The objective of the present research was to identify the effect of rural extension on the productive performance of Brazilian agricultural establishments, using as a measure of performance the technical efficiency of farms. The data used refers to the microdata of the 2006 Agricultural Census, accessed directly from the IBGE secrecy room. For this, an approach that combines the stochastic production frontier structure, taking into account the selection bias in the adoption of the rural extension (Heckman's approach), with the entropy balancing method was used. The results show that the rural extension contributes, in fact, to increase the efficiency in the use of the productive factors, with the producers adopting, more technically efficient than the non-adopters. When considering the differences according to the size of the establishment, an even greater effect was observed for the group of large producers. In addition, in general, public rural extension generated higher technical efficiency scores than those obtained by establishments attended by the private service. Acknowledgement :