Walmart Product Ecolabel Drive – The Good and the Impossible?

Walmat announced this week a drive to survey, and create from their supplies a comprehensive database and index of the environmental footprint of the products it carries.

Given the heft of Walmart and the ubiquity of its supply chain, I’d argue that its move may be the only way to bring such an index into being. For pushing for ecolabels they deserve a huge amount of credit.

But issues abound. The effort is essentially what is known as a lifecycle analysis (LCA), calculating for a given product or process the environmental footprint of the product and its own supply chain (its embedded or embodied energy, environmental or emissions content, as the case may be).

The problem is that with LCA the devil’s always in the details, a combination of the three things 1) data quality, 2) what conversion factors to use, and 3) what is known as the boundary conditions, ie, what is appropriate to count.

You know the joke about economists, put 3 of them in a room, ask 1 question, get 10 correct answers. Well with lifecycle analysts, those same 3 people give you 547 correct answers, not 10. Because those three items, especially number 3, tend to have lots of shades of legitimate gray.

That’s why rigorous lifecycle analysts generally shy away from using LCA techniques to compare 2 products or processes, as opposed to using them to assess trends in a single product. Because two products, both with perfectly reasonable assumptions as to what should be counted, often mean a “right” answer for one is not equivalent to a “right” answer for another. In fact, you know you have an idiot for a lifecycle analyst, if he or she tells you his or her answer is right, and product A is better than product B.

A simplistic example, let’s say you have a plant in Thailand, that ships 1 mm cotton shirts a year to several vendors, including Walmart, and uses 1 mm kwh of energy a year. Should the energy allocation be then 1 kwh/shirt? What if 20% of those shirts are extra large? Should the XLs get a higher allocation because they’re bigger? Do you make that allocation based on size, %, square footage, time to manufacture, or cost, or a combination? Even financial inventory accounting leaves room for differences. What if some shirts cost less than others to produce, should cost be included as a variable in the allocation, and if so, should average, LIFO, or FIFO be used? What if only 30% of the shirts go Walmart, and the others go to a place that doesn’t have ecolabels? How do we account for shifting allocations over time if products in one batch come up with different labels, or get shipped by different ships? What if one ship is 20% full and the other 100% full? And how do you allocate the energy footprint for product returns, shrinkage, or wastage? What periodicity do you pick? Allocate quarterly, monthly, annually? Not every answer is material, and not every answer is difficult in every case. That’s the whole point, it varies. All of these can have legitimately different answers depending on the nature of the business (and if we get comments on this article explaining the “right” answer, that will just highlight the point), and when you consider that multiple companies or plants supply components, and therefore part of the answer to each other to calculate the final footprint, the permutations of “right” blow out fast.

Part of the solution boils to down to better data and better standards, for which the Walmart effort will be a huge help. And don’t get me wrong, the LCA community and hundreds of different LCA databases and models have been wrestling with these issues for decades, and have a lot of experience setting standards, and are always hungry for more data. Unfortunately, the standards as a matter of necessity leave enough room for judgement to be applied, so that eco comparability across products may always be in the eye of the beholder. Bottom line: know the limitations of your data.

So huge massive kudos to Walmart for driving the world forward, again, but the real devil is that comparability within the bounds of the margin of error is virtually impossible to achieve. And it’s ripe for gaming.

Neal Dikeman is a partner at Jane Capital Partners, and has cofounded, run, invested in, or served as a director of multiple startups in cleantech and technology. He is Chairman of Carbonflow and, and a Texas Aggie.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!