Modelling farm structural change for integrated ex-ante assessment: review of methods and determinants

摘要

This paper provides a literature review of methods and determinants relevant for modelling farm structural change within an integrated modelling chain. Environmental and economic impacts at farm level and individual farm responses to agricultural and agri-environmental policies strongly depend on characteristics like farm size, specialisation, and production intensity. Consequently, up-scaling results of corresponding farm type models in ex-ante assessment exercises requires comprehensive and valid predictions of the farm types' future relevance under different scenarios. The paper reviews methods relevant to forecasting farm numbers in classes defined by farm typologies with the objective to identify (1) a preferable modelling approach and (2) empirically relevant determinants. Despite the literature's considerable size, even recent studies are rather limited in scope and typically restricted to a subset of farm types and one or very few regions. With regard to data availability, computational complexity and statistical validation procedures, Markov chain models are identified as the only generally suitable method for a broadly scoped modelling approach across European regions and a differentiated farm typology. However, other research on determinants of farm growth, the number of farm holders, farm succession as well as new multi-agent based simulation approaches hint at relevant explanatory variables previously not considered in Markov chain analyses. Their impact seems testable in more ambitious cross-regional and cross-farm type setups.