Incorporating uncertainty in a deterministic agricultural sector model

Abstract

Climate-induced uncertainty in crop yields is introduced in the Norwegian sector model Jordmod. The model is comprised of a supply module in which profits for more than 300 regional farms are maximized and a market module which maximizes social welfare in the agricultural sector. In the supply module, farmers determine their plant decisions and crop input levels (N-fertilizer) subject to a discrete number of weather outcomes affecting crop yields. After that, a specific weather distribution is chosen determining crop yields. The resulting input-output mix at farm level makes up the supply side of the commodity markets which together with linear demand functions determine equilibrium levels. The procedure is repeated for each discrete weather outcome. Note that plant decisions and crop input levels remain the same for all weather outcomes as farmers face the same uncertainty during all repetitions, but crop yield will vary. Hence, equilibrium prices and quantities will vary as well allowing their representation as stochastic distributions. In a preliminary empirical application, the stochastic results are contrasted with the deterministic results based on the mean values of the weather outcomes. This comparison will shed light on the potential error made by neglecting uncertainty at the farm level.