Modeling Path Dependence in Agricultural Adaptation to Climate Variability and Change

Abstract

Path dependence of farmers’ technical choices for managing climate risk combined with farmers’ difficulties in discerning climate change from natural variability might hamper adaptation to climate change. We examine the effects of climate variability and change on corn yields in the Southeast United States using a regional climate model nested within a global climate model (GCM) simulation of the equilibrium atmospheric CO2 concentration of 540 ppm. In addition to a climate scenario with normal variance, we modify the GCM outputs to simulate a scenario with a highly variable climate. We find that climate variability poses a serious challenge to the abilities of farmers and their supporting institutions to adapt. Consistently lower corn yields, especially in the scenario with a highly variable climate, illustrate that farmers’ abilities to make informed choices about their cropping decisions can be constrained by their inabilities to exit from their current technological regimes or path dependence. We also incorporate farmers’ responses to climate change using three adaptation scenarios: no adaptation, “perfect knowledge,” and a scenario that mimics diffusion of knowledge across the landscape. Regardless of adaptation scenario and variance structure, the most common result is a decline in corn production to the point where yield reductions of 1 percent to 20 percent occur across 60 percent to 80 percent of the region. The advantage of the perfect knowledge adaptation scenario declines through time compared to the diffusion-process adaptation scenario. We posit that the cost of path dependence to farmers, in the form of yield reductions, is likely unavoidable because the inherent variability of the climate system will result in adaptation choices that will be suboptimal for some years.