Drivers of Deforestation

The world loses 5 million hectares of forest to deforestation each year. What activities are driving this?

Related topics

Palm Oil

Explore palm oil production across the world and its impacts on the environment.

Deforestation and Forest Loss

Explore long-term changes in deforestation and deforestation rates across the world today.

Cutting down forests: what are the drivers of deforestation?

Every year, the world loses around 5 million hectares of forest. 95% of this occurs in the tropics. At least three-quarters of this is driven by agriculture – clearing forests to grow crops, raise livestock, and produce products such as paper.1

If we want to tackle deforestation, we need to understand two key questions: where we’re losing forests and what activities are driving it. This allows us to target our efforts towards specific industries, products, or countries where they will have the greatest impact.

In a study published in Global Environmental Change, Florence Pendrill and colleagues addressed both of these questions.2 They quantified how much and where deforestation occurs from the expansion of croplands, pasture, and tree plantations (for logging) and what products are grown on this converted land. They also combined this with global trade flows to assess how much of this deforestation was driven by international trade – we look at the role of trade specifically in a related article.

Here, we’ll look at both where tropical deforestation is happening and what products are driving it.

Brazil and Indonesia account for almost half of tropical deforestation

The study by Pendrill et al. (2019) found that, between 2005 and 2013, the tropics lost an average of 5.5 million hectares of forest per year to agricultural land. That was over a decade ago, but the world is still losing a similar amount today: using satellite data, researchers at Global Forest Watch estimate that annual global deforestation is around 4 million hectares.3 95% of this was in the tropics. But where in the tropics did we lose this forest?

In the chart, we see the share of tropical deforestation by country and region. It's measured as the annual average between 2010 and 2014.

One-third of tropical deforestation happened in Brazil. That was 1.7 million hectares each year. The other single country where large forest areas are lost is Indonesia – it accounted for 14%. This means around half (47%) of tropical deforestation occurred in Brazil and Indonesia. Again, if we look at more recent satellite data, we find that this is still true today: annual deforestation is over 4 million hectares, with Brazil and Indonesia accounting for the majority of it. The expansion of pasture for beef production, croplands for soy and palm oil, and, increasingly, the conversion of primary forest to tree plantations for paper and pulp have been the key drivers of this.

The expansion of pasture lands have also had a major impact on land use in the rest of the Americas – outside of Brazil, Latin America accounted for around one-fifth of tropical deforestation.

The expansion of agricultural land in Africa accounted for around 17.5% of tropical deforestation. This may slightly underestimate the loss of forests in Africa for two reasons. Much of Africa’s deforestation has been driven by subsistence agricultural activities, which are not always fully captured in national statistics. Secondly, depending on the permanence of agricultural activities such as slash-and-burn farming, some of this forest loss might be classified as temporary forest degradation rather than permanent deforestation.

Click to open interactive version

Beef, soy, and palm oil are responsible for 60% of tropical deforestation

If we want to tackle deforestation, we also need to know what causes it. That allows us to avoid the foods that drive deforestation or innovate the ways we produce them.

In the chart here, we see the breakdown of tropical deforestation by the types of agricultural production.

Beef stands out immediately. The expansion of pasture land to raise cattle was responsible for 41% of tropical deforestation. That’s 2.1 million hectares every year – about half the size of the Netherlands. Most of this converted land came from Brazil; its expansion of beef production accounts for one-quarter (24%) of tropical deforestation. This also means that most (72%) deforestation in Brazil is driven by cattle ranching.4 Cattle in other parts of Latin America – such as Argentina and Paraguay – also accounted for a large amount of deforestation – 11% of the total. Most deforestation for beef, therefore, occurs in Latin America, with another 4% happening in Africa.

Palm oil and soy often claim the headlines for their environmental impact. They are categorized as ‘oilseeds,’ which also include a range of smaller commodities such as sunflower, rapeseed, and sesame. They drove 18% of deforestation.

