What is Brazil’s soybean and corn production size? 

By Daniele Siqueira, Head of Market Intelligence with Brazilian consultancy AgRural 

In a normal year, I would answer the question above by saying it is still early to tell how much corn and soybeans Brazil will be able to harvest in its 2023/24 crop season. Soybeans were planted later than normal and the second corn crop, which accounts for nearly 75% of the country’s total corn production, is still being sown. 

All that is true. This year, however, we have to deal with another variable: the wide differences between estimates, including the forecasts released by sources considered “official” by the market: Conab, Brazil’s federal crop agency, and the USDA. 

On Feb 9, Conab made another cut to Brazil’s soybean production estimate, which was put at 149.4 million metric tons, about 13 million down from the initial estimates due to the impact of hot, dry conditions on several states, especially Mato Grosso, the country’s largest producer. 

On the same day, the USDA trimmed its soybean production estimate for Brazil to 156 million metric tons, 7 million down from the initial estimate — almost the same difference from the number released by Conab just a few hours earlier. To make things worse, private estimates vary from 135 million to 155 million metric tons.

Why so many differences? 

So, what is Brazil’s soybean production size? And why have there been so many differences between forecasts? I will start with the private estimates because the answer is simpler: people and companies looking for their 15 minutes (seconds?) of fame and/or believing their numbers can change prices in Chicago. 

Mild differences between major consulting firms (among which is the company I work with, AgRural) are normal and reflect different methodologies and/or the exact timeframe of the estimate. Our latest number, released in mid-January, is 150.1 million metric tons, by the way. That is one thing. Another thing is beginners and headline/click/like diggers who come up with numbers that nobody knows how were put together.

However, the differences between Conab and the USDA’s numbers are a little more complex and started a while ago. Conab’s soybean production estimates for Brazil over the five latest crops (2019/20 to 2023/24) are 124.8 million, 139.4 million, 125.5 million, 154.6 million, and 149.4 million metric tons. The USDA’s numbers for the same five seasons are 128.5 million, 139.5 million, 130.5 million, 162 million, and 156 million metric tons.

Supply and demand

If we make Conab minus USDA, we have the following discrepancies: -3.7 million, -0.1 million, -5 million, -7.4 million, and 6.6 million metric tons. Add to those calculations different numbers, depending on the source, for beginning stocks and domestic use, plus different marketing year calendars (Conab uses Jan-Dec, while the USDA goes with Oct-Sep) and we have a big mess in terms of supply and demand.

A few years ago, Conab’s supply and demand balance just imploded when Brazil exported significantly more soybeans than it would be possible to ship considering the agency’s supply numbers. It took more than a year for Conab to come up with a new balance, based on historical revisions that were more focused on the domestic use than on production estimates. 

In my opinion, Conab tends to underestimate Brazil’s soybean production. On the other hand, we have to consider that the agency has a consolidated methodology that uses several tools, including satellites and, more important than that, people with boots on the ground to see what is really going on with the crops across the country.

Conab x USDA

What limits Conab’s capacity to put together a supply and demand balance that shows the soybean reality in Brazil is the absence of a robust stock survey system (like the USDA’s quarterly stock reports) that could allow the agency to fine-tuning its production numbers at the end of each season. 

The USDA, as far as I know, doesn’t make field surveys in Brazil to estimate the soybean production, and that is something that makes me go more often with Conab’s forecasts (although always taken with a pinch of salt). It seems to me that the USDA’s numbers, particularly when fine-tuning the previous crop’s size, are based on demand figures, especially exports. There is nothing wrong with that, but the constant increases in Brazil’s last crop estimate (now seen at a stunning 162 million metric tons) have made me feel a little uncomfortable.

Also a problem for corn

Similar discrepancies have occurred with Brazil’s corn numbers. In early February, Conab estimated the 2023/24 production at 113.7 million metric tons, while the USDA works with 124 million. (AgRural, in turn, forecasts 117.4 million metric tons.) For the previous crop, Conab forecasted 131.9 million metric tons, below the USDA’s 137 million. 

In a year with plenty of corn everywhere (especially if we consider that Argentina might have a normal crop this season, after the severe drought that reduced the country’s yields last year), a 5 or 10 million metric ton gap between Conab and the USDA’s estimate might not result in a big change in the world market. But 5 or 10 million metric tons can make or break someone in a year when supplies are tighter. 

For that reason, and considering that Brazil is now too important in the world’s soybean and corn production and exports, it would be good to see numbers that everyone could, if not believe, at least take into account.

Check Also

Waiting on the proper soil conditions

On this week’s Pioneer Spring Agronomy Update, Dave Russell visits with Pioneer Field Agronomist Alex …

Leave a Reply

Your email address will not be published. Required fields are marked *