Zero-deforestation commitments and Brazilian soy


  • Coverage of soy traded under a zero-deforestation commitment is very uneven between regions in Brazil.
  • Less than half of soy exported from the Cerrado biome in 2016 was covered by any zero-deforestation commitment.
  • Between 2006-2016 soy traders with zero-deforestation commitments have been associated with similar levels of deforestation risk as non-committed companies.
  • Over the last decade, signatories of the Amsterdam Declaration were exposed to similar or higher levels of relative deforestation risk – than major consumer markets such as China, with no discernible decline in deforestation risk since the declaration came into force in 2015.


Companies and countries sourcing forest-risk commodities have come under increasing scrutiny and public pressure to eliminate deforestation in their supply chains. As a result, a growing number – more than 100 to date – have made public commitments to do so. These commitments are often referred to as zero-deforestation commitments (ZDCs).

Making ZDCs is ambitious and commendable. However, most ZDCs contain little information about how they will be implemented on the ground, or how progress will be measured.

Given the complexity and opacity of the supply chains involved, the impact of individual ZDCs remains difficult to track: it remains unclear whether they are leading to real reductions in deforestation.

Trase data offer a new opportunity to track the success of ZDCs, creating the measures of environmental performance by linking actors to places, and combining supply chain data with local indicators. This chapter provides the first systematic insights into the impact of ZDCs in relation to Brazilian soy.


Zero-deforestation commitments in the Brazilian soy sector

One of the first ZDCs was made in 2006, when major soy traders responsible for around 90% of all the soy sourced from the Brazilian Amazon signed up to the Soy Moratorium – a commitment not to purchase soy planted on land in the Amazon deforested after 2008.

The Soy Moratorium was a landmark, multi-stakeholder initiative which brought together companies, environmental NGOs and the Brazilian government. It combined an independent, robust monitoring system and powerful enforcement mechanism – through market exclusion of non-compliant soy. Together with strong-handed efforts by the the Brazilian government to reduce Amazonian deforestation, the Moratorium helped to reduce the proportion of soy planted on recently deforested land in that biome from 30% in 2006 to around 1% by 2014.

However, despite its success in helping to reduce the direct conversion of Amazon rainforest to soy plantations, the Soy Moratorium has several limitations. The most obvious is that most soy cultivation in Brazil takes place not in the Amazon, but in the neighbouring Cerrado, a highly biodiverse savannah ecosystem that is outside of the scope of the Moratorium.

Most soy production and soy-associated deforestation occurs outside the Amazon.


By not covering the Cerrado, the Moratorium therefore does not address most soy-associated deforestation and habitat conversion (including the clearance of large areas of savannah and shrubland).

The Soy Moratorium may even exacerbate soy-associated land conversion in the Cerrado, by displacing soy expansion away from the Amazon.

Most recent soy expansion has been in the Cerrado.


Zero-deforestation commitments are increasing in coverage, but not evenly

Given the limitations of biome-specific ZDCs like the Soy Moratorium, it is encouraging that an increasing number of companies and countries are making commitments to eliminate deforestation across their entire supply chains, and beyond the Amazon.

In the soy sector, four major traders – Cargill, Bunge, Archer Daniels Midland (ADM) and Amaggi, together responsible for almost half of Brazilian soy exports between 2006 and 2016 – have each recently made ZDCs that encompass their entire supply chain, though there remains a lack of clarity regarding their scope.

These commitments, together with the Soy Moratorium in the Amazon mean that a growing proportion of soy is now traded under some form of ZDC. In 2016, 28 million tonnes – 42.2% of all soy exports from Brazil  – were covered by ZDCs made by soy traders.


The New York Declaration on Forests was signed in 2014 by governments, companies and civil society. Its main goal is to halve the rate of deforestation by 2020. Photo credit: istock.com/franckreporter

Many more consumer-facing companies have also signed up to zero deforestation agreements such as the 2014 New York Declaration on Forests, including Walmart, McDonalds, and L’Oreal, who all manufacture and sell products which contain soy, or soy-fed meat.

When it comes to countries, seven European countries (Denmark, France, Germany, the Netherlands, Norway, the UK and Italy) – which together were the destination for 22% of soy exports from Brazil in the last decade – signed the Amsterdam Declarations in 2015 to promote the elimination of deforestation from agricultural commodity supply chains.

Even with the rise in global ZDCs, the geographic coverage of ZDCs remains highly uneven. While almost all the soy exported from the Amazon is covered by the Soy Moratorium, less than half of soy exported from the Cerrado biome in 2016 was covered by any ZDC.

These volumes reflect the soy exported by companies who are signatories of the Soy Moratorium, plus the four major traders who have publically available, written commitments: Cargill, Bunge, ADM and Amaggi.


