Effects of communication, group selection, and social learning on risk and ambiguity attitudes: Experimental evidence from Bangladesh

In this paper, we conduct lab-in-field experiments using a sample of 206 farmers in Bangladesh and perform four main tasks related to measuring the farmers’ attitudes toward risk and ambiguity in both gain and loss domains: (i) measure the effect of communication and group members’ characteristics on risk and ambiguity attitudes, (ii) investigate the effect of group selection on risk and ambiguity attitudes, (iii) investigate the effect of seasonal variation on risk and ambiguity attitudes, and (iv) study the gender disparity in the measured attitudes in different environments. Our experiments yield the following key results. First, farmers exhibit higher risk and ambiguity aversions when they make choices in groups compared to choosing alone. In addition, group members’ characteristics such as social stature, literacy, etc. are important factors that influence farmers’ measured behavioral attributes after communication. Second, farmers are less risk-averse but more ambiguity-averse when group members are self-selected relative to when they are randomly assigned. Third, farmers in our sample are substantially more risk averse during the pre-harvest season compared to the post-harvest season. Finally, women tend to be more risk and ambiguity averse than men in all dimensions as male farmers tend to make more extreme choices than females.