Here we see that Indonesian palm oil was the biggest component of this. In neighboring Malaysia, the expansion of oil seeds was also a major driver of forest loss. Soybeans are the most common oilseed in Latin America. While many people immediately think of food products such as tofu or soy milk, most of the global soybean production is used as feed for livestock or biofuels. Just 6% is used for direct human food. The impact of soy production is one we look at in more detail in a related article.

Combined, beef and oilseeds account for nearly 60% of deforestation.

If we add the third largest driver – forestry products, which is dominated by paper but also includes timber – then we cover almost three-quarters. Across Europe and North America, forestry products mainly come from managed plantation forests that have been established for a long period of time or are grown on previously unforested land. This is different from most tropical countries where forestry products also come from the logging of primary rainforests or their replacement with plantations. This destroys primary rainforests and, as shown in the chart, has been an important driver of deforestation in Indonesia and elsewhere in Asia.


We can tackle a lot of deforestation by focusing on a few key supply chains

If almost three-quarters of tropical deforestation is driven by the production of a few key products – beef, soybeans, palm oil, and paper – then we can achieve a lot by focusing our efforts on these supply chains.

There are some signs that progress is possible. Soybean production in Brazil was once also an important driver of deforestation in the Amazon region.5 In 2006, under pressure from retailers and NGOs, the world’s major soybean traders signed Brazil’s Soy Moratorium (SoyM) – the world’s first voluntary zero-deforestation agreement. Traders agreed that they would not purchase soy that was grown on deforested lands in the Brazilian Amazon after July 2006. Overall, it was a success: in the two years before the agreement, 30% of soybean expansion in the region came at the expense of forest; afterwards, deforestation declined dramatically, and by 2014 only 1% of expansion was turning forests into land for oilseed production.6,7

But, as we show on our page on the impact of soy, there are also lessons to learn about how to implement these commitments more effectively. There is evidence that while the moratorium reduced deforestation rates in the Brazilian Amazon, some of this deforestation may have ‘leaked’ to neighboring regions. Soybean production has shifted from the Amazon to the Cerrado region south of the Amazonas, often at the expense of forests there.8 This suggests that zero-deforestation agreements can be effective but must be considered in the wider context of how they shape forest and agricultural changes elsewhere. To combat this, researchers have suggested that the SoyM be expanded to include not only the Amazon but also regions such as the Cerrado.6

If we can take similar action in the other industries – beef, palm oil and paper – then there is the potential to cut out a large share of deforestation today.

Looking to the future, a shift in focus towards sub-Saharan Africa looks likely. The demands for increased agricultural production in Africa are going to be large and could come at the cost of forests.9 Solutions there will have to focus on major improvements in crop yields so African farmers can produce more food without increasing the amount of land they need to do so.

Alternative ways of making high-quality protein could also be transformative. Beef is the leading driver of deforestation, and the demand for meat across the world will continue to grow in the coming decades. Technological innovations in meat substitute and cultured meat products would allow people to continue eating meat-like products without the destruction of tropical forests that come with it.

Is our appetite for soy driving deforestation in the Amazon?

Soy has earned itself a bad reputation with many consumers. Its links to deforestation mean that, alongside palm oil, soy has become a product to avoid. Is this reputation justified?

In this article, we will take a look at the story of soy: how production has changed over time, where it is produced, what it is used for, and whether it really has been a key driver of deforestation. Although the research suggests that by far the largest driver of deforestation in the Brazilian Amazon has been driven by the expansion of pasture land for beef production, soy is likely to have played at least some role in the loss of forest.

More than three-quarters (77%) of global soy is fed to livestock for meat and dairy production. Most of the rest is used for biofuels, industry or vegetable oils. Just 7% of soy is used directly for human food products such as tofu, soy milk, edamame beans, and tempeh. The idea that foods often promoted as substitutes for meat and dairy – such as tofu and soy milk – are driving deforestation is a common misconception.

How has demand for soy changed over time?

Global soy production has exploded over the past 50 years. Global production today is more than 13 times higher than it was in the early 1960s. Even since the year 2000, production has more than doubled.

In the chart here, we see the change in global soybean production from 1961 onwards.10 Back in the 1960s, we were producing 20 to 30 million tonnes per year. This is now around 350 million tonnes.

