Soybeans: Who's Right, USDA or CONAB?

• The great debate following Thursday's round of imaginary numbers was the growing difference between USDA's and CONAB's Brazilian soybean crop production estimates.
• Meanwhile, noncommercial traders have shown no fear in adding to their already record large net-short futures position.
• Futures spreads continues to show concern from the commercial side over longer-term supply and demand.

I’ve said it before and I’ll likely say it again, I wasn’t expecting to continue the discussion of USDA’s February guesses into Friday morning. Yet here we are. Thursday’s post-report gnashing of teeth and rending of garments, while humorous to watch from afar, raised what many view as an important question: Which group is right about Brazil’s 2024 soybean crop production, USDA or CONAB? To help clear things up a bit, we all know USDA provides the “official” US government guesses on all things supply and demand related to agriculture. According to most in the industry, US markets cannot function without the guidance of USDA, which is why so many get so frothy over the release of new numbers every month. CONAB is Brazil’s equivalent of USDA. Enough said. 

In its February World Agricultural Supply and Demand update, USDA pegged Brazil’s soybean crop at 156.0 mmt, down 1.0 mmt from its January guess. At the same time, CONAB released its latest estimate of 149.4 mmt, reportedly a “decrease of nearly 3.8% from January and well below initial expectations…”. To those who view such numbers as important, this was too big a divergence, coming in at roughly a 4.5% difference. Some obviously thought it was substantial enough to cry “Foul!”, though don’t expect either side to make an immediate revision (though such a thing is not impossible). 

While my underlying question remains the same, “What difference does it make?”, I’ll address the other question that caused blood to boil Thursday afternoon and evening, “Which group is right?” 

As I talked about last time, the answer is relatively simple. What is the market telling us? Isn’t that what really matters? As of a week ago this past Tuesday, noncommercial traders (funds, investment traders, Watson[i], etc.) had moved to a record large net-short futures position in soybeans of 140,577 contracts. They did this largely without fear of underlying fundamentals. Additionally, through this past Tuesday’s close (CFTC data is pulled after Tuesday settlements) March soybeans (ZSH24) had dropped as much as 39.25 cents before finishing the week down 19.25 cents. Meanwhile, May (ZSK24) lost as much as 37.75 before closing 19.5 cents lower Tuesday-to-Tuesday. What does this tell us? Funds were adding to their net-short futures position, setting the stage for this afternoon’s CFTC Commitments of Traders report (legacy, futures only). 

But what about REAL fundamentals? The March-May futures spread closed January at a carry of 10.5 cents and covered 33% calculated full commercial carry, as compared to the end of December’s settlement of 9.25 cents carry and 28%, with 33% or less considered bullish. The commercial side of the market indicated it had grown more comfortable with Brazilian production and total supplies, albeit slightly, in relation to demand. That’s all that matters. At Wednesday’s close, before the release of the latest round of imaginary numbers, the March-May futures spread showed a carry of 8.5 cents and covered 27% cfcc. 

While the March-May futures spread is our best read on Brazilian production, and we can see the commercial side has consistently registered its awareness of a smaller crop, US supply and demand expectations can be seen in the May-July spread. Why? This spread covers the timeframe after Brazil’s production is generally known, and we start to see the possible ripple effects on demand for US supplies. As I discussed last time, the end of February is a key time for the soybean market as March moves into delivery, shifting the focus to the May futures contract. Also, it is the midpoint of the make-believe marketing year so many like to talk about, a time when the US tends to have shipped just short of 75% of what turns out to be total exports for the year. 

The May-July spread closed last December at a carry of 5.75 cents and covered 18% cfcc and this past January at 9.0 cents and 29% cfcc. Yes, commercial traders are still bullish, meaning the US could see a later than normal round of export demand if China pulls as much as expected from Brazil. (For the record, USDA upped its Brazilian exports to an incredible 100 mmt while lowering its US export guess to 46.82 mmt. Remember what we were told, “Trade wars are good, and easy to win.”) If, for whatever reason, Brazil’s soybean supply grows uncomfortably tight the US could start to whittle away the nearly 20% deficit it has been running to last year’s total sales (total shipments plus unshipped sales). As of the latest weekly sales and shipments update, for business through Thursday, February 1, total US export shipments were projected (by me) at 1.637 bb as compared to last year’s reported shipments of 1.918 bb. USDA’s February export demand guess was trimmed 35 mb to 1.72 bb. 

Who was right Thursday? Who cares. The only opinion that matters is the market’s. That’s the bottom line. 

[i] My name for the algorithm driven investment industry in general. 

Disclaimer: The copyright of this article belongs to the original author. Reposting this article is solely for the purpose of information dissemination and does not constitute any investment advice. If there is any infringement, please contact us immediately. We will make corrections or deletions as necessary. Thank you.