Mapping the ‘seams’ of retail’s global supply chain

If the last few years of global trade disruptions have taught us anything, it’s that behind every storefront, shelf, and product is a supply chain so complex that sometimes even major corporations can’t keep track of it all—let alone your average consumer.

In the interest of demystifying this murky world of logistics, Benjamin Y. Fong, associate director of the Center for Work & Democracy at Arizona State University, has taken on the herculean task of mapping out the supply chains of national retailers such as Walmart, Target, The Home Depot, and Dollar General.

The project, which Fong publishes on Substack under the title “On the Seams,” as well as in the publication Phenomenal World, has produced several primers detailing how each company’s supply chain functions—with a focus on the role of labor—and where exactly fulfillment and distribution centers are located across the US.

Retail Brew spoke with Fong about his methods for uncovering the physical footprint of supply chains, the state of automation in logistics, and the crucial differences between e-commerce distribution and traditional retail.

This interview has been edited for clarity and length.

How were you able to piece together these maps?

There are a few different interesting data sources. There are commercial real estate databases, and there is OSHA data. The Occupational Safety Health Administration keeps something called an injury tracking application for all reports of injuries at different work sites, and because the warehouses in question tend to be of a certain size where there’s at least going to be one major injury or accident every year, they have pretty good information about where facilities are, what their addresses are, what their head counts are.

This is not information the companies want to be public necessarily; there’s no real incentives for them to make it public. The more they show you how the sausage is made, the less appealing their products become in some ways.

Labor is one of the main focuses of your research. What did you learn about the distribution side of the retail labor force, and did anything surprise you?

One surprise would be that most warehousing work, despite what you hear in the news, looks more or less as it did 10 years ago. It’s pallet jacks. It’s people running the floors with headsets.

There are real disincentives for third-party logistics companies in particular to adopt the automation everyone talks about. The really interesting forms of warehousing automation are mainly being deployed by Amazon, but they’re pretty expensive and the margins for 3PLs are pretty thin.

What else accounts for this relative lack of automation in the industry?

We’re at a moment where we’re running up against a difference between two very different architectural logics. E-commerce demands certain facilities and a certain set of relationships between those facilities that really encourages automation, especially in Amazon’s world where the promise of Prime membership is to get stuff to your doorstep without extra delivery fees within two days or maybe even on the same day. That requires a lot of investment in fine-tuning the speed and efficiency of your distribution network.

In the case of traditional retailers…there’s not that same demand. You need to have stores replenish the inventory on a regular basis, but it’s not like you have to fine tune need to get something to individual customers. That’s its own logistical problem, and so it just doesn’t demand automation in the same way.

What are some geographic challenges facing retailers when it comes to distribution?

In general, geographically, there’s a physical coordination problem for all retailers that US geography poses, which is the fact that the majority of imports come in through the West Coast, but 65% of the country lives east of the Mississippi. That’s really the central problem…and we’re not very good at it because the ports are constrained in their capacity.

How do you see supply chains evolving as a cost center for retailers?

We’re at an interesting point where some of the key retailers are beginning to see that investing in new ways of doing things…is actually not that great for them. A, it doesn’t rapidly improve their operations; and B, it’s just tremendously expensive. I think one interesting story…is the fact that Kroger recently rolled back…automated grocery fulfillment. That’s one of those cases where the company was like, “Hey, this is the future. Everyone’s doing it.” And then in actual practice, it just doesn’t make sense.

I think that increasingly people are going to realize you can’t take logics that make sense in the e-commerce world and import them over to traditional retail without either bankrupting yourself or making your operation extremely complicated in such a way that it’s detrimental to the overall operation.