Why so much soy?

Rising competitiveness of Brazilian soy could be an invitation to reimagine the 87 million acres US producers currently dedicate to soy

Soy keeps coming up in conversations. 

Or at least the conversations I find myself in.

At the Sustainable Ag Summit in Charlotte in December 2023, soy was the topic of no fewer than six different discussions I had at a single cocktail party. Subsequently, soy continues to lurk at the root of discussions with clients about their greenhouse gas footprints, their suppliers’ footprints, their science based targets, and their water water impacts. The focus on soy is not because there is anything inherently wrong with soybeans, but because the production of soy increasingly comes at the cost of really any other crop in the United States and the Amazon rainforest and grasslands in South America and Canada. With poultry clients, soy features prominently in climate impact conversations as ~70% of their footprint comes from feed and 65-70% of their feed’s land footprint is soy.

These linkages give soy what’s referred to as high embedded land use change (LUC) emissions. For many regions, these embedded emissions from LUC exceed emissions associated with soy production many times over (check out Argentina, Brazil, India, Paraguay, and even Canada in the graph below). These regions are all converting forests and/or grasslands, which hold far more carbon in their biomass and/or soils, to croplands to produce soy. This means the key sustainability issue for soy is not how it’s grown – with or without a cover crop for instance – but rather how much of it we grow and at the expense of what other ecosystems and crops. 

So why are we growing so much soy?

If you, like me, enjoyed some tofu with rice noodles from time to time, I’ll go ahead and spoil things for you: You and all the tofu eaters out there were not the culprit. Neither were all the preschoolers eating their Dora the Explorer-branded edamame snack packs. Less than 2% of global soybean production goes directly to human consumption in the form of soymilk, tofu, tempeh, and the like, according to Our World in Data. 

Instead, the vast majority of global soy production proceeds from the grain bin to the crush plant to be separated into soy oil and soymeal. Most crush plants literally press the oil out of soybeans during processing. Many also further extract remaining oil from the meal using hexane and other solvents to make sure, as one crush plant operator once phrased it to me, the full “liquid gold” portion. 

Why did we go all in with a bean as an oilseed?

Soybeans account for more than 90% of US oilseed production. Compared to other true oilseeds like sunflower, canola or palm, soybeans are rather pathetic oil sources. Where sunflower seed can be upwards of 50% oil, soybeans yield a paltry 10.7 lbs of oil from every 60 lb bushel (just 17.8% oil). But, unlike other oilseeds whose meals have limited nutritional value to livestock, either due to excess fiber or the wrong ratio of amino acids, soybean meal is easily digested by both pigs and poultry. 

Booming demand for pork and poultry in the US and China has pushed soybean prices up as has booming demand for soy oil used as a cooking oil and in biofuels. Soybeans are also a legume, unlike other true oilseeds, which means nitrogen fertilizer, with its associated high costs related to the natural gas required for its production, is not needed. Though, US producers still apply an average of 17 lbs of nitrogen fertilizer to each acre of soybeans they plant. While this is largely unnecessary from an agronomic perspective, it is still far less than the 150 lbs of nitrogen applied to to each US acre of corn on average, which makes the comparative cost of soybean production lower than corn and further enhancing profitability, especially in these last few years when the Russia-Ukraine war disrupted gas supplies and prices. 

Policy shifts account for a big piece of market dynamics

And, in the policy arena and food system space, I am feeling a big sense of déjà vu. 

All the current policy efforts to push soy-based biofuels feels like corn ethanol, a la 2006-2007 when the initial Renewable Fuel Standard (RFS) inflated corn demand and acreage by nearly 16 million acres year-over-year (USDA-NASS). Subsequent analyses have found this land use change resulted in net negative impacts not just on the soil and water, but the climate and food security (here, here, here). Lark and colleagues (2022) estimated ethanol’s GHG emissions are likely about 24% higher than gasoline’s.  

