Sustainability labels, such as Rainforest Alliance or Fairtrade, are often found on chocolate wrappers, bananas or coffee. They indicate to customers that certain sustainability criteria were met during the production of these goods. For example, that no rainforest was cleared for cocoa cultivation, that no children work in the fields, or that no toxic pesticides are used. To ensure that these promises are kept, auditors collect data on farms, such as the number of newly planted trees. This is intended to ensure that a farmer meets the criteria for certification.
Conflicts of interest and a lack of incentives
Federico Cammelli has spent several years studying certification schemes and environmental programmes in cocoa farming and draws a critical conclusion: "There are conflicts of interest throughout the entire value chain, and there is a lack of incentives to ensure that the data collected on cocoa farms is independently verified." Until February, Cammelli was a postdoctoral researcher working with Johan Six, Professor of Sustainable Agroecosystems at ETH Zurich and Rachael Garrett, Professor of Conservation and Development at the University of Cambridge. He now works as an independent researcher but is still affiliated with the University of Cambridge Conservation and Development Lab. He is the lead author of a study on the integrity of sustainability initiatives, which has just been published in the journal Science (DOI link).
The study shows that one in four entries collected for the sustainability monitoring program in cocoa farming had been manipulated. Changes were substantially higher when the auditors and farmers were aware of the target value required for the corresponding certification, providing evidence of intentional false reporting.
For the study, the researchers analysed monitoring data from 407 cocoa farms involved in a tree planting programme carried out throughout Côte d'Ivoire. All the farms were participating in a sustainability programme run by an international cocoa buyer and exporter. As part of the programme, smallholder farmers were invited to plant shade trees to ensure the long-term health of the cocoa trees and to increase biodiversity.
The study manipulated the monitoring system by rewarding farmers for providing accurate information about how many trees they had.
The farms were then audited. The auditors entered the number of trees they counted via smartphone into a system that immediately showed them whether their count matched the farmer initial count, and therefore the farmer would have earned a reward. Auditors also had the option to make corrections retrospectively.
'In cases where the criteria (i.e. the initial farmer tree count) was met on the first attempt, retroactive changes were made in only 6 to 7 per cent of instances,' says Cammelli. 'It was quite different where the criteria wase not met on the first attempt – in 33 per cent of farmers' cases, we observed subsequent changes.' For comparison, the researchers also checked whether other criteria, not relevant to the audit and subsequent reward system, had been altered. However, this was not the case. "The auditors therefore very specifically adjusted only the data relevant to passing the audit," the scientist concludes. While the auditors themselves do not receive a reward for accurate reporting, they have incentives to ensure farms pass the audit. For example, to maintain good relations with the farmers or share the prize with them.
Concealing the criteria reduces false reporting
The researchers also wanted to conduct an experiment to find out whether false reports could be reduced through simple adjustments to data collection. To this end, they divided the farms into two groups. In the first group, the auditors were aware of the parameters used to assess whether a farmer passed the audit (i.e. their initial tree count); in the second group, this information was withheld. "As a result, the rate of manipulated data fell by more than half, from 25 to 11 per cent," says Cammelli. If auditors are unaware of the criteria required to receive a premium under a sustainability programme or certification scheme, the incidence of misleading information during monitoring can be significantly reduced.
Highly relevant to the EU Deforestation Regulation
"The study comes at just the right time," says Rachael Garrett, co-author of the study. "Under the EU Deforestation Regulation, trading companies must ensure that their monitoring systems function effectively throughout the entire supply chain." The EU Deforestation Regulation will come into force from the end of December 2026 for large and medium-sized enterprises and from the end of June 2027 for small enterprises.
The regulation was developed as part of the European Green Deal. It requires all importers and sellers of soya, coffee, palm oil, beef, rubber, timber and cocoa in the EU to be able to demonstrate that the products were produced on land that was not deforested for that purpose after 2020. To provide this evidence, the boundaries of the cultivation areas are measured manually using geoinformation systems. The forest cover around the surveyed plots is then monitored via satellite.
"When collecting data on site, auditors face similar incentives to produce inaccurate documentation as those we described in our study," explains Garrett. Areas that have recently been cleared could be incorrectly documented during the initial manual mapping. "Our research data is highly relevant for the practical design of audits and for monitoring under the regulation," says Garrett. She is currently advising the UK's environmental protection agency and supporting the development of a monitoring system for the UK Illegal Deforestation Regulation.
Supply chain transformation needed
The study's authors agree that adjustments at individual cocoa farms are not sufficient to solve major sustainability problems, such as the deforestation of the tropical rainforest, or to eliminate incentives for manipulation. "To achieve this, the entire distribution of revenue along the supply chain would need to be transformed," says Johan Six. Cocoa cultivation is still based on a neo-colonial economic system that offers producers in Africa no true means of a sustainable livelihood. For this reason, cocoa processing – for example, into chocolate – must increasingly take place in the producing countries themselves, so that they receive a greater share of the profits along the value chain. "To this day, producers in Africa hardly benefit from the sustainability promises made to consumers in Europe," says Six, who has been working with cocoa farmers in West Africa for years. "This means there are very few incentives for them to invest in more sustainable production systems themselves."
Reference:
Cammelli F, Six J, Garrett RD: False reporting undermines the integrity of supply chain sustainability initiatives, Science, 13. August 2026, doi: 10.1126/science.aea9565
