Understanding Inconsistencies in Risk Attitude Elicitation Games: Evidence from Crop and Fish Farmers in Five African Countries

Abstract

Recent empirical studies eliciting farmers’ risk attitude by means of lab-in-the-field experiments have reported high degrees of inconsistency in farmers’ responses. We investigate inconsistencies in risk attitudes elicitation games using data from incentivized lotteries with 2,371 small-scale farmers from Eastern (Kenya, Uganda, Tanzania) and Northern Africa (Tunisia, Morocco). Our sample farmers exhibit high levels of inconsistent behavior, with over 48% committing some type of inconsistencies. Depending on the country, inconsistencies are explained by poverty (inattentive choices), gender, and/or the interaction of gender and level of education (wrong choice). We find no significant impact (negative or positive) of education alone. Furthermore, session fixed effects significantly explain inconsistencies in many instances, suggesting that session-specific circumstances, including the enumerators’ performance, play a significant role in properly explaining the experiment. Our findings suggest that using risk attitude parameters without considering the existence and the causes of inconsistency may lead to unreliable results and should be given serious consideration. Additionally, the results of this study may inform practitioners of the farmer segments more prone to errors.

Keywords: behavioral economics, Risk attitude, poverty, smallholder farmers, field experiment