Soybean Stocks Overshadowed by Prospective Plantings

March 31st traders and farmers are likely to pay a great deal more attention to the number of soybean acres USDA expects will be planted this season than the number of soybean bushels left in the United States. However, the stocks figure may hold some surprises.

Last December the United States Department of Agriculture reported a surprisingly low Grain Stocks number for soybeans. The agency counts up available bushels of most crops once a quarter; in December, March, June, and September. University of Illinois Ag Economist Darrel Good says the December 1 soybean stocks number implied a record large residual use of soybeans during the first quarter (September-November) of the 2014–15 marketing year.

Quote Summary - Some have explained this low figure by suggesting a larger number of bushels of soybean were in transport on December 1 than in previous years. This explanation was apparently favored by the market and caused March 2015 soybean futures to close 36 cents lower on the day of the surprisingly small estimate. Another possible explanation is that the size of the 2014 soybean crops has been overestimated.

This argument might be supported by higher than expected soybean prices this year given the estimated size of the surplus projected to be generated by the large 2014 crop. In addition, basis levels have been generally strong for most of the year. Basis is the difference between the price of a futures contract in Chicago and the local cash bid.

USDA’s March 1, 2015 estimate of soybean stocks may add some clarity to this debate writes Darrel Good in his Weekly Outlook posted to the Farm Doc Daily website. Expectations for the magnitude of March 1 stocks are based on the estimate of December 1 stocks, imports during the quarter, and estimates of soybean consumption during the quarter.

If the size of the 2014 soybean crop has been accurately estimated, the March 1 stocks estimate should imply a large negative seed and residual use during the second quarter of the 2014–15 marketing year. That was the case in previous years of very large implied residual use during the first quarter of the marketing year. Seed and residual use during the second quarter of the marketing year, for example, was estimated at –38 million bushels last year, –22 million bushels in 2012–13 and –42 million bushels in 2009–10. A reasonable expectation this year might be near –90 million bushels. A March 1 stocks estimate near 1.41 billion bushels, then, would be consistent with the estimated size of the 2014 crop and known use of soybeans through February.

Given this, if the USDA’s Grain Stocks report shows something substantially different than 1.41 billion bushels on hand, then it should renew the debate over the size of last fall’s soybean harvest. Such a debate, however, would not be resolved for another six months. The USDA’s estimate of the crop size is frequently revised, but not until the release of the September 1 stocks estimate. It comes on September 30th this year.

Good says, historically, implied seed and residual use of soybeans during the first half of the marketing year has not been a good predictor of the size or direction of any subsequent change in the estimated size of the crop.

The March 31 Grain Stocks Report

The reports USDA releases March 31 will set the tone of agricultural trade for three months in Chicago.



Once every quarter the National Agricultural Statistics Service takes a census of the available bushels of corn, soybeans, and wheat. It is called the Grain Stocks report. It is not exactly a survey, but rather more of an actual accounting, in his case of what’s stored in Illinois, says NASS State Statistician Mark Schleusener, “…to measure the whole supply of grains and oilseeds USDA NASS does on farm surveys. Those are done with producers to find out what they have in their grain storage bins. Off farm storage tallies bushels in the mills and the elevators using a census as of March 1. All commercial storage facilities are contacted”.

Nationwide more than 9000 commercial storage facilities are contacted for the census side of the Grain Stocks report. The survey side - that done with farmers - is sent to more than 80,000 producers with an 80 percent response rate. The goal is to get a very accurate accounting of the bushels available for use.

Where the bushels are stored changes across the season. December 1 it is stored on farm. Through the winter months these bushels slowly move to the elevators and mills and eventually, in the case of corn, the bushels are shipped down the river for export, or fed to livestock, or turned into ethanol. The bushels are used.

If you add what’s used to what’s left - the Grain Stocks number - the sum should be the total available supply for the year. However, tracking the middle usage number for corn - bushels fed to livestock - isn’t possible. That’s why USDA calls this number Feed & Residual. This season it is supposed to be 5.3 billion bushels. The question is how much of that 5.3 billion has already been consumed. There in lies the guess says University of Illinois Ag Economist Darrel Good.