We can increase agricultural production in two ways: by improving yields (growing more on a given plot of land) or by expanding the amount of land we use. As we will see later, although countries have seen yield gains over time, much of the increase in production has been driven by the expansion of croplands. Unfortunately, some of this has been at the expense of forests.

Click to open interactive version

Which countries produce the most soy?

Global soy production has increased rapidly over the past 50 years. But to understand whether this has come at a major environmental cost, we also need to understand where this growth has come from.

Which countries are the main producers of soy?

In the map we see the distribution of soybean production across the world. Most of the world’s soy comes from only two countries: the US and Brazil. Combined, they account for more than two-thirds (69%) of global soy production.11 In fact, they produce almost exactly the same amount: in 2018, the US produced 123 million tonnes, and Brazil 118 million tonnes. Individually, they each account for around one-third of global production. The other major producer is Argentina, which accounts for 11% (at 40 million tonnes).

To understand soy as a potential driver of deforestation, it’s also useful to understand how each country’s production has changed over time. It is the change in production, and how this was achieved (either increased yields or cropland expansion), that is the potential driver of deforestation.

In the other chart we see the change in soy production in Brazil and the US. The US was already producing a lot of soy throughout the 1960s, 70s and 80s and so its growth in recent decades has been much slower than Brazil. The US was growing 20 million tonnes per year as early as 1960. Brazil did not reach this level of output until 1990 – three decades later. This means that its rate of growth in the last 30 years has been much faster.

Click to open interactive version

More than three-quarters of global soy is fed to animals

Before we look at the evidence for whether Brazilian soy is responsible for the clearing of the Amazon, we should first understand what products have been driving this growth.

When someone mentions soy, we often think about foods such as tofu, soy milk, tempeh, or edamame beans. This feeds into the argument that meat and dairy substitutes – such as switching from meat to high-protein tofu or from dairy to soy milk – is in fact worse for the environment. But, only a small percentage of global soy is used for these products. More than three-quarters (77%) of soy is used as feed for livestock.

In the chart, we see the breakdown of what the world’s soy was used for in 2018. On the left, we have total global soy production; in the middle, the three categories of uses (direct human food, animal feed, and industrial processes); and on the right, we have the end-use products. This data is sourced from an analysis published by the University of Oxford’s Food Climate Research Network (FCRN), which relies on the USDA’s PSD database.12 Over one-third (37%) of global soy is fed to chickens and other poultry, one-fifth to pigs, and 6% to aquaculture. Very little soy is used for beef and dairy production – only 2%.

One-fifth of the world’s soy is used for direct (i.e. not from meat and dairy) human consumption. Most of this is first processed into soybean oil. Typical soy products such as tofu, soy milk, tempeh, and edamame beans account for just 7% of global demand.

Soy can also be used for industrial purposes. Around 4% is used for biofuels, lubricants and other industrial processes. Biodiesel alone accounts for 2.8%.


We might therefore conclude that the increased demand for soy has been driven by a growing appetite for meat, dairy and soybean oil. But to double-check we should look beyond this static single-year view and see how demand has changed over time. Maybe demand for these products has always been high, and instead the growth in demand has come from the increased popularity of products such as soy milk and tofu.

In the chart we see the allocation of soybeans to three categories. Processed products include all animal feed from soybean cake (i.e. 70% of global demand); soybean oil; and industrial products such as biofuels. Direct human food includes all non-animal-sourced foods from soy excluding oils. Direct animal feed is soybeans fed directly to livestock (rather than been processed into soybean cake first).

We see that most of this growth has come from the increased demand for processed soy – animal feed, biofuels, and vegetable oil. This rise has been particularly steep since 1990. By 2013, it had increased from 88 million to 227 million tonnes. Over this period, demand for human food products such as tofu and soy milk increased by only 3 million tonnes (from 7.4 to 10.7 million).

This should not surprise us. Global meat production has more than tripled over the last 50 years. This increase has been most marked for poultry – the largest consumer of soy feed.

