Repeated Transaction in Rural Grain Markets of Ethiopia

This article uses the theory of repeated transaction to model exchange in rural grain markets. We examine the theoretical and empirical drivers of repeated transaction, as well as potential problems that may arise as a result of its widespread use. We develop a structural repeated game to represent buyer-seller relations in a developing country grain market, motivated by observations in Ethiopia. The model generates hypotheses regarding the logic that drives long-term tied transactions, as governed by information access, screening and investment costs, and time preferences. The model's predictions are tested using market transaction data collected in Ethiopia in 2009. Results support the view that a relational contract is driven by access to information and the costs of screening. Our major finding is that trust in relational trading can emerge through costly repeated interaction.