Ambiguity aversion and the adoption of weather index-based insurance: experimental evidence from Uganda

Purpose This study aims to estimate farmers’ ambiguity aversion and analyze its impact on the adoption of weather index-based insurance in Uganda. Design/methodology/approach The research question we address is: Does the ambiguity of insurance contract outcomes affect farmers’ decisions to adopt weather index-based agricultural insurance (WII)? To explore this, we use incentivized choice experiments to elicit ambiguity preferences and insurance adoption decisions. Our sample includes 291 small-scale coffee farmers from Central and Western Uganda. Given the panel data and the dependence of adoption decisions over time, we employ a dynamic probit model to analyze how ambiguity aversion influences WII uptake. Additionally, we use a pooled binary probit model to test the robustness of our results to different model specifications. Findings The results indicate that the adoption of WII decreased with each round of the choice experiment, particularly among ambiguity-averse farmers. We found that 58% of respondents were ambiguity-averse and that ambiguity aversion is negatively associated with WII adoption. Conversely, tertiary education positively influences the likelihood of adopting WII. Practical implications The results imply that farmers are not likely to invest in risk management strategies whose outcomes are not predictable with some level of certainty. Also, the negative effect of ambiguity aversion can be reduced as farmers gain more knowledge about the insurance products. Our findings highlight the need to provide sufficient information and education about the use of WII. Originality/value This study contributes to the limited literature on the role that behavioral factors play in the adoption of index-based agricultural insurance.