Using the ‘change country’ toggle, you can see how this allocation varies by country. For example, we see that this distribution is similar to Brazil's.

Click to open interactive version

Is soy production driving deforestation?

So far we’ve established that demand for soy has increased rapidly over the past 50 years; most soy is produced in the US, Brazil (and to a lesser extent, Argentina); and most of this has been driven by increased demand for animal feed, biofuels and vegetable oils.

The big question is whether this has been a key driver of deforestation.

The first step to answering this is to understand whether the amount of land we use to grow soy has increased. If improvements in crop yields were able to keep up with this rise in production, then we wouldn’t need any additional land, and there would be no need to cut down forests.

In the chart we see the change in soy production, yields, and area harvested. This tells us the percentage change in each of these variables relative to the first year shown. Here the starting year is 1961 but you can change this by adjusting the time slider at the bottom of the chart.

Crop yields have not been able to keep up with production. Since 1961 global yields increased by 150%. But production increased by 1200%. This means the area used to grow soy has more than quadrupled.

This is also true for Brazil [you can see this using the “Change country” toggle]. Since Brazilian soy production was very low in 1961, it makes sense to adjust the time slider to see the change since 1980. Soy yields have doubled since 1980. This is impressive but not enough to keep up with demand: soy production increased by 680%. Instead, Brazil has had to devote more and more land to soy production: land use has tripled since 1980.

Click to open interactive version

Brazil has devoted increasing amounts of land to soy production. But has this come at the expense of forests? Has the Amazon been cleared to make room for it?

This is a question many researchers have tried to answer. What makes it complex is that agricultural and land use systems are intricately linked: it’s possible to assess whether soy production has a direct impact on deforestation, but much harder to understand whether it is indirectly causing harm elsewhere.

In a study published in Science, Alexandra Tyukavina et al. (2017) look at the drivers of forest loss in the Brazilian Legal Amazon.13 The change in these drivers from 2000 to 2013 is shown in the chart. Note that this overall trend suggests a major decline in forest loss since 2000 – unfortunately, in the years since 2013, rates have increased again.

From this, we would conclude that the dominant driver of deforestation in the Brazilian Amazon was the expansion of pasture for beef production. If we look at forest loss from commercial crops – which is mainly soybeans – we see a significant decline, especially following the introduction of ‘Brazil’s Soy Moratorium’. The ‘Soy Moratorium’ was a campaign involving civil agencies and soybean companies, which stipulated that farmers who grew soy on illegal or legal deforestation areas would not be able to sell them to suppliers. Since 2009, satellite imagery has been used to help identify soybean crops being grown on deforested areas.14

Numerous studies have reached a very similar conclusion: it is pasture, not soybean production, that is driving most of the deforestation in the Brazilian Amazon.15 But, this only looks at the direct drivers of deforestation. In other words, the cutting down of forest today to make space for cropland for soy production.

Click to open interactive version

The indirect impacts of soy production

Soy may no longer be a direct driver of deforestation in the Brazilian Amazon. But we should also think about the indirect impacts of increased production. We know that the area being used to grow soy is still increasing and that land has to come from somewhere. In particular regions, such as Mato Grosso, studies have found that instead of replacing forest, these croplands are replacing pasture.16 If this pasture land is simply shifting into forested areas, we could argue that soy is still a major underlying cause of deforestation.

A recent study by researchers Nikolas Kuschnig, Jesús Crespo Cuaresma, and Tamás Krisztin reached this conclusion.17 They combined high-resolution satellite imagery from the Mato Grosso region in Brazil with socioeconomic panel data to quantify not only the direct impacts of soy production but also the spillover effects. The results suggest that by only looking at direct impacts, we underestimate the role that soy continues to play in deforestation.

It’s also the case that the Amazonas region in Brazil gets most of the attention. But most soy is now grown in other regions of the country. In 2015, only 13% came from the Amazon, while 48% came from the Cerrado region.18 Some researchers, therefore, make a strong case that interventions such as the ‘Soy Moratorium’ need to be extended to cover regions beyond the Amazon if they are to be effective. Without wider implementation of these policies, we will continue to see deforestation simply shift elsewhere.

