Individual and Social Learning in Bio-technology Adoption: The Case of GM Corn in the U.S

Genetically Modified (GM) technology has been widely adopted by the U.S. farmers within just a recent decade since the first generation GM varieties were commercially planted in 1996. Also, it has provided economists with various controversial issues: food safety, biotech industry concentration, labeling regulation, and environmental contamination. In dealing with them, it’s the analysis of farmers’ technology adoption behaviors that need to be studied fundamentally because it plays a role of the first step to evaluate the associated economic policies and suggest more efficient GM regulations. The high adoption rates of GM technology are believed to be driven by farmers’ expectations for more profitability than planting non-GM (conventional) seeds. In addition, according to the recent improving biotechnologies, the single trait GM seeds of herbicide-tolerant (HT) or insect-resistant (IR) are rapidly substituted by stacked gene varieties. Those trends of tremendous diffusion of GM crops and increasing access to the stacked seeds in such a short history comes with a question about which determinants have influenced farmers’ active adoption behaviors under uncertain profitability. Most of the previous GM adoption literatures have analyzed determinants affecting the diffusion of technology with regards to farmer characteristics such as farm size, education level, risk preference, and credit access. Another recent study pointed out GM crop characteristics represented as average yields, labors, or herbicide/pesticide usages. However, few studies paid attention to the role of externalities in technology adoption decisions; 1) learning process – a process of improving farmers’ ability to implement new technology and allowing them to make better decisions. They are composed of individual (learning-by-doing) and social learning (learning from others); 2) neighborhood effects – the tendencies that a farmer’s adoption is affected by his/her neighboring farmers’ behaviors in a peer group. These two concepts are worth while to be analyzed empirically in the sense that, in reality, individual technology adoption is affected not only by one’s own experiences but also by others’ behaviors through continuous social interactions. Also, the learning process requires introducing the dynamic framework into the analysis because farmers’ acquired information generates an ability to predict future profitability and leads to the situation that farmers are forward-looking. Therefore, this paper tries to develop a dynamic GM technology adoption model with externalities and explore the importance of learning and neighborhood effects under uncertain profitability. To the GM technology adoption studies, this paper makes the following contributions: first, externalities of learning and social interactions are directly specified in the empirical model; second, introducing dynamic framework expands the previous limited static level works due to lack of accumulated data in short history of GM technology; finally, the dynamic structural approach can suggest scenario evaluations in terms of various GM issues.