Amazon soy pact collapse points to 1.4 million hectares of new deforestation by 2036

The Amazon soy moratorium, the private-sector pact credited with sparing millions of hectares of rainforest over two decades, expired on 16 July 2026 without renewal. Modelling published the next day warns that its absence will unleash at least 1.4 million hectares (3.5 million acres) of additional deforestation in Brazil by 2036, releasing carbon emissions equivalent to Canada's annual fossil-fuel output.

The numbers convert a quiet administrative lapse into a structural climate event. The moratorium is not a treaty and has no enforcement arm; it is a commercial agreement among traders, retailers and farmers to stop buying soy grown on land cleared after 2008 in the Amazon biome. Its collapse exposes a familiar pattern: the heavy lifting of conservation has been done by supply-chain contracts, not by statute, and the contracts age faster than the forests they protect.

What the moratorium actually did

Signed in 2006 and renewed several times, the soy moratorium was a response to a previous deforestation surge in Mato Grosso and Pará, where satellite data had shown that soy expansion was the single biggest proximate driver of forest loss in the Amazon basin. By 2024 the agreement covered roughly 35,000 Brazilian producers and every major European and North American grain trader operating in the country. Its leverage was simple: any soy grown on land cleared illegally after the cut-off date would be excluded from the major supply chains.

Modelling published this week estimates that the pact prevented the conversion of roughly 17,000 square kilometres of forest over its lifetime, an area larger than Kuwait. That figure comes from a counterfactual exercise comparing observed land-use change against a no-moratorium baseline. The same paper now projects that, without renewal, an additional 14,000 square kilometres will fall by 2036 as frontier pressure migrates from the Cerrado savanna back into the Amazon itself.

Why it collapsed

The immediate cause is procedural. The Brazilian Vegetable Oils Industry Association (Abiove), which administers the pact, and the Ministry of Agriculture failed to agree on a successor instrument before the July deadline. Reporting from PHYS on 17 July 2026 attributes the breakdown to disagreements over how a renewed moratorium would treat land cleared in the Cerrado biome, where most of Brazil's recent soy expansion has occurred.

The deeper cause is political. Under the current administration in Brasília, the agricultural frontier has been treated as a strategic asset. Deforestation alerts from INPE, Brazil's space research institute, have fluctuated since 2023, but the structural drivers, cattle pasture, soy, illegal mining, road paving, have only grown. The moratorium functioned because Brazilian exporters wanted access to European markets operating under the EU Deforestation Regulation. That commercial pressure has weakened as traders diversify routes through China and as some European buyers soften their stance under domestic political pressure.

In other words, the agreement collapsed not because the forest became less valuable, but because the cost of ignoring it went down.

The counter-narrative

Brazilian agricultural interests frame the moratorium as a foreign-imposed constraint on sovereign land use. The argument has force: the pact was negotiated by European retailers in response to European consumer preferences, and its enforcement mechanism was the European market, not Brazilian law. Ruralista lawmakers in the National Congress have argued for years that the agreement benefited competitors in Argentina and the United States by capping Brazilian output without capping theirs.

Abiove, in its own communications, has stressed that the Brazilian soy sector has reduced its direct forest footprint to a fraction of what it was in 2006. That is accurate. The remaining pressure on the Amazon comes primarily from cattle ranching, with soy following cleared pasture rather than primary forest. The new modelling assumes that soy demand will again pull directly on forest frontiers once the supply-chain check is removed.

The structural counter is that the agreement, whatever its limitations, was one of the few instruments that actually moved the needle. Its absence leaves a policy vacuum that neither Brazilian environmental enforcement nor European regulation can plausibly fill by 2030.

What 1.4 million hectares means

The headline figure translates into roughly 0.8 gigatonnes of additional carbon dioxide equivalent over the decade, on the order of Canada's annual emissions from energy and industry. That is not a marginal number for the global carbon budget; it is the equivalent of adding a mid-sized industrial economy's worth of emissions, in a country whose own NDC commitments already struggle to constrain growth.

The figure is also conservative. The modelling assumes only partial enforcement failure and does not account for cascading effects: new roads, new illegal logging camps, new land-tenure disputes in the arc of deforestation that runs from Pará through Maranhão to Rondônia. Each of those tends to bring more fire, more cattle, more informal credit, and more political permission to clear.

The Global South framing here is uncomfortable. Brazilian policymakers are right that European demand helped create the original dynamic; they are also right that domestic politics, not foreign pressure, determines whether the forest survives the next decade. The two truths do not cancel.

What to watch next

Three dates will determine how serious the loss turns out to be. First, the 2026 planting season in Mato Grosso, which begins in September, will reveal whether traders quietly maintain soy-supply restrictions without the formal agreement. Second, the EU Deforestation Regulation's enforcement timetable, which has already been delayed and may slip again under pressure from member-state agricultural lobbies. Third, the 2028 Brazilian general election cycle, which will set the regulatory tone for the second half of the decade.

What remains genuinely uncertain is whether Brazilian federal environmental agency IBAMA can scale up enforcement fast enough to substitute for what the private sector was doing voluntarily. The sources do not specify whether enforcement budgets have changed in the past 12 months, and the modelling assumes enforcement capacity remains roughly constant. If that assumption is wrong on the upside, the 1.4 million hectare figure will be lower; if wrong on the downside, it will be higher.

Either way, the political economy of the Amazon has just shifted. A conservation instrument that cost the Brazilian government nothing, and that operated for two decades without a single coercive clause, has ended. Replacing it will require either a new commercial compact or a stronger state, neither of which is currently visible on the horizon.

*This article reports the modelling as published on 17 July 2026 by PHYS. Monexus will update the figures if the underlying counterfactual is revised or if a successor instrument is announced before the 2026 planting season.