Whilst the expansion of pasture for beef production is the leading driver of deforestation in Brazil, soy still plays a significant role when we take its indirect impacts into account. There are a couple of key actions we can take to end deforestation. For consumers, since most deforestation is driven by expanding pastures for beef or soy to feed poultry and pigs, reducing meat consumption is an effective way to make a difference. For companies and regulators, zero-deforestation policies must be more widely implemented (i.e., not only focused on the Amazon) and must be more carefully designed to take spillover effects into account.


  1. There are a number of estimates on what share of deforestation is driven by agriculture. The differences in these estimates are often explained by the different treatments of deforestation versus forest degradation. We cover the differences between these types of forest loss on a related topic page. Agriculture accounts for 70-80% of tropical deforestation – the permanent conversion of forested land to another land use. It accounts for a smaller percentage when degradation – the temporary loss of forest prior to regrowth – is included.

    One of the most widely cited studies on this comes from Noriko Hosonuma et al. (2012), who estimates that 73% of tropical deforestation is driven by agriculture. This is similar to estimates by Geist and Lambin (2002), who estimated that around 80% of deforestation in the 1980s and 1990s was driven by agriculture. In fact, over the longer analysis of deforestation from 1840 to 1990, they estimate that 96% of deforestation was driven by agriculture. Gibbs et al. (2010) estimate that during the 1980s and 1990s, 83% of agricultural land expansion replaced forest.

    Geist, H. J., & Lambin, E. F. (2002). Proximate Causes and Underlying Driving Forces of Tropical Deforestation. BioScience, 52(2), 143-150.

    Gibbs, H. K., Ruesch, A. S., Achard, F., Clayton, M. K., Holmgren, P., Ramankutty, N., & Foley, J. A. (2010). Tropical forests were the primary sources of new agricultural land in the 1980s and 1990s. Proceedings of the National Academy of Sciences, 107(38), 16732-16737.

    Hosonuma, N., Herold, M., De Sy, V., De Fries, R. S., Brockhaus, M., Verchot, L., ... & Romijn, E. (2012). An assessment of deforestation and forest degradation drivers in developing countries. Environmental Research Letters, 7(4), 044009.

  2. Pendrill, F., Persson, U. M., Godar, J., Kastner, T., Moran, D., Schmidt, S., & Wood, R. (2019). Agricultural and forestry trade drives a large share of tropical deforestation emissions. Global Environmental Change, 56, 1-10.

  3. The Global Forest Watch program categorizes forest loss drivers based on permanent deforestation – the conversion of forest to another land use – and degradation (which includes logging of tree plantations and wildfires). ‘Commodity-driven deforestation’ – which includes some activities such as mining but is predominantly agricultural commodities – totaled over 4 million hectares in recent years.

    A paper by Philip Curtis et al. (2018) discusses this classification in detail. We also look at these categories in more detail in a related article.

    Curtis, P. G., Slay, C. M., Harris, N. L., Tyukavina, A., & Hansen, M. C. (2018). Classifying drivers of global forest loss. Science, 361(6407), 1108-1111.

  4. Since Brazil accounts for 33% of tropical deforestation, and Brazilian cattle account for 24%, cattle account for 72% of Brazil’s total [24 / 33 * 100 = 72%].

  5. Tyukavina, A., Hansen, M. C., Potapov, P. V., Stehman, S. V., Smith-Rodriguez, K., Okpa, C., & Aguilar, R. (2017). Types and rates of forest disturbance in Brazilian Legal Amazon, 2000–2013. Science Advances, 3(4), e1601047.

  6. Gibbs, H. K., Rausch, L., Munger, J., Schelly, I., Morton, D. C., Noojipady, P., ... & Walker, N. F. (2015). Brazil's soy moratorium. Science, 347(6220), 377-378.

  7. Boucher, D., Roquemore, S., & Fitzhugh, E. (2013). Brazil's success in reducing deforestation. Tropical Conservation Science, 6(3), 426-445.

