Impact

Heterogeneity: Within AGRICORE each farm will be modelled as an autonomous entity with capacity to interact with their environment and other agents, and to evolve according to its decisions and their context dynamics. This will provide a means to better account for heterogeneity among farms, to enable modelling at a geographic scale ranging from individual farm to regional and to global, and to reduce bias when averaging multiple farms.

Structural change: The configuration of the AGRICORE agent-based model will allow the explicit consideration of structural change adjustments as a consequence of agricultural policies. Specifically, the AGRICORE model will explicitly consider the transfer of production factors such as land and the impact of policies on variables like number of active farms, territorial distribution of farms and their evolution over time; farm size distribution and its evolution over time; land rental process; hired and family labour allocation; on-farm and off-farm labour allocation; and farm income disaggregated by farm type/size/regions.

Financial viability: The outputs of the AGRICORE agent-based model will include all major financial indicators of the farm business derived from the FADN database which accounts for growth, profitability, liquidity, leverage and activity. Moreover, the dynamic nature of the AGRICORE model will allow considering the impacts of policies on solvency, liquidity and other financial indicators which ultimately affect the structure of production of modelled farms and their survival rates.

Dynamic investment: The farmers’ simulation process implemented in AGRICORE will rely on a multiperiod optimisation problem in which the investment demand for land and capital goods (such as machinery, buildings, facilities and equipment) are to be explicitly modelled at various time scales – from the short to long term. This will allow analysing the impacts that policy measures have in the investment decisions of farmers in the long run.

Risk and uncertainty: The AGRICORE’s AI-based behavioural rationale module will explicitly integrate uncertainty by considering the mean, variance and skewness of likely future framework conditions (prices, yields and input demands). This will allow not only to model policy tools explicitly targeted to farmers’ risk management (such as income stabilisation schemes as a response to high losses in crop incomes) but also to effectively assess the impact on producers’ attitude toward risk.

Interaction between farms: AGRICORE will explicitly consider farm agents not only to interact indirectly by competing on factor and product markets but also directly by sharing information related to expected production outcomes and expected technology adoption. This will allow the model to simulate the interactions that take place when farmers are brought into a dynamic relationship through a set of reciprocal actions and which has effects on their future evolution.

Transfer of production factors: The structure of the AGRICORE agent-based model, along with the explicit handling of interconnecting variables, will enable to effectively consider the transfer of production factors between agents. Specifically, the transfer of land will be implemented through the dedicated Land module, and the transfer of other production factors will be implemented through the dedicated Markets module. Moreover, and given that land is essential to most agricultural and livestock production, AGRICORE will exploit today’s availability of computing power and GIS to implement realistic representations of land properties. This way, AGRICORE will generate a 2D mesh of the area of interest using satellite information with dynamically-sized cells attending to the extension of the area of interest and the available computing power.

Market inter-linkages: The AGRICORE approach to consider market inter-linkages is twofold. On the one hand, the AGRICORE Markets module will account not only for final products but it will also consider other market feedbacks such as those for labouring, water, manure, fodder and young animals. On the other hand, it will implement market equilibrium model dynamics linked to the aggregated feedback from the demand side, so prices that seem exogenous at farm scale become endogenous when simulating policy impacts at larger geographical scales such as at EU extent.