Fashion is the world’s second most polluting industry, after oil. It shows up in panel introductions, brand sustainability pages, and student essays, usually with no citation attached, because the people repeating it inherited it from someone who also had no citation attached. Three separate journalists have gone looking for the study behind it. None of them found one.
Where the claim came from, according to the person who tried hardest to find it
Alden Wicker, writing for Racked in March 2017, set out to source the number the way a reporter sources any number: find the study, read the study. She could not find one. No study, no official report, nothing published under any agency or research institution’s name. She emailed sustainable fashion experts and advocacy groups asking where the figure originated. Several promised to follow up and never did. A couple pointed her toward the Danish Fashion Institute, which had used the claim in its own materials. When Wicker asked the Institute directly, it disavowed the statistic rather than defending it.
Fashionista revisited the same question in February 2020, three years after Wicker’s piece ran, and found the claim still circulating at the same rate, with the same absence of a source underneath it. The Danish Fashion Institute’s disavowal had not slowed anything down. A number does not need a citation to spread; it needs to be repeatable and alarming, and this one is both.
The New York Times ran its own trace in December 2018. Vanessa Friedman followed the claim back through the people and events most often credited with originating it: the 2015 documentary The True Cost, and the Copenhagen Fashion Summit, which has convened since 2008. Neither checked out as a source. Designer Eileen Fisher, whose team had repeated the line in public remarks, could not tell the Times where it came from either. Three independent investigations, run by three different outlets across three years, all ended at the same place: a claim with no author.
A peer-reviewed paper that had to fix its own numbers
The clearest evidence that this isn’t just a journalism story is that a scientific journal had to correct it too. Kirsi Niinimäki and coauthors published “The environmental price of fast fashion” in Nature Reviews Earth & Environment in 2020, a widely cited review of the industry’s footprint. The original text stated that fashion consumed 1.5 trillion litres of water a year and called it the second largest consumer of water and second largest industrial polluter. The journal later issued a formal author correction. The water figure changed from 1.5 trillion litres to 79 trillion litres, a correction of more than fifty times the original figure, and the ranking language came out entirely, replaced with the plainer claim that fashion is a major consumer of water. A journal’s permanent record now admits the number and the rank it supported were both wrong, in the same paper, at the same time.
The 8 percent figure that got pulled from publication
A separate statistic, that apparel and footwear account for 8 percent of global greenhouse gas emissions, has its own history and its own document behind it, which sets it apart from the rankless claim above: Quantis published it in a 2018 report called Measuring Fashion, produced with ClimateWorks Foundation. The number is higher than other published estimates, which cluster closer to 5 percent, and when Quantis was asked to provide the full accounting behind the 8 percent figure, it withdrew the report rather than defend the math in public. Quantis eventually republished Measuring Fashion with the same headline number, but without releasing the complete methodology that had been requested. ClimateWorks had its name taken off the study. An emissions estimate with a named consultancy, a named report, and a documented request for methodology that was not answered is a different kind of unreliable than a ranking nobody can trace at all, but it belongs in the same paragraph: cited constantly, defended rarely.
Both figures still describe an industry with a real and heavy carbon footprint; the dispute is about magnitude, not existence. Most articles that cite “8 percent” cite it as if it were the industry’s own confirmed disclosure, when it is closer to the high end of a contested range that its own authors could not fully document on request. A 3-point spread between competing estimates, on a base that large, is the same order of gap that made the water figure in the Nature Reviews correction worth a formal fix rather than a footnote.
What the industry’s own dyeing and waste numbers actually show
None of this is a case for leaving fashion alone. The figures that survive scrutiny, because they come from a named agency with a stated method rather than a viral rank, are bad enough to carry the argument without help.
The US EPA’s 2018 material-specific waste data found that textiles generated 17 million tons of municipal solid waste that year. Of that, 11.3 million tons went to landfill, 7.7 percent of everything landfilled in the country. Only 2.5 million tons were recycled, a 14.7 percent recycling rate, low next to categories like paper or metal that the same EPA report tracks in parallel. That is a federal agency’s own accounting of what happens to clothing once a household is done with it, and it does not need a “second worst industry on earth” framing to be a serious number. It needs the tonnage.
On the manufacturing side, the World Resources Institute’s summary of the industry’s footprint cites the Natural Resources Defense Council’s estimate that garment dyeing and finishing account for roughly 20 percent of global industrial water pollution, a specific process, tied to a specific body, rather than the whole industry lumped into an unranked total. The same summary carries the World Wildlife Fund’s widely used figure that a single cotton shirt takes about 2,700 litres of water to produce, close to what one person drinks over two and a half years. WRI also cites WWF on the input side of cotton farming: cotton uses about 3 percent of the world’s arable land but accounts for roughly 24 percent of global insecticide use and 11 percent of pesticide use, a lopsided ratio that has nothing to do with any industry ranking and everything to do with how one crop is grown. Every one of these numbers names the organization that produced it and the process it measured. None needed to claim a rank against the oil and gas industry to matter.
The distinction that keeps getting lost
Debunking a bad statistic and defending the industry it describes are not the same act, and the three journalists and one journal correction above never treated them as the same. Wicker’s piece, the Times investigation, and the Nature correction all land in the same place: the ranking claim is unsupported, and the underlying damage, measured properly, is real and large. Fashion does not need a stolen superlative to make the case against how much of it gets landfilled, or how much water goes into dyeing it, or how thin the accounting has been for its own carbon output. It has EPA tonnage, an NRDC water-pollution estimate, and a WWF per-garment figure doing that work already, each one attributable to an organization that will stand behind the number if you ask.
DRESS prints to order rather than against a forecast, which keeps one garment out of the 11.3 million tons the EPA counted going to landfill in 2018, and that’s a fact about our production model, not a claim about the industry’s rank against oil.
A statistic that traces to nobody is not evidence of anything except how far a number can travel once it sounds true enough to stop checking. The EPA’s 17 million tons did not need to travel; it was published, with a method, and it is still sitting there.