Do Changing Probabilities or Payoffs in Lottery-Choice Experiments Matter? Evidence from Rural Uganda

This study compares risk attitudes of smallholder farmers elicited from two different lottery designs (i) with fixed payoffs and changing probabilities and (ii) with fixed probabilities and changing payoffs. We utilize a combination of experimental and household survey data collected from 332 randomly selected smallholder coffee farmers in Uganda. Both methods reveal high proportions of farmers who are classified as risk averse. However, comparing the different risk categories shows that the two elicitation methods yield significantly different results. Furthermore, we relatively find low inconsistency rates in the response behavior for the two methods compared to other studies in the past. Specific socio-demographic and socioeconomic characteristics also affect farmers’ risk attitudes.