Food and agricultural industry locational determinants research: Aggregation bias and size measurement in the agricultural support industry

Abstract

Federal administrative data present a valuable opportunity for food and agricultural industry locational outcome research. We review issues with aggregated U.S. public data and summarize current methods. An example empirical approach combines federal administrative and secondary data. We compare results with differing levels of industrial aggregation. Results indicate locational determinants vary in magnitude, sign, and significance across industries and their sub-industries, as well as between employers and non-employers – nuances commonly missed with public data. We conclude by emphasizing that studies relying on public (more-aggregated) data may miss locational outcome relationships or inappropriately generalize to sub-industries and suggest data access changes.