Heterogeneous returns and the persistence of agricultural technology adoption

In this paper we explore whether low rates of sustained technology use can be explained by heterogeneity in returns to adoption.  To do so we evaluate impacts of the Cocoa Abrabopa Association, which provided a package of fertilizer and other inputs on credit to cocoa farmers in Ghana.  High estimated average productive impacts for treated farmers are found to be consistent with negative economic profits for a substantial proportion of the treated population.  By constructing an individual-specific measure of returns, we demonstrate that low realized returns among adopters are associated with low retention rates, even after conditioning on output levels and successful repayment.  The results are consistent with the hypothesis that high average returns mask substantial and persistent heterogeneity, and that farmers experiment in order to learn about their idiosyncratic returns.