Originally posted by Mistral75 You'll notice the same discrepancy each month for any and all categories: compact cameras, DSLRs, mirrorless cameras and lenses. That's because the reference price is not the same for production and for shipments.
I was intending to use this as an example of how misleading statistics can be without reference, but I didn't say that was my intent, so my bad. I did some more study and this "mark-up" is also not consistent when looking at only one month. The markup for all cameras declined from 34.3% in Jan. '18 to 25.6% in Jan. '19, even though average value per unit shipped increased 6.5%. If I look at MILCs, markup increased slightly from 31.5% to 33.7%, but average value ballooned by 31.6%. In other words, looking only at CIPA stats, it became much more expensive to manufacture cameras in January 2019 compared to January 2019, even when the technology isn't changing (markup for DSLRs declined to 28.8% from 37.6%, in spite of a 2.3% price increase) .
Because CIPA's stats represent almost all of the total population of new cameras, every statistic comes with a very high probability, even though the period being sampled only represents one-seventeenth of the period being analysed (a full calendar year). The point I am trying to make is that this statistical validity has absolutely no practical validity. Unless one has more information (such as a particular manufacturer's product mix, for both production and shipments) January's statistics are meaningless because they fail so many tests for reasonableness.
Originally posted by Mistral75 it's something one can observe each year in January: production has to catch-up after the holidays.
In general, yes, in detail, no. Total camera production in Jan '19 exceeded shipments by 6.4%, in Jan '18 by only 2.5%. For DSLRs,
shipments exceeded production by 0.25% (810 cameras) in Jan '19, in Jan. '18 the opposite occured with production exceeding shipments by 2.6%. For MILCs, Jan. '19 production exceeded shipments by 18.6%, Jan. '18 the opposite occured, shipments exceeded production by 2.1%. If we had a breakdown by model for every manufacturer, these discrepancies with common sense could be explained, but only CIPA staff have that information and if that level of detail ever became available outside of CIPA staff, CIPA members would stop contributing their data.
Yet Another Postscript: I used "markup" because that is what it is (the difference between the "trade" price and the "out the factory door" price), but I calculated it the same as calculating margin. In other words, the result is the ratio between the
difference in prices and the "trade" price. For a retailler, 50% markup is the same as 33% margin, but manufacturers are selling to distributors who then apply their own markups and the distributors' customers apply markups, so whatever this CIPA markup is, the manufacturers aren't getting it. Did anyone tell me that the value of production given by CIPA is the "out the factory door" price? No, but there has to be some consistency and suggested retail pricing and trade channel discounts are much more widely disseminated (and vary less between manufacturers) than internal cost accounting. Also, I correlated changes in this markup ratio to changes in manufacturing costs, which isn't necessarily the case, but trade channel discounts don't change dramatically (quite the opposite, in fact), so whatever is being observed here is a result of changes in product mix (both in production and shipments), which we have no way of knowing.
For the benefit of anyone who is able to make all the way through my post, the bottom line is that CIPA's statistics for January 2019 are meaningless without further data.