DigiTimes recently published an article in which it said that both Taiwan Semiconductor Manufacturing Company (NYSE:TSM) and its archrival Samsung (NASDAQOTH:SSNLF) are seeing lower-than-expected manufacturing yield rates on their new 10-nanometer manufacturing technologies.
This, the report suggests, could have an impact on Apple's (NASDAQ:AAPL) upcoming iPads.
"TSMC is scheduled to start making Apple's A10X chips for the next-generation iPad series slated for launch in March 2017," the report says. DigiTimes, citing industry sources, explains that "unsatisfactory yields for the foundry's 10nm process could disrupt the schedule."
But I don't think Apple investors ought to worry too much about a potential iPad schedule slip due to 10nm yield woes. Here's why.
The iPad is low enough volume that low yields can be compensated for
In fiscal year 2016, Apple shipped 45.59 million iPad units, or just under 22% the number of iPhone units that it shipped during that time.
From a raw volume-shipment perspective, iPad is nowhere close to iPhone.
Furthermore, while Apple tends to see a massive mix shift toward its latest iPhone models when new ones are released, the shift toward the latest iPad models is unlikely to be as significant once new models are released.
Apple doesn't break down its unit shipments by individual model, but a recent report from market-research company IDC claimed that even with the iPad Pro (9.7-inch and 12.9-inch versions) on the market, "the iPad Air and Mini lines have been the models with mass appeal, accounting for more than two-thirds of [Apple's iPad] shipments this quarter."
It is unlikely that iPad models with 10nm A10X chips will account for the bulk of Apple's iPad sales in the quarters following the launches of the new devices; it's more likely that most of the iPad models that Apple ships will use older chips, manufactured in more mature 14/16-nanometer technologies.
So, if manufacturing yields on the 10nm A10X chip are low, this will only impact a fraction of Apple's total iPad volume shipments. And since the required volumes are so low, Apple might very well be able to compensate for low yields by simply having TSMC run more A10X wafers (and potentially having TSMC eat the increased costs associated with the lower-than-expected yields).
Even if the new iPads are delayed, so what?
Although I don't think a chip-related delay of the new iPad models is likely, it's important to put the relationship of Apple's iPad sales to Apple's overall business into perspective. Shipments of iPads are far dwarfed by iPhone shipments, and average selling prices on iPad models are generally much lower than they are on iPhone models.
That means that, in fiscal year 2016, iPad revenue made up -- and this figure is rounded up -- just 9.6% of the company's total revenue. Unless the tablet market can be fundamentally reinvigorated, which seems unlikely in the near term (Apple's iPad sales have continued to plunge even as it introduces solid new products), being a month or two late with new iPads just isn't going to make a lasting difference to Apple's financial performance.