Search
Search
Search
Sophie Benson
You input data and you get a score, and in doing so you can identify hotspots which need addressing and support better supply chain decisions. But the content and quality of that data can vary massively, so comparing one LCA to another isn’t quite as straightforward as you might think. What’s more, as independent analyst for the sustainable apparel sector Veronica Bates Kassatly says, “If you put garbage in, you get garbage out.” Let’s delve into it.
While standards such as ISO 14040:2006 lay out the definition of the goal and scope of LCAs, they don’t specify techniques and methodologies. Nor is there one single body undertaking all LCAs. As such, approaches can vary wildly. Let’s imagine two brands use two different companies to conduct LCAs of plastic shampoo bottles. When looking at manufacturing, one company uses a sample of five factories in a highly regulated market, while the other uses a larger, international sample. This one difference in data gathering will have a huge impact on the results pertaining to the same product. As would looking at different countries for the source material, since the greenhouse gas impacts of the extraction and refinery of crude oil can vary by a factor of seven depending on the location. If we consider all the data which must be collected as part of the LCA inventory across a whole lifecycle, it’s clear to see how the quantity and quality of data can create wildly different results.
Even within a single LCA, data roadblocks occur. A 2018 LCA of Organic, BCI, and conventional cotton cultivated in India notes that “for many relevant aspects (such as soil types, nutrient content of soils, soil erosion) primary data were very hard to obtain, therefore for some of the data proxy values were applied, not necessarily representative of local conditions.” It goes on to add that “Indian geography has different climatic conditions, but the study focused only on the western part of the country.” Another LCA with more specific data or a more representative national sample may reach different conclusions.
Not all LCAs are forthcoming with their data or methodology limitations. The Higg Index, a fabrics and materials rating system which was widely criticised in 2022, reportedly used narrow data sets and data weighted towards more regulated markets, which meant fossil fuel-based synthetics scored ahead of natural fibres. While the ratings were available to all, the data came with a fee, meaning people made like-for-like comparisons without understanding the variability of the data. The Sustainable Apparel Coalition, which runs the index, has now “de-emphasised direct comparisons”.
While various software and models are used to calculate LCAs, people are still integral to the process, and people have biases and limitations. Factors such as a lack of knowledge of the subject at hand and different interpretations can prompt varied results. “On average, LCAs provide results with an uncertainty of at least ±10%,” says Baijia Huang, Sustainability Manager at Rockwool Group. “In other words, an LCA of the same product completed by 10 different LCA practitioners has the potential to yield 10 somewhat different results.” A paper on practitioner-related effects, meanwhile, states that “It is, somehow, a common experience among LCA practitioners that apparently negligible changes in the model setup can produce a large deviation in the final results.”
Simply put, different people get different results, but the financial and market-based benefits and disadvantages of LCA outcomes must also be considered in how and why results might vary. For instance, it would serve Plastics Europe (an association of plastics manufacturers) and its member companies to paint plastic in a positive light, sourcing data from regions and time periods which will yield favourable results. Its Eco-profiles, covering the main raw materials needed to produce plastic, are used in LCA databases by the likes of openLCA, ecoinvent, and GaBi. But data for crude oil and natural gas, for instance, was last calculated in 2005, undermining the accuracy and relevancy of LCAs which use it as an input.
Data from the association was also used to calculate polyester’s Higg Index rating, so it is perhaps unsurprising that it was rated as one of the world’s most sustainable fabrics. A rating which should have been globally representative was based on European data, when the majority of polyester production is in fact based in China, where fossil fuels represent 85% of the energy mix. In the EU, the GHG emission intensity of electricity generation in 2020 was 215.7 grams of carbon dioxide per kilowatt-hour, compared to China's 580 grams. In providing specific, localised data for a global rating, an association of plastics manufacturers made a plastic product look significantly more favourable and greatly misled an entire industry.
As we’ve covered, depending on the data used, one material can seem unexpectedly more environmentally favourable than another. But the data that isn’t used can have an impact too. You may, for instance, be comparing flooring materials for a store install and find a synthetic material performs better in an LCA. But the fact that the floor will be subject to high foot traffic and frequent waxing and cleaning, therefore emitting much higher levels of volatile organic compounds (VOCs), may not be taken into account at all for that material.
Impacts such as landfill runoff or the shedding of microplastics, which are harder to measure, may be ignored, as well as benefits such as a product being used as a part of a refill system. Other impacts are harder to quantify. For instance, a 2022 paper in the journal Sustainability states that “biodiversity, noise, and smell, are examples of impact categories that can be considered ‘under development’”, while a paper on the challenges of LCAs says that factors like the health effects of noise and odour - which include sleep disturbance and an increased risk of cardiovascular diseases - are often overlooked. Several other academic papers mention a lack of assessment of water use.
The difficulty in measuring certain impacts poses a number of problems. One LCA may attempt to quantify them by using proxy values, while another may omit them entirely for lack of solid data. And where health and environmental factors simply can’t be quantified, it can result in significant impacts being skimmed over entirely, leading to long-term health impacts across populations.
It’s self-evident that when the organisation funding the LCA has a vested interest in the outcome, boundaries and methods of allocation are likely to be selected that favour the material concerned. Seek analyses executed by a third-party, where data is sourced without involvement from the commissioning company.
You cannot compare LCAs produced using different boundaries or methodologies. Doing so can easily make materials that have very similar production inputs look vastly different in terms of purported environmental impact. If you want to accurately compare the impact of two separate materials, the LCAs should be conducted in the exact same way to minimise risk.
But the misuse, misunderstanding, and misrepresentation of LCAs that’s rampant in the sustainable apparel sector is not.
