Please note! This essay has been submitted by a student.
The terms “non-farm” and “off-farm” routinely appear seemingly in synonymous way. A majority of the literature used the terms “off-farm” and “non-farm” interchangeably. But the term “non-farm” should not be confused with “off-farm”. Off-farm refers to activities under taken off the owner’s own farm and it includes wage employment in agriculture earned on other peoples’ farms. But rural non-farm economy contains all rural economic activities outside agriculture (nonagricultural) and also it includes small and large scale activities of widely varying technological sophistication (Ellis 1998, 2000, Barret et al., 2001).
Rural non-farm activities can be classified as productive and non-productive non-farm activities (Lanjouw & Shariff, 2002); self employment and wage employment non-farm activities (Ellis, 2000; Barret et al., 2001; Beyene, 2008); and manufacturing, trading and services non-farm activities (Loening & Imiru, 2009). These activities are also characterized in terms of size, growth, composition and equity impact (Haggblade et al., 2002). Thus, when we say rural non-farm activities, it includes heterogonous collection of manufacturing, commerce, services provision and both formal and informal wage employment activities.
Non-farm activities play an increasingly important role in sustainable development and poverty reduction (Lanjouw, 1999; Gordon & Craig, 2001; Davis, 2003; Atamanov, 2011). Also, non-farm activities provide work in the slack periods of agriculture (Lanjouw & Lanjouw, 1995); contribute 35-50 percent of income for developing world (Haggblade et al., 2009); function as safety net through diversifying income source (Zhu & Luo, 2006); solve rural urban migration by providing seasonal or alternative employment for those left out of agriculture and lowers price to the poor (Lanjouw, J & Lanjouw,P 1995; Atamanov, 2011; Pal & Biswas, 2011); add value to the farm activities (through processing, trading and storing) and provide opportunities as to learn new skills, make new contact or gain entry to new markets (DFID, 2002).
Some studies characterized the drive for livelihood diversification into two: distress-push and demand-pull situation. Pull-factors include better returns in non-farm relative to farm sector while push-factors are inadequate farm output (Gordon & Craig, 2001; Atamanov, 2011). These characterizations of drive for diversification towards non-farm activities are undoubtedly an oversimplification, but, it is a useful reminder that participation in rural non-farm activities may be driven by quite different circumstances and have different outcomes.
Decomposition by income source is offered by Shorrock (1982, 1983) which was subsequently extended by Morduch and Sicular (2002) and Fields (2003) to regression based decomposition by factors of income. But this regression based decomposition dates back to Oaxaca (1973). It overcomes many of the limitations of standard decomposition by subgroups, because it is built on techniques used by inequality factor decomposition. Since Gini decomposition do not tell us the impact of uniform increase in any income source including equally distributed income components on inequality, regression based decomposition method complements the results (Morduch and Sicular, 2002). That is, Gini decomposition method fails to satisfy the property of uniform addition. In addition, it does not identify the factors of inequality. For example Gini decomposition cannot describe how household level variables such as age, family size, education, land size and the like affect income inequality. But the new regression based decomposition supplements the method by answering the question of how much a given determinant of income contributes to income inequality (Morduch and Sicular, 2002).
In Ethiopia there are limited studies on income distribution. Among the few findings Tassew (2002) found that non-farm income has an unequalizing effect on income distribution in northern Ethiopia due to entry barriers for the poor. According to the findings of Bigsten and Shimeles (2006) in Ethiopia household characteristics such as occupation of the head of the household, educational level of the head of the household and other unobserved characteristics were some of significant factors that played a role in determining the Gini coefficient for urban area. Rural areas with relatively high average land size tend to have lower consumption inequality, though higher land inequality translates directly into higher consumption inequality. Access to education plays an important role in driving the Gini coefficient upwards in rural areas. Villages with high concentration of educated family heads tend to be associated with high level of the Gini coefficient, which partly may explain higher degree of differentiation in earning potential as well as consumption preferences.