Evaluating Different Growth Scenarios for Organic Farming Using Bayesian Techniques

Different views exist on the future development of organic agriculture. The Dutch government believes that in 2010 10% of the farm land will be used for organic farming. Others have a more radical view: due to increasing emphasis on sustainable production in the end all farming will be organic. Others believe in a more pessimistic scenario in which the recent growth in organic was just a temporary upswing and that the share of organic farmers already reached its maximum. In this paper different potential scenarios for the further growth of organic farming are evaluated using Bayesian techniques. A nonlinear logistic growth model explaining the share of organic farms is estimated using available historical data for Dutch agriculture. Various scenarios imply different prior values for the parameters. Because of the non-linear model specification a Metropolis-Hastings algorithm is used to simulate the posterior densities of the model parameters. Finally, using Bayesian model comparison techniques probabilities can be attached to the different scenarios. The proposed methodology is a promising tool for analysing technology diffusion in general when different scenarios for diffusion are possible and limited data is available.