Commercial disappearance is a residual. It is:
Beginning Stocks + Production + Imports = Total supply.
Total supply - Exports - ending stocks = commercial disappearance.
First off, MPCs and caseins are not counted in the import data. That skews the data. Next export data tends not to be very accurate because no tariffs are collected. So the government pays little attention to exports.
Trying to obtain more specific information on the actual products counted yielded the following:
We follow four major product platforms, cheese, nonfat dry milk butter
and whey. We compute commercial disappearance on a milk equivalent
basis (both fats and skim/solids. Production is total milk production
which is published in the NASS Milk Production report. Commercial
disappearance is for all milk regardless of use.
There are supply and use tables for cheese, NDM, and butter in the Dairy
at a Glance tables. Production there comes from the NASS Dairy Products
report. We don't do and S$U for whey because we do not have a stocks
report from NASS for whey, but you could safely assume stocks are zero
if you wanted to do one.
If I understand, and I think I do, cheese, butter and NFDM are the only products counted. Those products are counted because there is "Cold Storage" data to yield beginning and ending stocks. Remember too, storage data is not audited.
So, the milk going into yogurt is not counted. As the latest Cheese Reporter noted in the latest issue, “Sales of yogurt last year totaled a record 3.599 billion pounds, up 123 million pounds from 2007 and the 11th straight year that yogurt sales have set a new record.”
Additionally, the milk going into milk chocolate, and there is a lot of that, is not counted.
When a so-called excess of 2% production causes a crash in farm milk price, the data used to claim should be 100% accurate.
The “commercial disappearances” are not accurate. “Commercial disappearances” are a reasonably good guess given the resources available to USDA. But, no one should take the information as gospel.