Vanilla and biodiversity loss: why global analyses don’t always tell the full story

- dr. DA (Dominic) Martin
- Assistant Professor
Much of our food travels vast distances, and its production can leave a heavy footprint on forests and biodiversity in countries far from our own. A recent global study suggested that internationally traded vanilla is a major driver of biodiversity loss in Madagascar. But new research from Wageningen University & Research challenges this conclusion, showing that the reality on the ground is more complex and has good news for vanilla-ice-cream lovers: farming vanilla may be far less damaging than it seems.
Global patterns, local realities
As food production increasingly takes place far from where food is consumed, environmental impacts are also displaced across borders. Recent global analyses have attempted to map this ‘outsourced’ biodiversity loss by combining data on trade, deforestation and species distributions. These studies highlight a striking pattern: countries in the Global North often drive more biodiversity loss abroad than within their own borders.
One of the most prominent examples identified in such analyses is vanilla production in Madagascar. According to that original study, vanilla exports to Europe and the United States were a key driver of biodiversity loss in this globally important hotspot.
What really drives deforestation in Madagascar
In a Matters Arising response published in Nature, researchers from Wageningen University & Research question this conclusion. Their analysis shows that the main driver of forest and biodiversity loss in Madagascar is not vanilla cultivation, but rice production through shifting cultivation. In this system, smallholder farmers clear forest, cultivate rice for a few years, and then leave the land fallow before moving on. When this cycle expands into forest areas, biodiversity is lost.
Vanilla plays a very different role. The crop is grown in a relatively small region in the north-east of Madagascar and is typically cultivated in agroforestry systems, where it is combined with shade trees. These systems can support relatively high levels of biodiversity and are often established on land that was previously used for shifting cultivation.
“It is important to understand what is really driving forest loss,” says Dr Dominic Martin, lead author of the study. “If we misidentify the causes, we risk designing policies that do more harm than good.”

Vanilla farmer in her agroforest. Many agroforests are established on land previously used for shifting cultivation and are therefore not causing deforestation and biodiversity loss.
Photo: Dominic Martin
Farmers at the centre of the challenge
At the same time, the researchers stress that smallholder farmers should not be blamed. “Farmers often face difficult trade-offs,” Martin explains. “Expanding shifting cultivation into forest areas can be a short-term survival strategy when alternatives are limited. But in the long term, they are also among those most affected by forest loss.”
This is where vanilla becomes part of the solution rather than the problem. As a valuable cash crop, it can provide farmers with income that reduces the need to clear forest for subsistence farming. In some cases, vanilla agroforestry even increases biodiversity compared to previously used agricultural land.
Why getting the story right matters
Correctly identifying the drivers of biodiversity loss is therefore crucial. If vanilla is wrongly portrayed as a major cause of deforestation, this could shift demand away from Madagascar. That, in turn, may reduce farmers’ income and increase pressure on forests. At the same time, it risks diverting attention from commodities such as cocoa, soy and palm oil, which are well-established drivers of deforestation in other regions.
Limits of global analyses
The study also highlights broader limitations of global-scale analyses. While such approaches are valuable for identifying large-scale patterns, they rely on assumptions and aggregated data that may not hold at local levels. As a result, conclusions can become less reliable when applied to specific regions or policy decisions.
“Engaging with local scientists and place-based research is essential,” says co-author Dr Sarobidy Rakotonarivo from the University of Antananarivo in Madagascar. “It helps ensure that global analyses reflect realities on the ground and avoid unintended consequences.”
Good news for vanilla ice-cream lovers, then. But an even more important lesson for science and policy. Biodiversity loss cannot be reduced to simple global patterns. It requires understanding local realities, acknowledging uncertainties, and working closely with those who know the landscape best.
Do you have a question?
Do you have a question regarding this topic, or do you see opportunities to collaborate with us? Please contact our expert.
Follow Wageningen University & Research on social media
Stay up-to-date and learn more through our social channels.

