Lifting quality constraints to agricultural technology adoption in the Ugandan market for maize seed

Agricultural technology remains under-adopted among smallholder farmers in sub-Saharan Africa. We investigate how the (perceived) quality of an agricultural input affects its adoption by means of two interventions. Firstly, agro-input dealers could lack knowledge how to handle and store maize seed, leading to poor average quality in seed shops. A dealer training is expected to improve seed quality and subsequently adoption on the farm. Secondly, as the sellers of maize seed have more information about its quality than the buyers, we hypothesize that this information asymmetry leads to poor seed quality offered by some dealers. To overcome this classic lemon technology problem, we implement a crowd-sourced information clearinghouse like yelp.com, meaning that we collect and disseminate farmers’ ratings of seed quality to make it more observable to dealers and farmers. Both interventions are tested in a randomized control trial among 350 agro-input dealers and 3500 smallholder farmers in Uganda. We find that the dealer training does not have a clear positive impact on seed dealers or their customers. On the other hand, clearinghouse treated farmers are more likely to switch to different seed shops. Treated dealers anticipate this and increase their efforts to outperform their competitors, and therefore sell and earn more. Clearinghouse treated smallholders are more likely to adopt improved maize seed, and thus harvest more than control farmers. Hence, a clearinghouse which solves the information asymmetry between sellers and buyers clearly improves the Ugandan market for maize seed.