Are users of market information efficient? A stochastic production frontier model corrected by sample selection

This article analyzes how information use affect farm productivity and efficiency. Our hypothesis is that farmers make better decisions when they use information (for example, choosing a high value crop combination or selling the products at higher prices) and that will enhance on productivity and efficiency. We use two techniques to mitigate the possible biases generated by observable and unobservable variables: Propensity Score Matching (PSM) for the first one and the stochastic production function (SPF) approach corrected by sample selection for the second one. We take advantage of the underused Peruvian National Agricultural Survey (ENA) which includes information about 12 877 farmers located in the Andean region. Our results show that farmers who use information are systematically nearer to their frontier than those who do not use information (0.50 vs. 0.47, on average). The analysis by plot size and age suggest that farmers with smaller plots and those who are middle age are more efficient in the users group; however, the relation is not clear among the nonusers of information. Thus, more research is needed about the complementarity of the agricultural inputs and information use. These results can contribute to the design of a cost-effectiveness evaluation of information extension programs.