Abstract

Farmers have to take decisions in an increasingly risky world. Climatic risk induces more frequent and deep negative shocks that farmers and systems will have to internalize. The production risk adds up to the increased exposure of farmers to price volatility following the market liberalization observed in many countries. Even though risk preferences are recognized to be important factors for understanding farmer’s decision-making, research on estimations of risk preferences has mainly concentrated on low-income countries. Moreover the state of current research does not enable to conclude on individual drivers of risk preferences because of the lack of robust estimations of heterogeneous effects.

I conduct an incentivized field experiment with farmers, free of learning-bias, to structurally derive parameters of risk preferences based on Cumulative Prospect Theory. The risk tasks follow the adaptation to the European context made by Bocquého et al. (2014) of the experiment developed by Tanaka et al. (2010). Farmers are found to be highly risk averse and to distort probabilities by overweighting small probability of desirable outcomes. However, contrarily to previous studies, subjects are not found to be loss averse in average. Moreover, I dig into heterogeneous effects and show that 15% of the sample completely deviate from the representative agent by being extremely loss-averse. Those individuals are relatively young and low educated farmers, having inherited a relatively small farm that they manage alone. The results of this piece of research have strong policy implications and induce the need of considering heterogeneity across and within sectors when it comes to assessing risk preferences.

Practical information