months and months immediately following harvest. The income earned from these activities improves the Well being of households and its role in reducing poverty is considered to be significant.

Ascertaining the impact of these types of service sector activities on poverty reduction is Considerable and it is possible to provide more information on the amount of income these activities generate and for which types of households. These individuals often have a primary categorization in agriculture and the non-farm income they earn is highly correlated with agricultural income, causing growth analyses to attribute this impact to agricultural growth. Simply ascertaining whether households with NFEs are poorer or richer than other households also does not address this question. If households with NFEs are richer it could be that operating NFEs is a means by which some poor, uneducated households grow their incomes and escape poverty. On the other hand, it could simply be the case that these households are already better off and are able to invest in high-return NFE activities, and are thus more likely to operate them.

Additionally NFEs in Ethiopia are largely complementary to agriculture and driven by growth in this sector. The close dependence of NFE activity on agricultural income means that this is not a driver of poverty reduction on its own. An initial assessment of constraints to NFEs suggests that interventions to increase demand will have the largest impact on increasing the vibrancy of this sector and its role in reducing poverty. On the supply side, NFEs appear to depend on agricultural income for inputs and investment capital. On the demand side, they rely heavily on increased local demand during the harvest period to generate household income. The need for capital does not appear to be a major cause for the current seasonality of NFEs, but many do report access to market demand as a major constraint. Increasing demand will require further investments in infrastructure, increased employment the manufacturing sector on non-seasonal service sector activities, and increased agricultural revenues.

The analysis uses detailed data on the livelihoods of households in rural and small-town. Ethiopian Rural Socioeconomic Survey (ERSS) sample is representative of rural Ethiopia and towns less than 10,000 people. The data includes both those counting the service sector as a primary occupation the service sector in rural Ethiopia and small towns comprises 67 per cent self-employed activities. NFE ownership is defined as the operation of a non-farm enterprise involved in the provision of non-agricultural services such as carpentry, the processing and sale of agricultural by-products such as flour, trade, professional services, transportation services, and food services. A household was considered to operate a NFE in the survey if it reported to have operated one or more of these types of enterprises in the twelve months prior to the survey, including those ventures that had been shut down permanently or temporarily during that time.

In Ethiopia, NFE activity is primarily concentrated in processing and sale of agricultural products, trade of other products or offering a service from home or a shop. The most prevalent NFE type is the processing and sale of agricultural byproducts, which is strongly tied to agricultural activities. Twenty-eight per cent of NFE operating households operate this type. A further 28.3 per cent of NFE-owning households offer a service from home or a household-owned shop and 24.2 per cent trade in a market or on a street. Stark differences in the prevalence of non-agricultural NFEs that are operated from home or a shop exist between rural areas and small towns with over 40 per cent of households in small town areas reporting to operate a NFE of this kind.

One in five households in rural Ethiopia own an NFE. NFEs dominated economic activities in small towns with 55 per cent of small town households operating at least one NFE. On the basis of the ERSS, it can be estimated that there exist approximately 2.9 million NFEs in Ethiopia with 20.2 per cent of all households in rural and small town areas owning at least one NFE. NFEs are found to be mostly young with a mean age of approximately six years and a median age of two years. The difference in median and mean arises because of the high number of very young enterprises and the presence of a few older NFEs. A third of all NFEs were started in the year leading up to the survey. In the absence of clear evidence of high growth in the proportion of households operating an NFE, this suggests significant churn in the operation of NFEs.

There is some indication that it is the less educated households in small towns that operate NFEs, as opposed to more educated households in rural areas. In rural areas, NFE household heads have an average of 0.5 more years of education than households without a NFE suggesting that better educated households may be better equipped to choose to engage in NFE activities. Conversely, in small towns they have on average 3.3 fewer years of education than households without a NFE, pointing to higher education potentially providing better access to public sector and wage jobs. Households in small towns do have higher access to wage jobs with over 15 per cent of those seven years and older working in wage jobs, compared to less than 3 per cent of those in rural areas; most of the wage jobs are with the government or private enterprises. NFE participation is more prevalent among households with lower landholdings per head, which may indicate some households are pushed by necessity into NFE operation. In small towns, where households generally have very little land, those that do not operate a NFE own on average more than double the land assets of NFE-owners.

This gap narrows but becomes statistically significant for households in rural areas. In addition house holds that operate an NFE are less likely to own livestock, suggesting these are alternate sources of livelihood for households. Households without a NFE own more livestock than NFE households in both sub populations with a comparably sized gap.

There is no significant difference in the rates of NFE ownership between male and female-headed households in rural areas, nor in small towns once other characteristics of households such as education and consumption per capita have been taken into account. In small towns female-headed house- holds represent a greater proportion of households with NFEs at 38.3 per cent than without NFEs at 29.3 per cent.

