Household livelihoods in semi-arid regions: options and constraints

Abstract

The overall aim of this study was to explore what the development community can do, or facilitate, to significantly improve livelihoods in semi-arid systems. We based our analysis on two case-study sites in the communal lands of southern Zimbabwe, areas fairly typical of a vast domain of granite/gneissic landscapes in semi-arid areas of numerous developing countries on three continents. Our main tool was a detailed livelihood questionnaire, supplemented by participatory appraisal and observation, action research, biophysical analysis and systems modelling. Human, financial, physical and natural assets are severely constrained for most households. An analysis of rules, leadership and committee operations indicates a system dominated by informal arrangements, and the ineffectiveness of a number of institutions. Most households rely on cash and subsistence income from a number of sources - dryland crop production, gardening, livestock production, woodland activities, wage or home industries and remittances/gifts. Marked wealth differentiation occurs, with local people recognising the different wealth groupings largely on the basis of various capital assets. One factor driving differentiation is whether a household has access to remittance income. The wealthiest quartile is the only one in which households derive more than a third of their gross cash income from crops, but their main cash income is remittances. Elements of change can be identified in numerous aspects of the capital assets and the livelihood strategies. We suggest that there are some key drivers of change, namely: (a) rainfall, (b) macro-economic changes, (c) changing institutional arrangements and social processes, and (d) demographic processes and HIV/AIDS. The study suggests that households have a rich and varied livelihood portfolio, with displays of infinite resourcefulness to make ends meet. We found little evidence of a downward spiral triggered by rapid population growth, but we do not see the poverty status of rural households improving in semi-arid regions, as part of intensification processes. Most households fall below various internationally recognised poverty lines. Rural poverty is the result of a suite of interacting social, economic and environmental factors and processes operating at a range of scales. The multi-faceted nature of poverty indicates that there is no silver bullet to rural development, and that an integrated, multi-sectoral approach to development is critical, with different, but complementary, activities pursued at different levels. We recognise that there are very low incentives to actively manage natural resources in the commons, and there are almost no technical or institutional interventions that could be used to change this situation. Removing ineffective national- and district-level regulations related to woodlands and grazing areas, and empowering local leaders, will be a step in the right direction. In pursuing poverty reduction, of paramount importance is the development of an enabling policy environment to allow farmers to experiment and capture opportunities. Our overall conclusion is that there are very few options for significantly improving livelihoods in semi-arid regions and that the poverty alleviation targets set by the international community are overly ambitious. Our analyses suggest that rainfall variation and the state of the macro-economy are likely to have a greater impact on livelihood status than local rural development interventions.