Home 9 Horizon projects 9 MIND STEP’s project: analytical tools and models for policy support

MIND STEP’s project: analytical tools and models for policy support

Jun 8, 2023 | Horizon projects

Author: Marc Müller (Wageningen University) 

The overall objective of the Horizon 2020 project MIND STEP is to support public decision-making in agricultural, rural, environmental and climate policies by taking into account the behaviour of individual decision-making units in agriculture and the rural society. To this end, MIND STEP develops a highly modular and customisable suite of Individual Decision Making (IDM) models that focus on the behaviour of individual agents in the agricultural sector and establishes linkages between the new IDM models and the sectoral models currently used by the European Commission. This includes improvements of the micro-economic foundations of the sector models and ensuring the overall consistency of the modelling toolbox.  

A major achievement of MIND STEP was the combined use of the entire model toolbox for policy scenarios regarding the reduction of GHG emissions and the use of nitrogen inputs at farm and sector level. Future scenarios could focus, for example, on modelling eco-schemes of the Common Agricultural Policy (CAP) to help the Directorate-General for Agriculture and Rural Development (DG AGRI) and Member States assess whether they can achieve the specific objectives and targets of their Strategic Plans

A crucial aspect of the development of new farm-level models in MIND STEP was the integration of empirical studies and the FarmDyn simulation model. The behavioural work covered farmers’ propensity to adopt certain technologies and management options to reduce GHG emissions, as well as their attitude towards risk. The former was based on a survey of Dutch farmers to understand the impact of cognitive and socio-demographic factors and their impact on adaption behavior, accompanied by a literature review of currently available mitigation options. These findings were combined with farm-level statistics and agronomic information on the costs of mitigation options to stratify farms in the Dutch Farm Accountancy Data Network (FADN) and parameterise the FarmDyn model for ex-ante assessment of policy measures to encourage farmers to adopt mitigating technologies and management practices. On risk attitudes, a new risk behaviour module for FarmDyn was developed and parametrised based on findings from farm surveys in Germany and Italy. 

In terms of improvements to the micro-economic foundations of the sector models, it was possible to split the standing dairy herds from the capital account in the economy-wide MAGNET model and to generate marginal abatement cost curves using EU-wide FADN data and the FarmDyn model. In addition, the cost structure of the newly introduced mitigation options was fed into the GLOBIOM model

All these model integration activities were guided by the principle to keep the newly developed model features as flexible and re-usable as possible. To this end, a conceptual structure for modular model development was put forward in MIND STEP deliverable 3.1 entitled “Specification of model requirements: Protocols for code and data”. This report elaborates on the principles of modular software development and the implications of this concept for the case of combining farm-level simulations and empirical models. The general idea was to define a farm programming model as the core model, to which new features could be attached in a modular manner. This requires the ‘identification of interfaces, the provision of default data, and the adherence to certain standards regarding the shape of input data and the structure of contributed model code. To facilitate the collaboration between the involved researchers, guidelines for good coding practices, version control systems, testing procedures, and similar practices for quality management have been discussed in this deliverable and made available to the developer’ teams.