Title: Eff ect of Cotton Farmers' Spatial Yield Variability Perceptions on Variable Rate Technology Adoption.
Rationale: The relationship between variable rate input application and yield perceptions of cotton monitor adopters is an important issue both for VRT (=Variable Rate Technology) dealers as well as extension education programs. If we find that there is indeed a strong relationship between VRT and perceived yields, then agribusiness firms have an incentive to provide promotions for yield monitor information to cotton farmers who are most likely to benefit from these technologies. Thus they can better see their true yield variability and be encouraged to purchase VRT products. Likewise, if more educated farmers are more likely to adopt yield monitors, there are incentives to develop fertilizer technology training programs or equip farmers with more computers.
Objectives: The objective of this paper is to estimate the effect of farmers’ yield perceptions on variable rate input decisions and infer about the influence of expectations on technology adoption. Specifically, we are trying to answer whether the expected yield variability of farmers who adopted site specific technology, may affect their variable rate input decisions. Site specific or precision farming technology is a strategy that uses new tools such as global positioning systems (GPS), aerial or satellite infrared imagery, green seeker etc. in order to monitor and identify high and low yielding areas within a field. Variable rate technology or VRT is the application of the appropriate input levels, such as fertilizer, seeding, chemical application and tillage, within a field. Previous studies have investigated the relationship between cotton monitor adoption and subjective yield variability. We will move one step ahead and try to answer whether the variable rate decisions of farmers who have already adopted this type of precision technology are affected by their reported yield variability.
Methodology: We will use a three-step sequential probit model, which accounts for possible endogeneity between yield monitor adoption and yield perceptions. Firstly, we will estimate the socio-economic factors that influence precision technology adoption. Then we will estimate the yield perceptions including the predicted probability of yield monitor adoption as an independent variable in the second step estimation. Our last step, which is also the interest question, will be a binary probit equation which models the effect of perceived yield variability of farmers who adopted yield monitors on VRT.
Data: Our data is collected from a survey sent to cotton producers in 12 Southeastern states in January 2009 and refers to crop period of 2007-2008. Among the 1692 farmers who returned the questionnaire, 631 did not adopt any precision farming technology and 1061 adopted at least one technology. Farmers also provided information about sociodemographic factors such as income, farm size, education etc.
Results: We expect to find a positive and significant relationship between VRT adoption and expectations about yield variability. Farmers, who believe that their within field yields are more variable, will most likely apply VRT instead of a uniform rate technology.
Key Terms (4): yield perceptions, precision farming, endogeneity, variable rate technology