Glen T. Shen
USDA, AMS, Dairy Programs
Seattle Milk Market Administrator’s Office

There is much concern and confusion over the upcoming conversion from total protein to true protein and its impact on handlers and producers. For the most part, this concern is unwarranted. Basically, all producers will experience a similar 0.18-0.19 decline in their protein component levels, but this decline will be compensated by a corresponding increase in the price of protein. The confusion stems from numerous technical details underlying the change. Here, for those who might be interested, is a more in-depth explanation.

The impending change affects a reference chemical determination used by all nearly all handler labs and Market Administrators to adjust their infrared analyzers for protein measurement. If reference determinations (in this case, total or true protein by kjeldahl assay) were used exclusively to determine/verify producer pay tests, the change would have a variable impact at the producer level. However, the MAs and handlers have and will continue to rely on IR analyzers to determine/verify producer pay tests, as reference methods are far too slow and expensive to allow testing of thousands of samples per week. Therefore, to understand precisely what will change next fall, one needs to understand how the total and true protein reference methods relate to standard infrared analysis.

An infrared analyzer "sees" nitrogen-hydrogen linkages that are associated with "true" protein in raw milk. Thus, the instruments used by Market Administrators throughout the last decade have always identified the correct nitrogen-bearing constituents in milk. The upcoming change bears upon how the IR instrument is calibrated. Historically, the MA has set the output of the machine to correspond with "total" protein as dictated by 10-12 "control" samples. Because this measurement includes non-protein nitrogen (associated with urea, creatine, creatinine, and other trace compounds), IR instruments calibrated on total protein attempt to estimate a small NPN component, even though the IR cannot see this component. This will be remedied by changing the reference method to true protein. The output of a single IR instrument for potentially thousands of raw milk samples is ultimately determined by 10-12 well-characterized samples (generally prepared every two weeks for freshness). Therefore, unless a handler or individual producer is selected as a control sample source, they will enjoy no influence or competitive advantage over any other handler or producer.

Even as a control sample contributor, "advantages" or "disadvantages" are non-existent under the new system. The reason is that both reference method and IR method will now focus specifically on true protein – levels of NPN will become irrelevant. In the short term, it is possible that a producer who formerly contributed high-NPN milk to an MA control set will suffer a tiny setback (along with the hundreds of other producers verified on that MA instrument) because "phantom" protein he formerly introduced into the calibration will no longer be there. Chances are, however, that this producer’s milk was counterbalanced by another producer with low-NPN milk and hence, the former "total" protein-based calibration was close to correct anyway.

So why make this change in the first place? There are two reasons. The first is that true protein better reflects the economic value of milk protein. The second is that the IR calibrations discussed above will become more precise. Since NPN can vary (especially as a function of feed), calibrations of infrared instruments against total protein are generally inferior to butterfat calibrations due to larger random error (total protein-based IR calibrations show more scatter and lower values of r^2). Thus, the accuracy of protein pay testing should improve slightly with the conversion to true protein.

Individual associations or coops do not gain or lose a significant advantage relative to peers due to the conversion to true protein in January 2000. On a producer/herd level, Jersey farms do not enjoy a competitive advantage over Holstein farms in spite of their slightly lower mean NPN levels. And finally, from a producer-to-producer standpoint, one farm does not benefit or suffer relative to another farm despite differing levels of NPN (unless of course, a farmer is spending resources to boost NPN under the mistaken belief that this can increase his protein payment). The farmer who cost-effectively maximizes true protein levels (e.g. through genetic selection and/or feed optimization) stands to gain the most. This has held true under the past system of payment testing and it will continue to hold true with the conversion to true protein reference methodology.

Home, Current News, Monthly Bulletin, Current Prices, Historical Data, Plant List, Order Language, Laboratory, Job Opportunities, Dairy References, Site Map, USDA, AMS, Dairy Programs