  8. Kuschnig, N., Crespo Cuaresma, J., & Krisztin, T. (2019). Unveiling Drivers of Deforestation: Evidence from the Brazilian Amazon.

  9. Clark, M. A., Williams, D. R., Buchanan, G. M., Ficetola, G. F., Rondinini, C., & Tilman, D. (2020). Proactive conservation to prevent habitat losses to agricultural expansion. Nature Sustainability.

  10. This data is sourced from the UN Food and Agriculture Organization (FAO).

  11. In 2018, global soy production was 349 million tonnes. The US produced 123 million tonnes [123M / 349M * 100 = 35%], and Brazil produced 118 million tonnes [118M / 349M * 100 = 34%]. Combined, they accounted for 69% of global production.

  12. United States Department of Agriculture. PSD Online. Available at:{/ref}

    The majority (77%) of the world’s soy is fed to livestock for meat and dairy production. 7% is fed directly to animals as soybeans, but the remainder is first processed into soybean ‘cake’.{ref}Soybean cake (sometimes referred to as soybean meal) is a high-protein feed made from the pressurization, heat treatment, and extraction processing of soybeans. The oil is extracted from the soybeans to leave a protein-rich product.

  13. Tyukavina, A., Hansen, M. C., Potapov, P. V., Stehman, S. V., Smith-Rodriguez, K., Okpa, C., & Aguilar, R. (2017). Types and rates of forest disturbance in Brazilian Legal Amazon, 2000–2013. Science Advances, 3(4), e1601047.

  14. Rudorff, B. F. T., Adami, M., Aguiar, D. A., Moreira, M. A., Mello, M. P., Fabiani, L., ... & Pires, B. M. (2011). The soy moratorium in the Amazon biome monitored by remote sensing images. Remote Sensing, 3(1), 185-202.

  15. Brandão, ASP, de Rezende, GC, Costa Marques, RW, & de Aplicada, IPE (2005). Agricultural growth in the period 1999-2004, explosion of the area planted with soybeans and the environment in Brazil.

    Müller, C. (2003). Expansion and modernization of agriculture in the Cerrado–the case of soybeans in Brazil’s center-West. Brasília: Departamento de Economia, Universidade de Brasília.

    Barona, E., Ramankutty, N., Hyman, G., & Coomes, O. T. (2010). The role of pasture and soybean in deforestation of the Brazilian Amazon. Environmental Research Letters, 5(2), 024002.

  16. Barona, E., Ramankutty, N., Hyman, G., & Coomes, O. T. (2010). The role of pasture and soybean in deforestation of the Brazilian Amazon. Environmental Research Letters, 5(2), 024002.

  17. Kuschnig, Nikolas & Crespo Cuaresma, Jesus & Krisztin, Tamás, 2019. "Unveiling Drivers of Deforestation: Evidence from the Brazilian Amazon," Ecological Economic Papers 32, WU Vienna University of Economics and Business.

  18. Soterroni, A. C., Ramos, F. M., Mosnier, A., Fargione, J., Andrade, P. R., Baumgarten, L., ... & Carvalho, A. X. (2019). Expanding the soy moratorium to Brazil’s Cerrado. Science Advances, 5(7), eaav7336.

Cite this work

Our articles and data visualizations rely on work from many different people and organizations. When citing this article, please also cite the underlying data sources. This article can be cited as:

Hannah Ritchie (2021) - “Drivers of Deforestation” Published online at Retrieved from: '' [Online Resource]

BibTeX citation

    author = {Hannah Ritchie},
    title = {Drivers of Deforestation},
    journal = {Our World in Data},
    year = {2021},
    note = {}
Our World in Data logo

Reuse this work freely

All visualizations, data, and code produced by Our World in Data are completely open access under the Creative Commons BY license. You have the permission to use, distribute, and reproduce these in any medium, provided the source and authors are credited.

The data produced by third parties and made available by Our World in Data is subject to the license terms from the original third-party authors. We will always indicate the original source of the data in our documentation, so you should always check the license of any such third-party data before use and redistribution.

All of our charts can be embedded in any site.