UNU-MERIT Working Paper Series, #2011-011, Maastricht, Atamanov, Aziz (2011)

The participation of the working population in non-farm employment in rural areas of Kyrgyz Republic is unambiguously governed by income distribution and differential access to assets and infrastructure.

The discourse on livelihood diversification in recent times has enquired into the causal factors or determinants behind different livelihood strategies, in particular a shift from agricultural to non-farm employment. What set of factors play a crucial role in facilitating or preventing such a transition in employment status have been a subject of keen interest. From the perspective of rural poverty reduction, the growth of non-farm employment is an important phenomenon, particularly in countries where more than half of the force is employed in the agricultural sector with low returns.

The post-Soviet economies in Central Asia are predominantly agricultural, unlike some of the other countries which were formed after the disintegration of the Soviet Union. In the Kyrgyz Republic, more than 60 percent of the population lived in rural areas, dependent primarily on agriculture. Although the unemployment rate in rural areas (6.4%) was much lesser than that in urban areas (11.1%), the poverty rate in rural areas is much higher at 36.8%. The poverty rate in urban areas was 8 percentage point lower than the rural rate. This implied that the extent of underemployment in agriculture was much more widespread.

Market imperfections with respect to inputs like land and credit have remained as major impediments to raising productivity in agriculture. In this context, rural non-farm employment (RNFE) is considered by policy-makers and economists as critical for reducing underemployment, diversify rural incomes and enhance the standard of living of the rural population. Atamanov, in this paper, attempts to address the determinant factors and barriers to the generation of RNFE in the Kyrgyz Republic through a quantitative analysis.

The author uses two representative and detailed household budget surveys conducted by the National Statistical Committee in 2005 and 2006, each survey having around 1800 observations. These surveys provide quarterly data on employment including time spent, sector of occupation and type of employment at the household level along with other household details ranging from demographic and locational characteristics to ownership of assets. This renders the data useful for a quantitative analysis of the determinant of employment diversification and more specifically the entry of household members into non-farm employment. The author employs a regression analysis for undertaking such an investigation.

The model for enquiring into the determinants of household participation in RNFE include a range of variables that can be broadly categorized as individual characteristics, household characteristics, household assets, location characteristics and access to infrastructure.

A descriptive enquiry into the data reveals that the share of RNFE in total employment was a significant 37.8% and 39.6% in 2005 and 2006 respectively. In terms of time spent also, the share of RNFE was nearly 50% in 2006. Within primary employment, the structure of RNFE shows that more than 60% employment in generated from commercial non-farm enterprises; the remaining can be attributed to non-commercial services like education and health, largely in the form of public employment. Within commercial employment, which mostly engaged male workforce, the important contributing sectors are trade, processing and transport.

The analysis in the study indicates the important role played by ‘push’ factors behind the participation of households in non-farm activities. A distribution of the population into five income quintiles exposes the linkage between poor asset ownership, including a meagre land-base, and engagement in RNFE. Households in the top three quintiles with higher agricultural incomes had a much lower share of their income coming from non-farm activities. Ownership of cattle herds also acted as a disincentive for household to diversify their income sources. In contrast, in the two years of survey, households in the bottom two quintiles (particularly the bottom 20% population) participated more in RNFE and more than half of their income was generated from employment in non-farm enterprises.

In comparison, occasional ‘pull’ factors were also located by the study. The presence of a resort in an area generated better paying non-farm activities towards which the workforce was attracted. Moreover, higher educational levels acted as a facilitator for household members to participate and diversify to non-agricultural activities. Additionally, the access to better infrastructure and markets not only enable the workforce to engage in non-farm activities but also enhance the income that can be earned from such activities.

Finally, the study points out to a paradox where ownership of adequate agricultural and livestock assets and high farm income act a cushion which creates a disincentive to diversifying occupation. However, given that a household is participating in RNFE, better asset ownership, including larger cattle herds and consequent livestock income, actually provided investible funds for generating high productive non-farm activities  or accessing better paying jobs in the non-farm sector.

The paper concludes on the note that better education, especially for women and increased access to infrastructure can allow the non-farms sector to play a more meaningful role in reducing rural poverty in the Kyrgyz Republic.