One out of every six households (16 per cent) in the bottom 40 per cent operate an NFE, but rates of NFE ownership are higher among non-poor households: one out of every four households (24 per cent) in the top 40 per cent own an NFE. Prevalence is higher among higher quintiles with 26.3 per cent and 21.3 per cent of households in the fourth and fifth quintiles, respectively, operating a NFE.

Analysis using panel data also finds that higher consumption growth is positively correlated with a higher initial share of non-farm enterprise income. Households with NFEs in rural areas consume an average of 280 Birr more per annual adult equivalent than those without NFEs, which rely primarily on agriculture. The difference in consumption is significant when household size and education and age of the household head have been controlled for. It could be that operating NFEs is a means by which some poor, uneducated households grow their incomes and escape poverty. On the other hand, it could simply be the case that it is those households already better off, that are able to invest in high-return NFE activities, and are thus more likely to operate them.

There is no difference in the consumption level of households with and without an NFE in small towns. In small towns, households that operate NFEs, on average, consume approximately 250 Birr less per annual adult equivalent than households that do not. However, these differences are not significantly different.

Over half (54 per cent) of NFE operating households report that NFEs generate approximately a quarter of their income: these are households for whom service sector activities contribute significantly to household welfare but who do not report their primary sector of occupation as services. Households in small towns, however, report more often to be generating a share of around half or three quarters of total household income through the operation of a non-farm business, and 21.9 per cent indicate that it generates the household’s entire income. Calculating average annual incomes per NFE, we find a median annual NFE income of 700 Birr. Median annual NFE income in small towns is 1600 Birr, relative to a much lower value of 650 Birr in rural areas, indicating that most small town NFEs are generating more income than their rural counterparts.

However there are some NFEs that earn much higher levels of income and this is indicated in the fact that the mean income in rural areas is much higher than the mean income in small towns. Using the source of consumption data in the 2011 Household Consumption Expenditure survey suggests that nationally, 10 per cent of consumption is funded through non-agricultural household enterprises.

Households for whom NFE activities comprise more than half of their income will report this as a primary sector of occupation in national surveys. NFEs generate, on average, one sixth of the returns generated by a hectare of land used for agricultural production. The median NFE increases household income by 20 per cent and generates the income equal to about 0.16 to 0.18 hectares of land. In a number of settings NFE income allows households to become more resilient in the face of agricultural shock such as weather.

The association between NFE start-ups and the main agricultural period suggests that business activity was taken up in anticipation of or in response to highly active agricultural activities and heightened local demand. In rural areas, most NFEs list the top three months of activity as November, December and January and in small towns NFEs list December, January, and February as the most important months for activity. There thus appears to be a small, one month lag in peak NFE activity between small town and rural sub-populations. This lag may indicate a rural supply-chain trend.

On the supply-side, most households rely on agricultural income to fund the creation of NFEs. Overall, agricultural income is reported to be either the primary or secondary source of start-up capital for 64 per cent of NFEs. NFE households report the next important source of start-up capital to be non-farm self-employment income, noted as a primary or secondary source of funds by 18 per cent of households. This result can be explained by the fact that some households operate multiple NFEs and may thus use the income from one NFE to start another.

Rural NFEs tend to rely more heavily on agricultural income for start-up capital than small town NFEs, with 65.7 per cent of rural households citing agricultural income as a main source of funds for NFEs, as opposed to only 13.7 per cent of small town households. This result can be explained by the greater prevalence of non-farm activities in towns, and the stronger direct links with agriculture in rural areas.

Although there is a statistically significant difference in the proportion of households reporting September, October, and November as a high month for NFE activity, there is no significant difference in the overall trend throughout the year for farming and non-farming households. Despite the fact that non-farm households cannot rely on agricultural income to fund the operation of their NFEs, they still exhibit increased NFE activity from November to February. The customer base of most NFEs appears to primarily comprise the local market, local consumers or passers-by, and traders, indicative of the local nature of the markets they serve. Locals and passers-by constitute a somewhat higher share of the customer base in small towns than in rural areas, with 41.6 per cent and 29.5 per cent of NFEs reporting this as one of their two main customer bases, respectively. Additionally, selling to traders appears to be more common for rural NFEs, as 16.8 per cent of rural households reported traders as a main customer base, relative to 10.4 per cent of small town NFEs. Generally, NFEs appear to depend on agricultural income for inputs and investment capital. On the demand side, they rely heavily on increased local demand during the harvest period to generate household income. The need for capital does not appear to be a major cause for the current seasonality of NFEs, but many do report access to market demand as a major constraint.