Non-farm income is clearly important to farm households in developing countries, yet no clear pattern has emerged to help trace its effect on the distribution of rural incomes as a whole. Sample surveys of mixed arable and livestock farms in Chiweshe and Gokwe, in Zimbabwe's Communal Lands, were used to examine the distribution of income in each area. Chiweshe has more fertile land, higher rainfall and closer access to Harare. Gokwe is more remote and is in an area of relatively poor soil and low rainfall.

Main income sources in both cases are farm production of crops and animals (both for subsistence and for sale) and non-farm income, including remittances. A survey of household and individual incomes broken down by source for the two Communal Lands revealed that:

• distribution of farm income to households was far more equal in Gokwe than Chiweshe
• income from animals was more unequal than that from crops in both areas
• both animal and crop income were less equal in Chiweshe than in Gokwe

Equal Opportunity? Data on the distribution of household incomes in the form of Lorenz curves suggest that farmers who live closest to a town not only find it easier to get other employment but also enjoy greater access to markets for their agricultural products. The picture is not consistent with a scenario in which farmers with poor land look elsewhere to bolster incomes. In the Zimbabwe Communal Lands, Gokwe is remote from Harare and has poor soil and low rainfall. The Lorenz curve for farm income is closer to the line of total equality than the curve for total income. Here including non-farm income results in a less equal distribution. In the case of Chiweshe, on the other hand, the curve that includes non-farm income is closer to total equality than that for farm income only. This indicates greater equality when non-farm income is included and suggests this source of income is available to a greater proportion of this community.

When non-farm and remittance income were included, the degree of inequality in the distribution of total household income increased in Gokwe (see charts, above). This can be accounted for partly by a highly unequal distribution of remittances in this region, although only a few people from this region leave their families to work outside the household. The opposite effect was found in Chiweshe. Here, including non-farm income and remittances in total household income results in a more equal distribution of income. Significant lessons for policymakers are that:

• Orthodox assumptions that non-farm income alleviates rural inequality are overly simplistic.
• In more remote areas with traditional agriculture, those who have higher farm incomes also tend to be better placed to exploit non-farm income opportunities.
• Where there is better infrastructure and closer access to urban markets, commercialisation of agriculture can mean that there are more opportunities for non-farm employment. Thus, households wishing to supplement income from the farm can do so, resulting in lower levels of inequality overall.

In more general terms, spurring entrepreneurs on to provide enough skilled and unskilled non-farm work opportunities remains a crucial policy goal, although for the more remote regions, this requires the necessary infrastructure to enable access to these. Collecting more micro-level income data and analysing it in ways that separate out the component sources of income will make it easier for policymakers to identify the causes of inequality.

Contributor(s): Jenifer Piesse, Colin Thirtle

Further information:
Jenifer Piesse
Birkbeck College
University of London
Malet St
London
WC1E 7HX
UK

Tel: +44 (0)171 631 6760
Fax: +44 (0)181 995 3437
Email: j.piesse@bbk.ac.uk
Birkbeck College, University of London

Colin Thirtle
University of Reading
Reading
RG6 6AR
UK

Tel: +44 (0)118 931 8965
Fax: +44 (0)118 975 6467
Email: C.Thirtle@reading.ac.uk

Other related links:
Search Eldis for sources on Agriculture