Quote Summary - If the most recent pattern is being followed this year and USDA’s 5.3 billion bushel usage for the year is correct, then use for the first half the year should total 3.9 billion bushels with 1.7 of that used in the second quarter. If that is the case, the total use during the second quarter would have been 3.75 billion bushel and leave March 1 stocks at 7.45 billion.

On-the-other-hand, if the usage pattern is more like it was prior to 2010, there could be another 200 or 300 million bushels of corn accounted for in the Grain Stocks figure because it hasn’t yet been consumed. It will still be consistent with a 5.3 billion bushel usage figure for the year.

The Grain Stocks report for corn has a wide range then of acceptable figures from around 7.4 to 7.7 billion bushels. It makes the Grain Stocks number not so important, and puts a great deal more weight on the Prospective Plantings report to be released on the same date, March 31.

How Much Would a Corn Acre in 2015 Make

The ag economists at ILLINOIS have done an interesting exercise to see how much an acre of corn might gross in 2015. Or maybe it might be better explained as what would happen in 2015 if this year was like 1979.



Or what if it were like 2012, or 1983, or 1995, or just pick a year. The idea is to give farmers some hard data on how variable gross revenue from a corn acre is over time by moving that time into 2015. So that’s what U of I ag economists Gary Schnitkey did.

He wanted to look and see what gross revenues would be like for 2015 considering crop revenue, crop insurance, government payments, and price risk. The goal was to know under what conditions would a corn acre produce higher gross revenues this year?

The question then is, “In 2015 what would revenue be like this year if a year like 1972 happened?”.

"When we looked at it, 50% of the revenues were above and 50% of the revenues were below $825 per acre."

Schnitkey put those all into a table on the Farm Doc Daily website from 1972 to 2014. It shows how much of gross revenue would come from price x yield, crop insurance, and government payments.

At $825 most $300 an acre cash rented farms in central Illinois would lose money. Over the span of the years this would happen about 75% of the time and a big yield does not solve the problem - it takes higher prices from some other force. You may read the “Gross Revenues in 2015” article on the Farm Doc Daily website.

How USDA NASS Counts Acres

USDA has just wrapped up its survey of more than 80,000 U.S. farmers. The agency uses the information to develop the March 31st acreage forecast.

In the spring USDA’s National Agricultural Statistics Service division contacts farmers in hopes of learning how much of each crop they expect to plant. The agency contacts farmers across the United States. Corn and soybean farmers are of particular interest. This year more than 4000 Kansas farmers were tapped, along with around 3700 in Nebraska and about 3000 in each of the Dakotas, Iowa, and Illinois. Another 2000 farmers each were contacted in Indiana and Ohio.

Quote Summary - Our goal is to make sure we are measuring small, medium, and large farms. So, we use what’s called a stratified sample.

That’s NASS Illinois State Statistician Mark Schleusener.

Quote Summary - That is a fancy way of saying for the biggest farms, we are going to talk to all of them; for the large, but not biggest we will talk with one out of three of those and for the medium, maybe one out of ten; and for the smaller farms we might measure one out of twenty-five of those.

Each farmer surveyed is asked how many acres they operate. How much of that land they intend to plant to corn or soybeans, and how much might already be in wheat. They’re also asked about oats, sorghum, and hay. The response rate goal, and usually achievement, is an amazing eighty percent.

Quote Summary - Yes, our goal is an 80% response rate on all surveys and we use several methods of data collection. Every producer in the sample receives a letter with a planting intentions questionnaire. The letter also has instructions for reporting to a secure internet website. These are both inexpensive ways of gathering data. The people that do not respond will be called. If this doesn’t work then someone will make a farm visit for a face to face. Both these methods are more effective, in general, but also are more expensive.

The biggest problem NASS faces when taking the acreage survey is that farmers usually haven’t yet made all their planting decisions. The agency knows this and is satisfied with best estimates. The individual reports are confidential by law and the data collected is exempt from legal processes.

The data can be aggregated at the county, state, and national level. Computers flag any large acreage changes at the individual level so that an analyst can check for a data entry error or make a follow up call. The state statisticians review the total number of crop acres for any major changes - total crop acres generally remain constant - and then submit the estimates in an encrypted file to USDA NASS in Washington, D.C. There more analysis is done and the final report is produced for release March 31.