Approaches to evaluating the impact of climate change on food security

Quantifying the impacts of climate change on food security is an important issue on the agenda of many governments and non-governmental organisations. Precise magnitude and direction of the impact of climate change on food security can be attained through careful review and selection of analytical approach. This chapter reviews methodological and empirical approaches that can be used to evaluate the average effects and causality between climatic factors and food security. Four broad categories of methods are discussed. These methods include simulation models, global economic models, parametric linear and non-linear models (without-causality) and impact evaluation methods (with-causality) for assessing impact of climate-smart interventions and programs on food security. The simulation models consist of crop and climate simulation models which utilise production and climate data such as temperature and rainfall into computer programs. The commonly used economic models are the Computable General Equilibrium (CGE) and Partial Equilibrium (PE). Linear and non-linear models include multiple linear regression, Tobit regression, Binary regression and multinomial logit/probit regression models. These models do not establish causality but explain how climatic variables correlate with food security. The impact evaluation methods that can be used to measure the impact of climate-smart interventions and programs on food security include randomised control trial (RCT), and non-experimental methods including propensity score matching, double-difference, instrumental variable, regression discontinuity and endogenous treatment and control function methods.