Why Online Retail Sales are Much Larger than US Census Data Report
39 Pages Posted: 18 Jan 2016 Last revised: 9 Feb 2016
Date Written: February 7, 2016
Based on the publicly available information on the data sources and methodology used by the Census Bureau to estimate e-commerce and physical retail sales, and based on detailed e-mail communications with the Census official responsible for the retail trade data, we found, in a version of this paper released on January 28, 2016, that the Census Bureau significantly undercounted online retail sales; we also provided what we believed were conservative estimates of the undercount. The Census Bureau has subsequently “corrected” the e-mail communications from the Census official and informed us that they follow different procedures and report different figures than we, and we believe any reasonable researcher, would have surmised. Assuming that Census does what they now say they do, we still believe that the Census Bureau undercounts online retail sales but it is not possible to verify that this the case or provide an estimate of the magnitude. We make several recommendation for the Census Bureau in light of our research, and our experience trying to validate their data, including that the Census Bureau commission an independent audit of the calculations of its e-commerce data for retail trade.
Keywords: Internet economy, e-commerce, Census retail data, retail, online sales, online retail, digital commerce, retail reinvention
JEL Classification: L10, L11, L22, L81, L86, M21, N82, Y1
Suggested Citation: Suggested Citation