Last month's US Supreme Court ruling in favour of the makers of the weedkiller Roundup (formerly Monsanto, now owned by German pharmaceuticals group Bayer ) upended thousands of lawsuits based on the cancer risk associated with the product's active ingredient glyphosate .
But the ruling also highlights a longstanding controversy about how the product's use is approved, including in New Zealand.
The US Environmental Protection Agency ( EPA ) has judged glyphosate " not likely " to cause cancer, while the World Health Organization's International Agency for Research on Cancer ( IARC ) describes it as " probably carcinogenic to humans ".
The two agencies weigh the evidence differently. IARC uses only publicly available studies and tests animal data for dose-response trends, while the US EPA leans heavily on unpublished studies commissioned by the manufacturer itself.
As a small country, New Zealand cannot itself conduct full risk assessments of thousands of chemicals in use, and leans on findings from bigger agencies overseas.
In the case of glyphosate, New Zealand's approval process is based on several reviews shaped by Monsanto. But I argue this is no longer tenable, as one of them has now been retracted and two are under investigation.
Compromised evidence in regulatory decisions
Monsanto-produced evidence has a questionable track record. Its early tests during the 1970s and '80s were conducted in contract laboratories that have since been convicted of fabricating data .
In 2017, litigation forced Monsanto's internal documents to be unsealed . These exposed how the company interacted with researchers who gave favourable assessments and produced ghostwritten reviews .
In this context, ghostwriting is the practice of putting names of independent scientists on work produced by employees of the company. It is explicitly prohibited under the standards of scientific practice .
Faced with the divided evidence, New Zealand's Environmental Protection Authority ( NZ EPA ) commissioned its own report in 2016. It concluded glyphosate is "unlikely to be genotoxic or carcinogenic to humans".
New Zealand scientists objected at the time. In 2018, six New Zealand public-health researchers, several of whom have connections to IARC, challenged the authority's report .
They listed many concerns, including reliance on unpublished industry studies and ghostwriting. None of this moved the authority's position. In 2024, it again found no grounds to reassess the use of glyphosate. A High Court challenge to this decision was unsuccessful .
However, the ground beneath this debate is shifting.
Three key Monsanto reviews under scrutiny
Over the past year, together with colleagues, I have traced how Monsanto-influenced reviews shape public discourse and regulation around the world.
We found they had been effective in their intended purpose of convincing people and regulators of glyphosate's safety.
We called for them to be retracted because the first was written by Monsanto and signed by outside academics. An internal email describes the practice of writing the sections so the named authors "would just edit & sign their names".
The journal retracted this review in November 2025 , noting Monsanto employees "may have contributed […] without proper acknowledgement as co-authors", among other issues.
Another two have recently come under investigation .
One was also shaped by Monsanto , with a company scientist who helped write it deliberately kept off the byline to make it look independent.
A recently disclosed memo shows the US EPA's own investigator reached the same conclusion that the paper "hid Monsanto's role as an author", but the agency kept citing it.
Another review published in 2015 discredited a study that had found glyphosate caused a significant rise in malignant lymphoma in mice, instead invoking a " viral infection ". But subsequent independent reanalysis found no evidence of this in the original study.
Bayer continues to reject the ghostwriting characterisation , saying Monsanto's involvement was appropriately cited in the acknowledgments which thanked the company's scientists for "significant contributions".
NZ should reconsider glyphosate regulation
Institutional trust varies significantly. US EPA emails from 2025 show Bayer promising the agency " a small thanks " for updating its glyphosate webpage.
By contrast, the judge in Australia's 2024 Roundup class action warned that a company dealing with regulators cannot be assumed to "play with a straight bat," and said the evidence gave him "concerns about accepting anything from Monsanto at face value". Yet the 2015 review was used by the applicant's own statistician.
New Zealand's 2016 report also leans on these three reviews. But now, in 2026, it is much harder to treat them as legitimate evidence in an honest scientific debate about glyphosate's safety.
When New Zealanders are told dietary exposure sits well under 1% of the " acceptable level ", they should keep in mind two things.
First, this level is based on industry-sponsored studies. Second, an average hides how unevenly exposure falls on sprayers, rural communities and people with particular diets. These factors should be weighted in, but New Zealand holds almost no exposure data .
Glyphosate is only one of 16 high-use pesticides in New Zealand flagged as suspected carcinogens by at least one major agency.
The costs of using them should be estimated explicitly: an uptick in illnesses, strain on the healthcare system, harm to the environment and lost quality of life.
The intellectually honest process would be to assess risks with broad scientific consultation and acknowledgement of where evidence is compromised or plainly missing.
For New Zealand to keep avoiding a comprehensive reassessment of glyphosate is to make wilful ignorance a matter of national policy.
Sasha (Alexander) Kaurov is affiliated with Motu Economic & Public Policy Research and his research on this topic has been partially funded by the Rockefeller Family Fund. He co‑authored an unpaid amicus brief in the US Supreme Court case involving Bayer.
