Page 107 - Climate Change and Food Systems

process the outputs of two GCM models from the fourth IPCC assessment into the IIASA-FAO AEZ crop simulation models, which simulate impacts on land resource use and potential yields. These outputs are then used as inputs into the BLS model. Three scenarios are explored: a baseline reference point with no climate change; SRES A2, which describes unmitigated climate change; and SRES B1, which is a proxy for mitigated climate change. The recursive dynamic BLS CGE model is then solved in annual steps for 1990-2080. In the baseline, the number of people at risk of hunger
is projected to decrease globally to 555 million people. Using these modelling tools, the authors conclude that climate change will reduce but not eliminate that progress.
One challenge for CGE models relates to intra-seasonal climate variability. Because CGE models use annual time steps, they typically lack the ability to account for intra-seasonal variability and short-term climate extremes, such as heat spells, which crop yield models have shown to be key determinants of seasonal yields. Block et al. (2006) demonstrate the importance of brief weather extremes in a CGE-based analysis of climate change impacts on Ethiopia that contrasts outcomes with and without accounting for the incidence of flooding. To describe threshold
effects in precipitation, the authors develop annual climate-yield factors (CYF) by crop and agricultural zone within Ethiopia, based on crop sensitivity to water shortages at four stages of growth, choosing the most water-sensitive stage to calculate the CYF. Just as water shortages can reduce yields, so can floods. A flood factor is therefore added
to the CYF, decreasing the CYF if the year is significantly wet during the flowering or harvesting stages. The flood factor is also applied to non- agricultural sectors, to represent higher costs of transportation and other damage due to flooding. Their recursive dynamic, single-country CGE model describes the impacts during 2003-2015 of an ensemble of weather experienced during nine historical 12-year time periods. Both drought and flooding reduce productivity and GDP and increase poverty. These effects are more negative when
flooding is accounted for. Without the flood factor, the mean increase in poverty rises from 42 percent to 43 percent between 2003 and 2015; with it,
the mean projected increase in poverty reaches 54.7 percent.
Another criticism of CGE models is their tendency to generate relatively smaller changes
in price effects and welfare even under large crop yield effects of climate change. This is because of the structural nature of the models and their neo- classical perfect market integration assumption in which inputs (capital and labour) are fully utilized and mobile across sectors and activities. Such an assumption lacks realism in light of market rigidities so characteristic of most economies, especially
in developing countries. In addition, the functional forms used to describe factor demand and trade also tend to diminish quantity responses to price changes. To address these limitations, and given much of the uncertainty around climate impact analysis, CGS models can make more use of a well-developed tradition of sensitivity analysis and the use of the probability distributions of outcomes in climate change impact studies. This also underscores the importance of efforts to account more comprehensively for the multiple pathways through which climate change may affect an economy’s productivity or factor supplies.
■ Climate and trade
One of the advantages of CGE modelling is its contribution to trade and trade policy, and several CGE analyses on climate impact also reported on the implications of trade as a potential adaptation mechanism. Ahmed et al. (2011b) use a static, single-country CGE model of Tanzania to analyse alternative trade regimes in the presence of climate change, finding that open trade policies help to negate the poverty impacts associated with climate shocks. Valenzuela and Anderson (2011) also address the adaptive role of trade in
a study that finds that climate change will cause a substantial 12 percent decline in the food self- sufficiency ratio of developing countries by 2050. They use a static, global GGE model to analyse the world economy in 2050 under two scenarios:
chapter 3: economic modelling of climate impacts and adaptation in agriculture: a survey of methods, results and gaps
87