Altering Farmers' Pest Management Practices Through Educational Outreach Programs: Evidence from a Panel Selection Model

In this paper we investigate whether or not an extension outreach program intended to educate California almond growers regarding alternatives to organophosphates (OPs), a highly toxic class of pesticides, affected pesticide use decisions. Using a panel dataset of 6708 growers observed during an 11-year period, we demonstrate that the outreach program has significantly contributed to the decrease in use of OPs. Grower education programs intended to elicit voluntary disadoption of pesticides that are particularly environmentally harmful are an alternative to banning a pesticide entirely or relying solely on use restrictions to mitigate a pesticide's harmful effects. Our analysis makes three contributions to the economic literature in addition to its contribution regarding the impact of extension programs on farmers' decisions. First, it contributes to the applied econometrics literature by illustrating the use of panel selection estimation methods in an application outside of labor economics. Second, it contributes to the empirical literature on adoption by implementing new techniques to address the presence of individual-specific effects in both the participation and the primary equation of the sample selection model that have been discussed in the adoption literature, but not applied. The third contribution of our analysis, also to the empirical literature on adoption, is that it demonstrates that data with relatively little grower-specific information can be paired with appropriate econometric techniques to address questions of technology choice.