This rant by a cattle analyst perfectly illustrates one of the reasons why I have loved the cattle market for decades.
CATTLE ON FEED NUMBERS HIT A DISCONNECT
If behind the scenes there is a analysts who can gather all the data points surrounding the cattle and beef markets and put together an econometric price model that provides positive results for trading cattle futures, it is a deep dark secret. If such a model truely exists, you would not expect the owner to disclose its existence. Because there is good evidence that some of the best mathematical and modeling minds in the country have tried and failed to uncover the secrets of pricing in cattle and beef and because of its nature [a dynamic production model allowing quick changes in production by increasing or decreasing days on feed], it is doubtful such a machine exists.
Nonetheless, people who play in this space like to think they are all looking at valid numbers when viewing reports such as the monthly cattle on feed report. For many consecutive months the pre-release guesses done by experts have understated the USDA number by 5-8 percent. These are large variances and there are no explanations for the differences. The report includes three components only one of which provides the data gathering challenge -- the placement number. Marketings are simply a additional problem solved by totalling all the required daily slaughter renderings from the processors that breaks down slaughter into fed cattle categories of steers and heifers.
Analysts submitting pre-release guesses never accompany those guesses with the methodology for making their predictions. USDA does explain how the data is gathered but the methodology leave more questions that answers about the accuracy. USDA uses mail out surveys, phone calls, auction barn receipts, feeder association records and other means for gathering the data. So why the large variances in the placement number?
The largest feeding states have cattle feeder organizations that cover most of the feedlots in the states and should be a solid foundation for gathering placement numbers. Surprisingly, those numbers frequently have large variances to the USDA numbers. USDA would benefit the industry by reconciling those differences and providing more detail explaining the differences. Death loss and how it is handled and reconciled is an important wild card in cattle on feed numbers.
Ultimately, animal identification is the only and final solution to rendering valid numbers to the industry. Placements, marketings and death loss would be reported daily and the on feed numbers would be available to anyone on demand at any given point in time.
As I've said for a very long time - it's not the numbers, it's the changing perception of the numbers.