Upstream, Downstream: Diffusion and Impact of the Universal Product Code

Timothy Simcoe and Emek Basker

The humble and ubiquitous barcode, named one of “50 Things That Shaped the Modern Economy,” first appeared in 1969.  Barcodes allowed both manufacturers and retailers to track with unprecedented accuracy and precision the movement of goods through the supply chain to the consumer.  Its success after its introduction in the grocery industry led to its use on nearly all retail products today. Despite its ubiquity, its spread has been the subject of very little quantitative analysis.  Timothy Simcoe and Emek Basker analyzed the effects that barcode adoption had on employment and innovation.

The Universal Product Code (UPC), or barcode, provides a classic case study for two-sided network effects.  These imply that there are benefits of adoption for upstream firms (manufacturers) supplying a “UPC-ready” downstream sector (retailers), and vice versa.  Manufacturers needed to know that it would be worth it for them to change their packaging to add barcodes, and retailers needed to know that it would be worth it for them to invest in large-scale renovations to introduce checkout scanners.  

To look for these effects, Simcoe, Professor of Strategy & Innovation at Questrom School of Business, Boston University, and senior fellow of the Technology & Policy Research Initiative at Boston University School of Law, and Basker, Principal Economist in the Center for Economic Studies at the U.S. Census Bureau, examined information on UPC registrations, employment, and trademarking, for approximately 779,000 manufacturing firms and 867,000 wholesalers for the period 1975 to 1992.  Simcoe and Basker found that although only a small number of firms registered for a UPC during that period, it included all of the largest firms, and that by 1980, more than 90% of all grocery products had a barcode.  By 1983, barcodes had spread widely to other industries as well.

To test whether manufacturer adoption drove retailer adoption or vice versa, Simcoe and Basker relied on the implication that benefits of UPC adoption are higher for firms when more of their competitors have also adopted:  if all a company’s competitors used UPCs, then it would behoove it to do so as well. Simcoe and Basker found statistically significant evidence that a manufacturer’s adoption of barcodes increased the likelihood that its rival would adopt them in the next year.   

For retailers, they had to compare the cost of installing barcode scanners at checkout with the expected benefits of doing so, which depended on how many of the store’s suppliers had barcodes on their products.  The data here show two-sided network effects, with larger retailers leading the way in adoption, which is consistent with the economies of scale in scanner deployment.

To analyze the effect of UPCs on employment, Simcoe and Basker compared firms of similar size and age, using firms that did not adopt UPCs as the control group.  They found that UPC adopters grew faster than non-adopters before they adopted UPCs, and continued to grow faster for a time after adoption.  This applies to both manufacturers and wholesalers, and is consistent with the observation that firm size and UPC adoption are strongly correlated, since faster pre-adoption growth necessarily produces a larger firm.  

Simcoe and Basker also analyzed trademark applications and found that both manufacturers and wholesalers had a large and statistically significant increase in trademarking following UPC registration.  Indeed, the likelihood that a manufacturer filed for a new trademark in a given year increased by 230% following UPC registration, and the number of trademark applications increased six-fold following UPC registration, both of which indicate that manufacturers were able to experiment with new products and stores were able to stock a larger variety of products after adoption.  

Although all the results point to strong positive effects of UPC registration on employment and trademark registrations, there may still be some omitted variables that explain the correlation.  Companies that adopted UPCs may be more innovative in general, and may have had those same effects in the absence of UPC adoption. However, the findings do suggest that once stores installed checkout scanners, manufacturers’ adoption of UPCs helped stores, manufacturers, and employees.  

SSRN Working Paper