Yet, here we are again at a moment when, driven by policies like California's Low Carbon Fuel Standard, the next RFS, and the recent COP28 commitment to phase out fossil fuels, the US has doubled down on these bad decisions, this time with an even less land-efficient crop. Industry analysis of announced crush plant expansion estimates between construction of entirely new crush facilities and expansion of existing ones, the US is on-track to increase domestic soy crush capacity by more than 34%. 

Such an increase in demand may be partially offset by decreasing exports, but still the US (and global!) land use ramifications could be tremendous considering the US already harvests more than 86 million acres of soy, a crop that yielded just under 50 bushels per acre on average in 2022, compared to corn’s 173+ bushels per acre. 

Comprehending land efficiency

I find it helpful to think about how a typical product of US cropland — poultry feed — translates into corn and soy land foodprints. For typical broiler chicken feed rations that are typically more than 65% corn by weight, soy accounts for more than two thirds of the product’s land footprint, assuming average yields and typical crush efficiency. Yet, booming domestic poultry demand and somewhat recovering pork demand, augmented by strong exports of both meat and soy over the last decade, has supported a strong market for this US soy.

But even increasing Americans’ annual meat consumption from 167 lbs per capita in 1960 to 227 lbs per capita in 2022, nearly 100 lbs more than US Dietary recommendations, even assuming all protein requirements were met by meat consumption, has not been enough to sop up our booming soy production. Instead, the US relies on export markets to purchase more than half of the US crop, mostly from China. In recent years China alone purchased more than 26% of US soy, but trade wars, African swine fever, and China’s softening economic growth slowed its booming demand for soy to feed pigs and chickens. 

Additionally, Brazil surpassed the US as the global leader in soybean production. In the last 10 years, Brazilian soybean production nearly doubled, enabled primarily by increases in the area planted to soy, which expanded from 74 million acres to about 113 million acres, as well as by a 17% yield increase. Where did all this new soybean ground come from? Well, most recently much of it was cattle pasture. 

Prior to being converted to cattle pasture, it was forested. In the last 10 years, Brazil lost more 79 million acres of tree cover. Removal of those trees and the carbon they held in their trunks and branches likely emitted more than 18 Gt of CO2 to the atmosphere, a steep climate cost to all these soybeans. For context, the US, which faces far less scrutiny for its tree cover loss, lost about 56 million acres in the last 10 years, likely associated with 8 GT of CO2 emissions to the atmosphere, according to Global Forest Watch. So, while the US lost less than half as much tree cover as Brazil, what it did lose was not insubstantial. 

Photo credit: Alison Grantham

A key shift in the global power dynamic

Brazilian soybeans are competing powerfully with US production. This shift is evident in the financial disclosures of major fertilizer and agriculture chemical companies, as well as commodity businesses. Many of them acquired Brazilian companies and assets or been acquired by Brazil- or China-based companies over the last decade (e.g., here, here, here, here, and here). These shifts were temporarily assuaged by record government payments to US soy producers (and all farmers) and Russia-Ukraine disruption-related record high soybean prices. 

But in recent months, there is growing anxiety about falling soybean prices. El Nino is giving Argentine production a good outlook while Chinese economic indicators continue to be weak. This dynamic is driving lobbying efforts for soy-based biofuels in an effort to buttress demand for the crop that is now grown on nearly half of harvested US crop acres. This lobbying for shifting liquid fuels from fossil to soy, could further expand acreage of this crop, eroding or eliminating any climate benefit from the shift. 

If climate benefit is the intended policy outcome, our tax dollars would be far better spent on investments in high speed rail, mass transit, and electrification. All those acres of soybeans? Retirement or conversion to perennial protein and oil crops like hazelnuts, pecans, or even hickories could support many more climate benefits, alongside water quality and enhanced animal welfare pork and poultry production systems. There are ways we could all win from the declining eminence of the American soybean, but not if we push back with policy-driven biofuel mandates to inflate market demand.