As Environmental Product Declarations become more widely used, manufacturers are increasingly able to get their new EPDs created using data from existing EPDs already developed by their suppliers.
Instead of only relying on a background dataset for materials sourced elsewhere, their suppliers' EPDs should theoretically provide the most accurate environmental data for it.
But that's not actually always the case.
Writing on LinkedIn, LCA specialist Fabian Diaz highlights the fact that EPD verifiers (including us here at LCA Support) are increasingly concerned about suppliers presenting EPDs with unreliable environmental data.
"Because of that, some verifiers are starting to almost desk-reject the use of EPDs from specific consultants and tools, as inputs for the product systems under review," says Fabian.
This is the right thing for an EPD developer to do. They can (and should) just outright reject unreliable data, even if it’s already published as an existing EPD.
We all have a duty to prevent unreliable data cascading downstream in the supply chain because that ultimately harms the reliability of building LCAs, which are so crucial for a sustainable construction sector.
In addition, do bear in mind that some EPD programme operators let you flag issues with EPDs even after they’ve been published.
If these processes are more widely available and we all get into the habit of using them, we add more accountability and trust to all EPDs (or at least those published by respectable operators).
There is still a dilemma for EPD developers though. Often, we'll come across a supplier EPD that is just slightly flawed and should have been subject to better verification. There are may be a few simple errors and a lack of transparency about the data. Or it could just be that the EPD wasn't published by a proper programme operator.
As an EPD verifier myself, my instinct is to reject data from these EPDs too. The dilemma, however, is that the data is likely to still be far more accurate than a background dataset.
So how do we fix all this systematically, instead of just firefighting by flagging issues after they are published?
Fabian has good suggestions:
- Harmonise verification processes to ensure better quality and fair competition.
- Strengthen and clarify verification procedures around the use of EPDs as primary data.
- Demand the declaration of data specificity for transparency and data quality assessment,
- And educate manufacturers on how to recognise a high quality EPD (and LCA consultant and tool) from an EPD that just works for box-ticking compliance.
I want to back up a little bit first though. In my perspective, we first need identify foundational problems.
EPDs were pioneered by the private sector, first as a voluntary initiative. That means:
- The underlying rules and standards are loose, inconsistent, and open to interpretation.
- There are no qualifications required to be an EPD developer, nor too much training taking place.
- EPD verifiers are paid by quantity of verifications without enough incentives for quality.
This system functioned relatively ok in the past. But EPDs have fast evolved and merged with mandatory regulations. Their value is increasing and the ecosystem around them is expanding rapidly.
So the underlying structural issues are just becoming more visible now. You can’t keep building higher on shaky foundations.
But we don’t need to start from scratch, nor fix every contributing factor. And we certainly can’t wait for regulations to be perfected.
The market can again take the lead and raise EPD quality - including by embracing new technology, developing new tools, and growing as an EPD community with enhanced standards.
More automation is inevitably going to assist in certain areas.
Personally, I think we really need to focus on point 3 by safeguarding the quality of EPD verification. Get that right and point 2 isn’t so much an issue.
Because we shouldn’t raise barriers to EPD development. On the contrary, we need to encourage a much wider understanding of the process and principles of embodied carbon calculations (and reductions), not silo it among specialists.
That’s how we build a sustainable construction sector with adequate checks and balances to ensure all the good work (and only the good work) is properly verified.
We’re currently working on one such solution for that at LCA Support and I can’t wait to share more publicly soon.