What the channel changes also helpfully tell us, is when retailers have ordered too many for current demand. Or been nudged into doing
Quite often when Jobs or Cook needed a numbers boost, suddenly that quarter had unexpectedly good sales. But then the next quarter would usually have a large increase in channel inventory, because the stores overbought. Didn't matter, because by then the necessary great sale headlines were past.
The key point as far as stores go, is that they will not continue buying something that won't sell. Which is as true for other phones as it is for iPhones. That's why it's silly for people to claim that stores keep buying tens of millions of say, Samsung phones, just to put them on a warehouse shelf without sales.
When it comes to channel inventory changes for iPhone or iPad (Apple hasn't been reporting changes for Mac), for the most part they've been about what we'd expect based on, e.g., new product releases and seasonal sales trends. They typically draw down channel inventory when sales are slower (recently, typically in the 2nd and 3rd fiscal quarters) and build it when sales are faster and when they're releasing new products (recently, typically in the 4th and 1st fiscal quarters). In earlier years, when sales growth in general was greater, they built inventory in most quarters trying to catch up with increasing demand. In recent years, when iPad sales have been declining, they've been drawing down iPad inventory in most quarters.
But there haven't really been any out of whack channel inventory changes. Over the last 5 years, the largest (quarterly) channel inventory builds for iPhone were still less than a week's worth of supply and left channel inventory either below or at the low end of Apple's target range (which range they've been transparent about, it isn't something they pulled out when they needed an excuse to build channel inventory). So that could hardly be considered channel stuffing. It's consistent with how Apple has long said it tries to operate and responsive to fluctuations in the pace of sales.
The only quarter I can recall that someone might question would be Q1 2016. That was the first full quarter for the 6S models and Apple built channel inventory 3.3 million units to put it just ahead of its (sell-in) unit number from the year ago quarter. But even that build, which was less than a week's worth of supply, just barely got Apple into its target range. It was larger than the year ago build (i.e. shrink) by 3.5 million units, but the year before had seen the 6 models creating extraordinary (and perhaps somewhat unexpected) demand which Apple had a tougher time keeping up with. Had Apple been able to build inventory in that previous quarter I suspect it would have.
I lost track a few years ago, but yeah it was headed to by way past 16 million by now.
View attachment 711275
That graph (as you suggest) stops quite a while ago. iPhone channel inventory accelerated after that, being well above that trend line and eventually getting above 20 million units. It's at a seasonal low point now.
iPad channel inventory peaked around 6 million units, but it's been trending down for over 3 years to under 3 million units as of the end of the March quarter. I would note that Apple didn't tell us the 3rd quarter channel inventory change for iPad in Tuesday's conference call. I didn't hear it mentioned when I listened to the call so I searched the transcript and couldn't find a number given. That's unusual, it's the first time Apple hasn't given us either a channel inventory or a channel inventory change for iPad. (I keep a spreadsheet with such information.)
That makes me wonder whether Apple had a significant iPad channel inventory build in this last quarter which helped with that big iPad unit number. Apple did say in the call that it exited the quarter in its target range for iPad channel inventory. Unless iPad sales dropped off significantly toward the end of the last quarter or beginning of this quarter (or Apple expected them to), that - Apple being in its target range - also suggests a large channel inventory build in the previous quarter. At a 875,000 iPads per week sales pace (which was the average sell-in for last quarter), Apple would have had to build inventory by over 1.6 million units to reach the bottom of its target range. Even at 700,000 iPads per week it would have had to build inventory by 0.8 million units. And for those keeping track, a 1 million plus channel inventory build (in this past quarter) would mean that iPad's sell-through was actually less than it was in the year ago quarter.
