Comparing the impacts of biofuels using survey and non-survey data

Abstract

This paper utilizes survey data to compare several non-survey methods of modeling the economic impacts of biofuels plants. It examines differences in the input coefficients derived from the survey versus the trade coefficients generated through the non-survey methods. It finds that of the three non-survey methods examined, the Swenson (2006) model input coefficients most closely represent those found in the survey based on the performance of the non-survey model input coefficients in a variety of statistical tests. Further, it examines the economic impacts (multipliers) generated by these models compared to those generated from the survey. Based upon statistical tests of the multipliers, the Swenson model's estimated impacts most closely represent the impacts derived from the survey.