A Systematic Review of Key Spatial Econometric Models for Assessing Climate Change Impacts on Agriculture

Abstract

This paper explores the limitations of traditional econometric models, such as the Ricardian and profit-based approaches, in accurately quantifying the impacts of climate change on agriculture. While these models offer valuable insights, they often neglect spatial dependencies, heterogeneity, and spillover effects. We argue that spatial econometrics provides a more comprehensive and robust approach to analyzing climate change impacts. By explicitly incorporating spatial relationships between agricultural units, spatial econometric models capture the influence of factors such as proximity to markets, resource sharing, information diffusion, and spatial correlation of climate variables. We review pioneering studies employing spatial econometric models, including SAR, SEM, SLX, SARAR and SDM, which reveal significant discrepancies between spatial and non-spatial estimations. These studies demonstrate that neglecting spatial dependence can lead to biased estimates and inaccurate predictions of climate change impacts. Moreover, the incorporation of spatial effects often results in smaller marginal effects of climate variables, suggesting that traditional non-spatial models may overestimate negative consequences. This paper contributes to the ongoing research on climate change impacts on agriculture by highlighting the significance of spatial econometrics and emphasizing its potential to inform robust and effective adaptation strategies.