Analyses of the impact of European policies on agricultural change are most often based on agricultural sector models. Such models have their limitations: they cannot specify the interaction between agriculture and the rest of the economy, and their spatial dimension is usually limited. Land use simulation models, on the other hand, usually depend on other models for assessing the demand for land. The consistency of those models with the assumptions and databases of the land use model is often not examined. This article reports on a research project where the links between a macroeconomic model, an agricultural sector model and a land use model were explicitly explored in order to arrive at a consistent model chain. This integrated framework was put to the test by applying it to two contrasting scenarios, which compare impact on agricultural incomes, land use and land management.