Pesticide resistance can be viewed as an open-access resource problem. While traditional economic incentives are the standard prescription for this market failure, non-pecuniary behavioral approaches have also shown promise in managing these resources. I empirically evaluate the performance of an intervention in the latter class of instruments to promote compliance with refuge regulations in the context of genetically engineered Bacillus thuringiensis (Bt) corn. Refuge regulations are important policies for reducing the risk of Bt resistance. To encourage refuge compliance, the agricultural company Monsanto piloted a behavioral intervention in 17 North Carolina counties in 2013/2014. Using seed sales data, I estimate econometric models combining difference-in-differences with propensity-score-matching (PSM) to identify the effect of the program on grower behavior and overall refuge planting. A simple difference-in-differences (DID) estimator implies the program increased the share of refuge (non-Bt) seed sales sold to the average grower by 2.9%, whereas the DID-PSM estimator implies an effect of 5.6%. Non-pecuniary behavioral instruments deserve further consideration as a means of managing Bt resistance.