This is particularly problematic because without strong action, there is a very real risk that the Cerrado will suffer further widespread deforestation. While only a small fraction of total deforestation in the Cerrado can be directly attributed to soy, this proportion is much higher in some areas. In the frontier region of Matopiba, at least 37% of soy expansion between 2005 and 2016 was through direct conversion, and 20% of the total area cleared since 2000 is now under soy (see also Chapter 5).

Further, the coverage of ZDCs in the Cerrado may well be even lower than we estimate. It is often not clear whether zero-deforestation commitments are intended to include biomes such as the Cerrado and Gran Chaco that are densely forested in some places but also include more open savannah-tree formations in others. If these commitments exclude these important biomes, their ability to foster a sustainable soy sector would be significantly weakened.

Are zero-deforestation commitments already having an impact?

If ZDCs are working, then we would expect that the rate of soy-associated deforestation would be low, or at least going down, in areas where the actors that have made commitments are sourcing large volumes of soy.

Trase allows, for the first time, the monitoring of deforestation risks in the areas where different companies are active. By linking all of Brazil’s soy exporting municipalities to the companies and countries that trade with them, it is possible to estimate the risk that they are exposed to deforestation in their sourcing regions – their overall ‘deforestation risk’.

For the period 2006-2016 soy traders in the Brazilian market with zero-deforestation commitments have been associated with similar levels of deforestation risk as companies that have not made such commitments. During this period the four traders that have made commitments (ADM, Bunge, Cargill, and Amaggi) were exposed to a deforestation risk of 326,000 hectares of direct soy-associated deforestation and land conversion. That means their exports of soy were potentially associated with deforestation of an area more than twice the size of London across the Amazon and Cerrado in Brazil. This is two-thirds of the deforestation risk across all traders.

These companies are also exposed to some of the highest relative deforestation risk per tonne of exported soy in both biomes. In fact, until 2015, these companies were exposed to higher deforestation risk in Brazil than all non-committed companies put together (see also Chapter 5).

While it is too early to draw firm conclusions about the impact of the ZDCs these companies have made – Cargill in 2014, ADM and Bunge in 2015, and Amaggi in 2017 – it is encouraging to see that collectively their level of exposure to deforestation risk has been falling since 2012. Whilst no conclusions can be drawn from a simple visual inspection of the trends it is perhaps promising to see that their exposure to relative deforestation risk fell below that of non-committed companies – and therefore the rest of the market – for the first time in 2015.

Of course, this change in pattern may well also reflect an increase in deforestation risk exposure among non-committed companies (see Figure 6.5) – which highlights the need to encourage all companies, big and small, new and old, to make commitments and monitor their supply chains.

In either case, the combined direct soy-deforestation risk exposure of these committed companies still exceeded 9,000 ha in 2016 (45% of the total), suggesting that companies with ZDCs have a long way to go to reassure their downstream customers that their soy is truly deforestation-free.

Similarly, the Amsterdam Declaration was signed only at the end of 2015, making it too soon to confidently assess what impact it may be having. But Trase data can be used to monitor trends and assess its future performance. Over the last decade, signatories were exposed to similar or higher levels of deforestation risk as other EU countries and major markets like China, with no dramatic decline in deforestation risk among signatories since the declaration was signed.

Despite the stark challenges that ZDCs face – the New York Declaration on Forests and Amsterdam Declarations, for example, target an end to commodity-associated deforestation by 2020, which is now just around the corner – it should not be forgotten that these commitments are new and ambitious. They can be a powerful tool for achieving sustainable commodity production.

But to do so they must be strengthened and enforced. Systemic weaknesses that need to be addressed in current commitments include: the need to move from zero-net deforestation targets, which allow reforestation to compensate for forest loss, to zero-absolute deforestation; ambiguity regarding the types of habitat encompassed by commitments, and the importance of explicitly including wooded savannahs and other natural habitats that fall under commonly agreed definitions of forests (e.g. by the FAO); the need for annual targets to ensure year-on-year improvements; a need to encompass the entire supply chain, not just direct suppliers; and a need to broaden implementation strategies beyond a strong reliance on certification programs which place an undue burden on producers. A broader challenge influencing the effectiveness of commitments is the need for transparency in how deforestation risks linked to supply chains change over time.

Whatever their scope, as these ZDCs evolve, transparency initiatives like Trase will be critically important for assessing progress, identifying priorities for greater investment, and strengthening accountability for both companies and governments.

In this yearbook, we have focused on soy in South America – for the most part, in Brazil. In future, Trase will expand cover to other commodities and geographies in more detail, where it can similarly be used to monitor the risks of actors, performance of sustainability commitments, and ultimately support the transition to the sustainable, deforestation-free production of all major tropical commodities.