Adaptation to human-induced climate change is currently receiving a lot of attention in international development circles. But throughout human existence, natural resource-dependent people have exploited and coped with the effects of climate variability on the ecosystems from which they derive a living. Learning from this experience can help inform the design of appropriate policies for responding to human-induced climate change. This paper presents the results of a World Bank study which sought to better understand the role of local institutions in supporting adaptation to climate variability and change in Ethiopia, Mali and Yemen. The study raised three questions. First, what strategies have been adopted by rural households in the past to adapt to climate variability? Second, to what extent do institutions of various sorts assist households in adopting adaptation strategies? And third, what are the factors that prevent households from adopting appropriate adaptation strategies? For the purposes of this paper, institutions are defined as structured, formal or informal organizations. The study followed a three-step approach. First, drawing on original data from field surveys, focus group discussions and institutional stakeholder interviews, household vulnerability to climate variability was characterized in terms of its three constituent elements: exposure to climate-related shocks and stresses, and sensitivity and adaptive capacity in the face of such stressors. Sensitivity refers to the degree to which people are affected by climate variability and change. High levels of exposure and sensitivity and low levels of adaptive capacity generally result in high levels of vulnerability. But a high level of exposure need not necessarily result in a high level of vulnerability if the household’s adaptive capacity is also high. Using data gathered on each of the constituent elements of vulnerability, cluster analysis was conducted to identify types of rural households sharing similar vulnerability profiles. Second, the reasons for differences in households’ choice of adaptation strategies were analysed. And third, the role of institutional assistance in adaptation was investigated. Virtually all households in all three countries engaged in at least some forms of adaptation, such as improved seeds or adjustment in planting dates, to render crop yields less susceptible to climate variability. But the particular adaptation strategies pursued differed markedly among households, both within the study sites in each country and among the countries studied. In Yemen, fewer adaptation strategies were pursued than in either Ethiopia or Mali. Wealthier households adopted capital intensive strategies such as the use of irrigation pumps. In Ethiopia, wealthier households engaged in more communal strategies, such as control of soil erosion, communal irrigation or reforestation, for which external assistance is needed. Pastoralists also reduced exposure through the adoption of collective water harvesting and rangeland management strategies, which cannot be undertaken by individual households on their own. In Mali, the only communal strategy chosen was irrigation, adopted by only a few, wealthier households. Income diversification beyond agriculture alone was also pursued by only a few households. Finally, migration as an income-diversification strategy was practised more frequently in Mali than in Ethiopia, and very little in Yemen. In all three countries, wealthier households tended over time to migrate to urban areas. Significant differences were observed in the extent and forms of institutional assistance provided to households in dealing with climate variability and change. The network of public extension agencies is particularly well developed in Ethiopia, where most rural households received such assistance, notably in terms of training. In Mali, on the other hand, public extension agencies are virtually absent and their assistance is focused narrowly on providing inputs for irrigation and home-garden agriculture. The role of NGOs, cooperatives, micro-finance institutions and religious groups was limited in all three countries. In Yemen, households hardly received any assistance from outside institutions, whether from formal government agencies or from NGOs. In all three countries, wealthier households enjoyed significantly more access to outside assistance. In all three countries, people living in rural areas are both exposed and sensitive to climate variability. Various strategies are pursued for dealing with this, and wealth is an important explanatory variable in adoption decisions. However, differences in coverage of institutional assistance and in the types of assistance provided were substantial, thereby affecting prospects for future development independently of the level of household wealth
