Climate variability, social capital and food security in Sub-Saharan Africa: household level assessment of potential impacts and adaptation options
Climate variability and poor distribution of rainfall often causes serious agricultural production losses and worsens food insecurity. Given that the direct effects of climate change and variability are transmitted through the agricultural sector, improving farm households capacities to adapt to the adverse effects of climate-related shocks is an important policy concern. This thesis applied a stochastic Agent-based Model (ABM) that is capable of simulating the effects of different adaptation options by capturing the dynamic changes of climate and prices, as well as the dynamic adaptive process of different farm households to the impacts of these changes. The agent-based simulations conducted in this thesis address the special challenges of climate and price variability in the context of small-scale and subsistence agriculture by capturing non-separable production and consumption decisions, as well as the role of livestock for consumption smoothing. To ensure the reliability and usefulness of results, the model was validated with reference to land-use and overall poverty levels based on observed survey values. In particular, the study used disaggregated socio-economic, price, climate and crop yield data to quantify the impacts of climate and price variability on food security and poverty at the household level. Furthermore, the study explicitly captured crop-livestock interactions and the "recursive" nature of livestock keeping when examining the effects of climate and price variability. The thesis additionally examined how specific adaptation strategies and policy interventions, especially those related to the promotion of credit, improved seed varieties, fertilizer subsidy and off-farm employment, affect the distribution of household food security and poverty outcomes. In addition to impacts on household food security and poverty, the study further considered indirect impacts through changes in the price of agricultural inputs and livestock holding. In terms of coping strategies, the simulation results in this thesis show that the effects of climate and price variability on consumption are considerable, but smaller for those households with relatively large livestock endowments. In addition, the study also found that farm households with a large plantation area of eucalyptus were able to cope with the effects of variability. Therefore, our results suggest that self-coping strategies are important but not sufficient and should be complemented with appropriate policy interventions. In terms of policy interventions, the study found that policy intervention through the expansion of credit and fertilizer subsidy along with innovation through the promotion of new crop varieties that are resilient and adapted to local conditions are the most effective adaptation options for the case of Ethiopia. In addition, the simulation results underscore that adaptation strategies composed of a portfolio of actions (such as credit and fertilizer subsidy along with new technologies) are more effective compared to a single policy intervention. For Ghana, the study suggests that if expansion of production credit is complimented by irrigation, it can provide a way to achieve food security under climate and price variability. In order to design a best-fit intervention instead of a 'one size fits all' approach, it is important to capture the distribution of effects across locations as well as households. The great strength of this study is its agent-based nature, which enables exploration of how effects are distributed across farm households. The simulation results clearly show that poor farms are vulnerable to climate and price variability, under which they suffer food insecurity, while a small group of wealthy farms are better off due to higher prices achieved when selling crops. The result from this thesis further underscores the need for improving adaptive capacity, as a large proportion of farm households are unable to shield themselves against the impacts of price and climate variability. In what follows, the study further applied standard micro-econometric techniques to examine the role of social capital and informal social networks on consumption insurance and adoption of risk mitigating land management practices. In particular, the thesis provides evidence of the effects of different dimensions of social capital on the adoption of soil and water conservation practices across households holding different levels of risk-aversion. The results of the study underscore that social capital plays a significant role in enhancing the adoption of improved farmland management practices and suggests that the effect of social capital across households with heterogeneous risk taking behaviour is different. Finally, by combining household panel data, weather data, self-reported health shocks and detailed social capital information, the last section is able to analyze how social capital buffers some of the implications of weather shocks.
