The First Look at S. Am. Production 2023
Easy Newz & Wt360: the Starting Point for 2023’s South American Production Forecast
Here is the initial forecast for South American soybean, corn, sunflower, and wheat production. We track minor crops (barley, rapeseed, etc.), but these updates will pertain to major crops. All figures are million metric tonnes.
Soybeans: Brazil (168.4), Argentina (47.72), Paraguay (9.64), Uruguay (2.87), Bolivia (3.4), Total (232)
Corn: Brazil (136.8), Argentina (55.32), Paraguay (5.07), Total (199.39)
Wheat: Brazil (10.05), Argentina (17.06), Total (29.7)
Sunflower: Argentina (4.38), Total (4.98)
“Generating an output is the easy part. How do we know it is any good?”
A clear starting point: why is this so important?
There needs to be an anchor or starting point. The first forecast is generated months ahead for two reasons.
The trend or evolution of the growing season sets the tone before the critical development stages of the crops. Once plants get through this period, markets have usually priced the risk premium. Either propelling prices higher if the weather is poor or selling off if the weather is benign. It is too late to prepare. Think of the starting point as “if everything went as normal and the realized weather followed our long-range models, what would production be?”
The second reason is the “probabilistic outcome” of things. Suppose any forecast (long-term or 7-day) shows a clear bias hot and dry or cool and wet. The more time an operation has to plan, the better. Rarely do people make good decisions when they are rushed or losing money.
Therefore, the initial forecast has to be early enough for businesses and farm operations to plan for the season ahead. Cutting 2 million tonnes from an arbitrary national figure does little to facilitate this.
There is already a similar scenario to this past May and June in the United States. Crop conditions were falling, and analysts were already cutting yields. Why? How vital are crop conditions and weather during this period? As it turns out, not very important. Yet, here we are at the end of October, and social media accounts are sounding the alarm (again).
Background on the Data and Modeling
The models are trained on four decades of weather data at nearly 2,000 locations across South America. The production history is a combination of USDA and private data sets. More data is built upon each day, improving the model’s robustness.
At the end of November, Easy Newz will split out Safrina and the main crop corn in Brazil. Easy Newz will be adding cotton and softs in the months ahead. Here is what to watch going forward.
Acreage Switching
Early weather anomalies such as too hot, dry, or wet will impact planting decisions and limit double cropping before having material impacts on yields. In Argentina, corn planting went all the way until January last year, and the corn seeds handle stress better than soybeans due to IP characteristics.
Right now, Argentina is at risk of switching 1 million or more hectares from corn to soybeans as dry weather persists. Ignore the social media accounts showing cumulative precipitation. Argentina’s weather is exhibiting characteristics of transitioning to a wetter pattern.
Weather Risk in El Niño Years
Brazil’s dry weather will force replanting in areas of Mato Grosso already sewn and delay planting in the north. Farmers in Brazil will risk missing the Safrina planting window if this drags on through November.
This year has started considerably dryer and hotter than a year ago. The forecasts are for central and southern Brazil to get wetter in the months ahead. The same goes for Argentina. If this does not happen, prices will have a function to move higher.
Other Considerations
Western hemisphere soybean stock build of more than 15 million tonnes.
Argentina soybean crush potential near 40 million.
Argentina's corn exports of 40 million and wheat exports of 10 million.
Brazil's soybean and corn exports could reach 100 and 56 million each.
Role of the MOG (Man on the Ground)
Humans still need to work with the computers. The financial situation around the elections will lead to more soybeans and less corn due to fertilizer shortages and farmers reducing risk. Brazil’s ability to harvest its most sewn soy acreage quickly will determine the upside on Safrina acreage.
Models cannot capture this because there is no data to train it on. Humans and expertise is still required to build a complete production picture. But it starts with robust data sets and modeling expertise. Excel cannot handle billions of data inputs, and opinions from analysts in the Midwest add very little value. South America’s production prospects will lay the foundation for a bullish or bearish price outlook in 2024.
