Econometrics and archival data: Reflections for purchasing and supply management (PSM) research

Purchasing and supply management (PSM) has faced unprecedented disruption over the past two years due to COVID-19 pandemic, input shortages, extended supplier lead times, record international transportation costs, and commodity price increases. Studying such phenomena is often best completed using archival data, such as data from government agencies or international organizations. This manuscript emphasizes how leveraging archival data often necessitates an iterative research process whereby researchers must first familiarize themselves with the data to ensure their scientific hypotheses can be appropriately tested. We further provide recommendations regarding how researchers should formulate generalized linear models (GLMs) to test theoretical predictions. Our approach emphasizes mapping scientific hypotheses to statistical hypotheses, as opposed to centering on issues of omitted variable bias (OVB). An illustrative example is provided where Census Bureau trade data are compiled to test whether the insurance and freight costs for waterborne containerized imports from Asian nations that enter through West Coast ports have risen more than the same products imported through East Coast ports. The research suggests the need to reorient how GLMs are formulated to better ensure researchers structure them to appropriately test their theory, in contrast to the current zeitgeist that overly emphasizes OVB.