The lack of adoption of new farming technologies despite known benefits is a well-documented phenomenon in development economics. In addition to a number of market constraints, risk aversion predominates the discussion of behavioral determinants of technology adoption. We hypothesize that ambiguity aversion may also be a determinant, since farmers may have less information about the distribution of yield outcomes from new technologies compared with traditional technologies. We test this hypothesis with a laboratory experiment in the field in which we measure risk and ambiguity preferences. We combine our experiment with a survey in which we collect information on farm decisions and identify market constraints. We find that ambiguity aversion does indeed predict actual technology choices on the farm.
Risk Preference Measure: Derived from five options presented to subjects, each containing different payoffs corresponding to levels of uncertainty provided by the probability of the state of nature occurring. Ambiguity Preference Measure: Derived from five options presented to subjects, each related to different payoffs but the subject is unaware of the probability distribution over the state of nature occurring. Dominated Alternative Preference Measure: Measure of subjects' ability to understand the decision-making problem in rational terms. Choice to invest or not invest in new technology: Survey data about investments in new technology (modern crop varieties) taken by the subjects on their actual farms.
Effect of risk, ambiguity, and dominated decision preferences on farmers' choice of diversification across (modern) varieties of crop: Ambiguity aversion is found to be correlated with technology choice, but risk aversion is not. The more ambiguity averse farmers are, the less likely that they diversify across modern varieties of crop, while risk aversion does not influence this decision.