Reaching out to socially distant trainees. Experimental evidence from variations on the standard farmer trainer system

The farmer trainer (FT) model has gained momentum as a cost-effective alternative to traditional agricultural extension systems. However, there may be friction in the transmission of information, whereby farmers closer to the FT may benefit more than socially distant farmers. This study explores whether variations on the standard FT model facilitate the diffusion of information outside the FT's pre-existing social network. On top of a standard FT training, a sub-sample of voluntary FTs in rural Uganda was randomly assigned to receive (i) vouchers for accessing professional extension agents, (ii) a signpost advertising the trainer services or (iii) further training aimed at tailoring the training to the specific needs of the trainees. The results show that the FTs assigned these treatment variations trained more farmers, a larger proportion of whom belonged to the FT's own close circle. The FTs who received vouchers were the only ones to reach out to more socially distant farmers and were also those who organized the most training sessions. We show that these effects are independent of any FT prominence in the village. Nevertheless, further evidence suggests exercising caution regarding the presence of friction in the transmission of knowledge, since knowledge and technology adoption appear to increase only among farmers closely connected to the FT.