Farmers’ learning preferences and risk management choices

Purpose: The relation between farmers’ learning preferences and risk management (RM) decision-making remains largely unexplored. We assess whether and how different learning preferences, notably learning by doing, from other farmers and through social media, influence farmers’ RM choices. Design/Methodology/Approach: Based on a survey of 320 farmers in Spain and the UK, we employed Poisson models with instrument variables to assess the effect of learning preferences on the level of diversification of RM portfolios, and Multivariate Probit Regressions to assess the effect on single strategies. Findings: We found that all learning preferences are significantly correlated to RM choice, with learning through social media and from peers leading to more diversified RM portfolios and learning by doing leading to less diversified ones. Moreover, the results show that each learning preference affects different specific RM strategies. Practical implications: Policies could leverage informal learning channels, especially learning through social media and from others that tend to more diversified RM portfolios. Yet, our results are limited in that other (formal) learning modalities that might play a relevant role were not accounted for in the analysis. Theoretical implications: The learning factor should be formalised and accounted for in theoretical frameworks modelling RM decision-making. Originality: The paper proposes theoretical advances to encompass the learning factor into RM decision-making frameworks and provides the first empirical evidence of the interaction between the three learning preferences and RM behaviour.