Confused by How USTR, USDA, and Treasury Officials Have Been Describing the U.S./China Soybean Deal?
Trying to make sense out of flip-flopping timeline
This item attempts to make sense out of changing timelines for China’s apparent commitment to purchase U.S. soybeans over multiple years.
The U.S./China agreement between President Donald Trump and Chinese leader Xi Jinping initially said via a fact sheet that China committed to purchase (not ship) 12 million metric tons of U.S. soybeans by the end of December 2025, and 25 MMT annually in 2026, 2027, and 2028.
There is no cumulative or multi-year pool; it is an annual target.
The 12 MMT timeline has shifted. Initially, the White House and USTR described the China commitment as: “12 MMT of U.S. soybeans by the end of 2025.”
But in subsequent briefings — especially those by Treasury Secretary Scott Bessent and USTR Jamieson Greer — the phrasing shifted to:
• “by the end of the season” (USTR Jamieson Greer)
• “by the end of February” (Bessent)
This is not an accident, I am told. It reflects internal recognition that China was not on pace to hit 12 MMT by the end of December.
What they’ve effectively done is move the goalposts from a calendar-year 2025 deadline to a crop-marketing-year 2025/26 deadline.
So what does “the season” mean? In U.S. agricultural trade and USDA reporting, “the season” always means the Marketing Year. For soybeans, the Marketing Year = Sept 1 to Aug 31. This is the definition used by:
• USDA’s World Agricultural Outlook Board
• USDA Foreign Agricultural Service export reporting
• Congressional Research Service
• All WASDE tables and export projections
Of note: One source emailed: “USTR doesn’t live in the MY world that USDA does, so it was a lack of precision on their part. But that’s merely speculative.”
So when an official says:
• “by the end of the season” → means by Aug. 31, 2026
• “by the end of February” → means by the peak period of the season, not the official end
Why was Bessent talking about “the end of February”? Two possible reasons:
1. October-February is the peak window for U.S. soybean availability. The U.S. usually dominates global shipments between October-February before Brazil’s big crop comes online (March forward). If China is going to close a 12 MMT gap, they must act before Brazil’s beans hit global markets.
2. Politically, “end of February” buys China time while still sounding immediate. If China can front-load 6–8 MMT by late winter, the Trump administration can say the commitment is “on track,” even if the remaining tons slide into spring/summer.
So while the official deadline has blurred into “the season,” Bessent emphasized February because that’s the only window in which the U.S. can realistically supply the bulk of 12 MMT.
What does this shift actually mean? It’s a de facto extension. China was nowhere near the pace needed for 12 MMT by Dec. 31. Shifting the metric to “the season” pushes the deadline eight extra months (Dec → Aug 31). It allows re-labeling normal seasonal flows as “compliance.”
U.S./China soybean shipments ordinarily surge in Oct–Feb anyway. By moving the benchmark, the Trump administration can say China is meeting its commitment even if they only return to typical seasonal purchasing behavior.
It reduces pressure on China to buy aggressively in a short window. It also reduces near-term price impact.
A December deadline would have compressed purchases into a few weeks — bullish. A season-long window spreads demand across months — much less price-moving.
So, current reality: 12 MMT by the end of the 2025/26 soybean marketing year → Aug. 31, 2026.
What Bessent meant: China must show meaningful volume by February 2026 because that’s the last competitive window before Brazil shifts the world market. Bessent’s “end of February” remark refers to the practical purchasing window, not the legal deadline. But the shift dramatically dilutes the meaning of the original 12 MMT commitment.
Next up: Is that 25 MMT measured on a calendar-year basis or a marketing-year basis?
Sources say it is almost certainly measured on a marketing-year (MY) basis — NOT calendar-year. Here’s why:
A. Every USDA enforcement, verification, and reporting mechanism uses marketing years. Also, USTR learned in 2020–21 (Phase One agreement with China) that calendar-year targets are unworkable for a commodity whose exportable supply varies dramatically across crop cycles.
B. All internal references from USDA, Treasury, and USTR to “the season” or “annual commitment” default to Marketing Year definitions.
The same people who shifted the 12 MMT deadline from “2025” → “this season” (meaning the 2025/26 MY) are applying that same structure to the 25 MMT annual commitments.
C. A calendar-year definition would double-count or under-count the U.S. export window. Again, the U.S. ships most soybeans Oct.–Feb., so if targets were calendar year, China could backload purchases each year in ways that distort compliance measures. A MY basis avoids that.
D. This is exactly how the 2020 Phase One soybean commitments were interpreted operationally. Even though Phase One text referenced calendar years, actual enforcement and USDA monitoring used Marketing Year math, because the market cannot be evaluated any other way.
Conclusion:
The 25 MMT commitments for 2026, 2027, 2028 are being treated as Marketing-Year commitments (MY 2026/27, MY 2027/28, and MY 2028/29).
Does the 12 MMT commitment factor into the 2026 25-MMT commitment? Short answer: No — the 12 MMT is separate and does not count toward the 25-MMT 2026 commitment. Here’s why:
A. The 12 MMT is tied to clearing the 2025 backlog. It applies to the 2025/26 Marketing Year, created to compensate for China’s near-zero purchases during most of calendar 2025. It is effectively a “catch-up tranche” before the new multi-year structure begins.
B. The 25-MMT commitments start the next marketing year.
2026/27 → 25 MMT
2027/28 → 25 MMT
2028/29 → 25 MMT
Nothing in the language suggests the 12 MMT is a “floor” feeding into the 25. It is conceptually:
12 MMT = make-good for 2025 (But one source emailed: “For many, it effectively means that the U.S. will lose a half a year of exports to China for 25/26.”)
25 MMT = forward-looking annual baseline starting in 2026
C. Using 12 MMT as part of the 2026 target would undermine the negotiated structure. If 12 counted toward 25, China would only need to buy 13 MMT in 2026. U.S. officials — especially trade economists — explicitly rejected that interpretation when drafting the commitments, I am told.
D. Officials refer to the 12 MMT as “this year’s requirement” and the 25 as “next year’s minimum.” They are consistently described as separate buckets.
Summary
Commitment 1
12 MMT by the end of the 2025/26 soybean marketing year (Aug 31, 2026).
(This is the shifted, more realistic interpretation of the original 12-MMT pledge.)
Commitment 2
25 MMT in MY 2026/27
25 MMT in MY 2027/28
25 MMT in MY 2028/29
The 12 MMT does NOT roll into the 25 MMT.
The 25 MMT is a fresh, independent annual baseline starting with MY 2026/27.
Front-loaded “real world” benchmark (70% Oct–Feb, 30% Mar–Aug)
Assumptions:
Oct–Feb = 5 months
Mar–Aug = 6 months
70% ships Oct–Feb, 30% ships Mar–Aug
1. 12 MMT year (MY 2025/26)
Oct–Feb (70%) → 8.4 MMT over 5 months
≈ 1.68 MMT/month
Mar–Aug (30%) → 3.6 MMT over 6 months
≈ 0.60 MMT/month
2. 25 MMT years (MY 26/27–28/29)
For each 25 MMT year:
Oct–Feb (70%) → 17.5 MMT over 5 months
≈ 3.5 MMT/month
Mar–Aug (30%) → 7.5 MMT over 6 months
≈ 1.25 MMT/month
Of note: One industry contact emailed: “It could very well be that Cofco U.S. has already purchased the soybeans and has yet to transfer them to Cofco China, which would trigger a flash sale. Most likely the 12 MMT would ship by the end of Feb but it’s possible some could ship in the summer. The biggest question may be will they take 25 MMT given the large South American crop.”
