The reduction of adverse health and environmental effects from pesticide use is currently a top priority on the agricultural policy agenda. Efficient pesticide policies must take into account farmers’ application behavior, especially effects of pesticide use on economic risk. However, previous results regarding the direction of risk effects of pesticides are ambiguous. We show that the ambiguity in earlier studies could be due to the pesticide indicator selected. Indicators which fail to account for the heterogeneous properties of pesticides may be inapt for interpreting farmers’ pesticide use decisions. Our analysis, based on a rich panel dataset of Swiss wheat producers with highly detailed information on pesticide use, considers different pesticide indicators and multiple sources of uncertainty. Our key finding is that indicator choice affects the magnitude and sign of estimated risk effects. Estimates of pesticide productivity and risk effects are significantly higher for fungicides, and even reversed for herbicides when we measure pesticide use in simple quantity units (kilogram per hectare) ‐ compared to the quality and intensity corrected Load Index. This means for example, that farmers will ceteris paribus use lower quantities of herbicides, but will increase the overall toxicity of the products applied with increasing risk aversion. We discuss implications of our findings for the design of pesticide policies and agricultural risk management instruments.     «

The reduction of adverse health and environmental effects from pesticide use is currently a top priority on the agricultural policy agenda. Efficient pesticide policies must take into account farmers’ application behavior, especially effects of pesticide use on economic risk. However, previous results regarding the direction of risk effects of pesticides are ambiguous. We show that the ambiguity in earlier studies could be due to the pesticide indicator selected. Indicators which fail to account...     »