Modeling Land Use with Multi-Agent Systems Perspectives for the Analysis of Agricultural Policies

This paper addresses the adequacy and perspectives of multi-agent system (MAS) models in the context of policy support for agricultural policy makers. The paper starts illustrating what MAS are and how they may be used. The general presentation is followed by the description of an exemplary spatial dynamic model of a large number of interacting farms (agents). This model is simulated for different policy scenarios and farm sizes. The results are analyzed with regard to the farm structure, land use, efficiency, and farmers' incomes. It is concluded that policy analysis should not only address problems of price distortions. If agricultural policies affect structural change in the way the simulations show, the dynamic efficiency and income implications may be much more severe. In particular, well meant subsidies may in the long run create a dependency on further support.