The Hidden Burdens of Well-Intentioned Standards
Supplier Meet-ups are private, off-the-record conversations open only to suppliers, held once per month and hosted by the Asia Garment Hub. Each month, one supplier shares a specific challenge they’re facing while the rest of the group shares feedback and offers support.
The challenge shared this month was about the difficulty of establishing data supply chains for LCA analysis. The supplier began with some general context: They are a vertically integrated mill doing spinning, weaving, processing, and garment manufacturing. Their focus is denim. They started working on LCA calculations last year. Though much of the data they need for this is within their control (because they are vertically integrated), the main data outside of their control is related to cotton and chemical inputs they procure. They shared that they are buying from globally renowned chemical suppliers (and not from unreputable chemical suppliers). They approached the five global and reputable chemical companies from whom they source products with data requests. Only one was able to provide the data needed for their LCA calculation, however, the data shared was absolute whereas they wanted normalized data. Normalization is a way of allocating data for a given facility to specific products per impact category.
This was followed by some clarification questions: is it that the chemical companies have the data but are unwilling to share it, or is the problem that the chemical companies themselves don’t have the data? Though the supplier couldn’t be certain, they speculated that the problem was that chemical companies themselves do not have the information.
Another supplier remarked: the only way forward, at least for now, is average numbers. We need to set up research bodies to do LCAs on basic chemicals. Even absolute numbers are an over-expectation. The supplier emphasized: ensuring the accuracy and reliability of the collected data is crucial for a meaningful LCA. However, data quality can vary, and it is challenging to verify the accuracy of the information obtained. Different data sources may have varying degrees of precision, completeness, and consistency, leading to uncertainties in the assessment.
Someone asked whether they could work together with other suppliers sourcing these products from the same chemical companies to see whether, together, they could more effectively apply pressure for the data they needed. The supplier responded that the chemical companies have plants in different locations. This means that even if two mills are sourcing the same product from the same chemical company, it’s possible that the chemical products are coming from different plants and therefore would have different datasets. This was a barrier to centralizing pressure for data.
Someone then remarked: there are lots of people and resources for figuring out HOW to do LCA calculations. But nobody is giving guidance about what to do in the event that data is not available. Suppliers are left to their own devices to make certain assumptions. Are these the right assumptions to make? Some guidelines would help with this uncertainty. Furthermore, collecting data from a diverse range of sources may require the use of various technology tools and platforms. Integrating data from disparate sources, ensuring data interoperability, and managing large datasets can be technologically challenging as well.
Someone else suggested: could you create incentives for your suppliers to give you the data? The supplier responded that they had considered this, but in order for those incentives to be effective they would have to be financial. This was hard for them to do given that they themselves were not getting any price premiums from their own customers (brands and retailers). Frankly, they were struggling just to get orders – sometimes due to just a few pennies.
As for me, the universe seems to be hitting me over the head with this challenge. For example, I recently heard a different supplier in an entirely different context present almost exactly the same challenge. At a Transformers Foundation event back in April, Ebru Debbag of Soorty Denim candidly shared during a panel open to the public that she too was struggling to get data from upstream suppliers. She expressed that, as a result, she and her team were forced to make numerous assumptions throughout their LCA calculations. Like the supplier in this group, she reflected that it was hard to know whether those assumptions were correct and that this made her feel quite alone and isolated.
In parallel, while at Kingpins, I had another conversation with Rashid Iqbal of Naveena Denim NDL who said that he was interested in coming together with a small group of other denim suppliers to agree on and standardize certain assumptions needed to do an LCA for the most basic denim fabric. He then wanted to work with a research institute to establish a transparent methodology and baseline impact calculation. His perspective was that this would help give some guidance to suppliers who must currently make these decisions on their own and with little guidance.
At the end of the Supplier Meet-Up, the presenting supplier remarked: What’s important? The audits? The data collection? Or the progress? This remark pushed me to reflect on the bigger picture: standards created by well-intentioned sustainability professionals not working directly in production tend to create hidden burdens for suppliers (I would be remiss not to mention that long conversations with Dr. Divya Jyoti have been critical in helping me to articulate this point). The irony is that these burdens risk shifting resources away from the activities that might actually drive more meaningful change.
Join us for June’s Supplier Meet-up (8 June at 9:30CET)! It’s a special edition. We’ll be joined by Olivia Windham Stewart and Sarah Dadush, consultants from the Responsible Contracting Project. They will briefly present the model contract clauses they’ve been working on for STTI, and ask for supplier input and feedback. Email me and I’ll share the calendar invite.