FORT COLLINS, Colo.- Less than two years ago the United States was drowning in excess soybean supplies, mostly due to trade interruptions with top buyer China.

But relative to domestic demand, U.S. stockpiles are forecast to be record-tight later this year, and according to the U.S. Department of Agriculture, not even a record crop this summer can ease the situation much by mid-2022.

USDA on Friday tentatively slated U.S. soybean acres to rise 8% on the year, which could lift the 2021 harvest to a new record when combined with the second-largest ever yield. 

However, despite a slight downtick in demand, the agency pegged domestic ending stocks for the upcoming 2021-22 year at 145 million bushels, the second-smallest in eight years after the current year’s projection of 120 million.

This prompts the question of what exactly would be required to raise the carryout levels more substantially, and how long that would take. The answer depends on forward Chinese demand, disease control and production recovery in China’s hog herd, top soy grower Brazil’s expanding production and export capacity, U.S. summer weather and corn prices, among other items.

Those things are nearly impossible to predict, but there may be some lessons from the past that could help traders be ready for the next supply shift. Two recent events stand out, including the 2017 easing of Chinese imports and complacency over the 2019-20 U.S. carryout reduction, both of which contributed to the sharp surge and decline in U.S. stockpiles over the last three years.

DEMAND MISS

The soybean market several years ago finally got used to China’s booming appetite for the oilseed and was unprepared for any hiccups in that trend, though signs of the slowdown that occurred late last decade are easier to see in hindsight.

One clue was that in the typical October-December U.S. soy shipment peak in 2017, exports to China were actually sitting at a four-year low. That can partially be explained by an increased volume out of Brazil, but USDA forecasts were also predicting the Chinese demand pullback at the time.

In late 2017, USDA saw 2017-18 Chinese soybean imports rising just 3-4% on the year compared with an average increase of nearly 12% per year in the previous four seasons. In the end, the 2017-18 volume rose only fractionally on the year due to reduced trade with the United States and the outbreak of African swine fever (ASF) in its hog herd.

But the soybean market in early 2018 was distracted from China’s slowing imports by a crop failure in Argentina, the world’s leading exporter of soy products, and prices signaled U.S. farmers to plant another massive soybean area. That led to a record U.S. crop and drove stockpiles to unfathomable highs as trade was choked off with China in mid-2018.

It is unclear why China’s demand trajectory slowed in late 2017, but it is possible that imports in the prior year had finally caught up to and surpassed the needed levels for the time being.

A more sinister though plausible theory is that the first ASF impacts may have arisen as early as 2017, regardless of whether China was aware at the time. Beijing first officially acknowledged ASF in August 2018, though the wider industry did not understand its severity until seven months later.

STOCKPILE MISS

Following 2018, a correction to the massive stockpiles was needed. But the soybean market a year ago probably underestimated the risks of cutting U.S. ending stocks in half from the prior year, since the reduced levels were still elevated versus history.

Soybean prices were weak in early 2019 and farmers intended to cut acres. But historic spring flooding and relatively better corn prices caused growers to plant 8.5 million fewer soybean acres that year than originally planned, an unprecedented change of course.

At the time, that acreage loss seemed like a disguised blessing to adjust the enormous inventory, which ended up falling more than 40% year-on-year. But looking back, the market showed too little concern in early 2020 over the effects of slashing carryout so sharply, basically leaving no room for error on the 2020 U.S. crop.

Soybean prices were not sufficiently attractive a year ago to gain enough acres as uncertainties over trade with China and the unfolding pandemic loomed. Still, the market thought at several points during the 2020 season that farmers had planted more soybeans than they did, perhaps misjudging the true weakness of the economics.

The late-summer 2020 drought amplified the underplanting effects, as yields were clipped in several major states, producing just an average crop instead of the strong one that was expected. The return of robust Chinese demand in mid-2020 capped off the recent supply squeeze.

2021-22 PREP

The soybean market is clearly still learning how to adjust to the exceptional supply and demand swings that resulted from recent events related to ASF, the trade war and weather, ultimately leading to the overplanting in 2018 and the underplanting in 2020.

But in terms of price impacts, the year-on-year changes in U.S. supply and Chinese demand probably deserve more attention. USDA currently sees China’s soybean imports at 100 million tonnes in 2020-21, up 1.5% on the year.

U.S. soybean stocks-to-use in 2020-21 is seen at an all-time low of 2.6%, slightly edging 2013-14, and that is down from 13% a year earlier and nearly 23% in 2018-19. The 2020-21 ratio had been predicted as high as 13.7% back in August.

USDA’s latest figures place 2021-22 stocks-to-use at 3.2%, the third-lowest on record.

Following 2013-14’s low of 2.6%, the ratio was at 5% in each of the next two years before climbing to 7% in 2016-17 and 10% a year later. But soybean prices were significantly lower in those subsequent years than the ratios might suggest because USDA had a habit of underestimating soybean demand at the time.

Under business as usual, however, the historically tight soybean stocks mean that supply risks will linger for at least another year despite the slight expected increase. Summer weather will undoubtedly be under more scrutiny than normal, and it may take more than a few decent weather forecasts by mid-July for confidence on a big U.S. output.

(Editing by Matthew Lewis) ((karen.braun@thomsonreuters.com; Reuters Messaging: karen.braun.thomsonreuters.net@reuters.com; Twitter: @kannbwx))