How to read a WASDE balance sheet (and what the trade actually watches)

What WASDE actually is

The World Agricultural Supply and Demand Estimates is a monthly USDA publication released the second Friday of every month at 12:00 ET. It contains balance sheets — supply and demand line items — for every major US-relevant commodity (corn, soybeans, wheat, sorghum, rice, cotton, beef, pork, dairy) plus world-level balance sheets for the same commodities.

Every balance sheet follows the same structure: beginning stocks + production + imports = total supply; total supply − domestic use − exports = ending stocks. The line items change weight by month — early in the marketing year, supply estimates carry most of the signal; late in the year, demand revisions and ending stocks dominate.

An anatomized example

Here's the shape of a US corn balance sheet entry on release day. Columns labeled "May" and "Apr" are the new and prior month's estimates; the column labeled "Trade" is the pre-release Bloomberg / Reuters survey average. The unit is million bushels (mil bu) for everything except yield (bu/acre).

Line	Apr	Trade	May	vs Trade
Beginning stocks	1,540	—	1,540	—
Production	15,150	15,250	15,340	+90
Yield (bu/acre)	181.0	181.5	183.0	+1.5
Imports	25	—	25	—
Feed & residual	5,800	—	5,825	—
Ethanol	5,500	—	5,490	—
Exports	2,400	2,380	2,310	−70
Ending stocks	2,025	2,100	2,290	+190
Stocks-to-use	14.6%	—	16.3%	—

The headline most retail readers focus on is the ending-stocks figure: 2,290 vs 2,025 prior. The bigger pile of corn is "bearish." That's not wrong, but it's incomplete and frequently miscalibrated.

The four things desks actually watch

1. Surprise vs trade estimates — not vs prior month

Markets price the trade survey before the release. The price impact of WASDE day comes from the difference between USDA's number and what was already in the curve. In the table above, ending stocks at 2,290 vs trade estimate 2,100 is a +190 million bushel surprise — that's what moves price, not the +265 mil bu vs April.

A rule of thumb: ending-stock surprises of more than 5% vs the trade estimate move corn ~10-25 cents on release. Soy moves ~30-60 cents on similar surprises. Wheat is noisier. The actual move depends heavily on the prior week's positioning — see the COT briefing for why a heavily-long market reacts harder to a bearish surprise.

2. Stocks-to-use, not absolute stocks

The 2,290 mil bu ending-stocks figure is meaningless without scale. The metric that matters is stocks-to-use: ending stocks divided by total domestic + export demand. In the example, that's 16.3%, vs 14.6% the prior month.

Why this matters: 2,290 mil bu of carryover sounds large in isolation, but if total use is climbing alongside, the cushion is unchanged. Historically, US corn stocks-to-use below 10% is "tight" (price-supportive), 10-13% is "adequate," and above 14% starts to compress new-crop premium. The May print at 16.3% says new-crop is comfortably supplied — that's the real signal.

3. Feed & residual + cross-commodity demand

The feed-and-residual line is the demand category most prone to monthly revisions and the one where corn and soy substitution shows up. When wheat is cheap relative to corn at the farm gate, US feed wheat usage climbs and corn feed demand softens — the WASDE will telegraph this in the corn feed line and the wheat feed line moving in opposite directions across months.

If you trade the corn-wheat or corn-soybean spread, this is the line you watch first. The headline corn balance sheet might look bearish in isolation while the spread story is bullish for corn relative to wheat. Reading WASDE without the cross-commodity lens is reading half the report.

4. The world lens, not just the US balance sheet

The US is a big producer but the price is set globally. WASDE's world tables (separate document, released same time) cover Brazil, Argentina, Ukraine, Russia, China, EU. Two world-balance signals matter most:

• SA production revisions — Brazil and Argentina soy and corn estimates move materially through the SA growing season (Oct-May). A 2 MMT revision to Brazilian corn changes US export prospects.
• Chinese demand assumptions — China is the marginal soy buyer. WASDE imputes their import demand based on crush margins and stockpile policy. Small Chinese revisions cascade into US export expectations and basis.

What not to do

• Don't trade WASDE on the headline number alone. Wait for the surprise vs trade — the screen tickers report it within seconds.
• Don't compare month-over-month without checking marketing year. The May report contains old-crop (current marketing year) and new-crop (next marketing year) balance sheets. Most of the action late spring is in the new-crop column.
• Don't ignore the world tables. If you only read the US tables, you're missing the half of the report that increasingly drives price.
• Don't fight the curve in the first 15 minutes. The initial reaction is high-frequency desks taking surprise + positioning; the real read often emerges in the next session as the cross-commodity and world implications get priced.

Where to find it

Direct from USDA at usda.gov/oce/commodity/wasde. The PDF, an XML companion file with the tables, and historical archives are all free and public. Bloomberg, Refinitiv, and ag-specific services like AgriCensus repackage the same data with their own commentary — but the source is free, and reading the actual report puts you closer to the data than the secondary coverage.

Set a calendar reminder for the 2nd Friday of every month, 12:00 ET. Have the trade-estimate survey open in another tab. The first thing you compare is the new-crop ending-stocks line vs the trade estimate. Everything else builds from there.