In my Hogs and Pigs Report review sent earlier this week, I alluded to the big swing between the year-over-year change between the inventory of 120-179 lb. pigs (-0.1%) and the inventory of pigs weighing over 180 lb. (+3.8%). That 3.9% total change between successive inventory groups struck me as very large and I questioned whether our industry, with its "constant throughput" goals, would actually see that much variation. So I went back to the historical data to see just where these numbers fit.
Figure 1 shows year-over-year changes in the inventories of pigs of different weights. This graph compares one quarter's under-60 lb. inventories, for instance, to the under-60 lb. inventory for the same quarter one year before. Data for the four weight classes are included. A few conclusions can be drawn:
- As with virtually every other measure of production in the U.S. pork sector, this one demonstrates a decreasing level of variation. Annual swings of 20% were common prior to 1980, and 10% swings persisted even through 1998-99. The largest year-over-year changes this decade have been just 4%, and even those swings appear to be a thing of the past.
- The weight categories generally lead one another in a quite logical fashion. While the graph is pretty busy, the blue line representing under-60 lb. inventories always moves first and then is followed by the heavier inventories.
- The magnitude of the percentage inventory changes are very close, but recent changes in the heaviest category appear to be getting larger than the changes in the lighter inventories.
That last finding was rather curious, so I computed the differences in percentage changes for successive weight classes in the same quarter. For instance, I compared the percent change in the under-60 lb. category to the percent change in the 120-179 lb. category for the same quarter. Comparing weight classes this way should capture the cumulative effect of changes in pig crops and survival for previous time periods. I expected to see these numbers get smaller over time.
One potentially confounding factor, though, is imports of weaner and feeder pigs from Canada. Their dramatic growth since 2000 would suggest that negative class-to-class shifts should be less frequent and positive shifts may be larger.
Figure 2 shows the frequency counts for changes of 3% or more for each pair of successive weight categories. As expected, the number of changes of 3% or more, regardless of direction, has declined dramatically over the years. But the frequency decline for the difference in the top two weight categories (i.e. 120-179 vs. 180+ line) has been much smaller than for the lighter weight categories.
The second portion of Figure 2 shows the frequencies of declines of 3% or more between the weight categories. These, too, have fallen over time due to the relative stability of today's high-investment production systems and the fact that our industry has generally been growing. But note that the decline in frequency is less for the 180 lb. and over category than for the lighter-weight categories.
More to Learn
So is there any lesson to be drawn here? I think so. Year-to-year and group-to-group variation in the U.S. pig supply is definitely smaller than it once was. That comes as no surprise. However, changes of 3% or more in the year-over-year shift in weight classes are still not rare, especially for the 120-179 lb. vs. 180+ categories.
I'm still a bit concerned about the shift from +3.8% to -0.1% for the two numbers in last week's report. Much of that concern, though, lies in the fact that porcine circovirus-associated disease (PCVAD) was causing widespread death losses and that porcine circovirus vaccine is now on the scene.
We shouldn't expect the decline in slaughter in mid-April to be as abrupt as the report suggests but, if those numbers are correct, slaughter totals should begin a pretty sharp decline soon. Federally inspected slaughter through Thursday was 1.4% lower than last week, but that's a far cry from 3.9%.
Click to view graphs.
Steve R. Meyer, Ph.D.
Paragon Economics, Inc.
e-mail: [email protected]