- Intel has reportedly delayed the mass-rollout of its 7nm chips.
- This could hurt its pricing power and promote the growth of competition.
- It could very well shrink Intel's manufacturing lead further.
Intel (NASDAQ:INTC) has been the world’s leading chipmaker for several years now. Its huge investments geared towards improving chip fabrication technologies and introducing node shrinks at an aggressive timeline, ahead of its peers, have pretty much given it the leading spot amongst all the players operating in the segment. But its hegemony may not last for long. Intel has reportedly delayed the launch of its 7nm chips to 2022, from the previously estimated 2020, which could have serious implications for the chipzilla, both financially and operationally. Let’s take a look.
- Let me start by saying that Intel has historically operated with about a 2-year cadence, which is essentially the time it took to complete the entire “tick-tock” cycle and was followed by a subsequent node shrink. This was an aggressive launch schedule, which Intel’s cash-strapped competitors couldn’t match, that pretty much gave the chipzilla its leading spot in the semiconductor industry. I compiled and attached a chart below for reference. Now the thing to note here is that Intel’s next “tick” cycle at 10nm node is expected for mass release next year. If its 7nm chips are delayed to 2022, then it basically means that Intel’s 10nm node will have a cadence of anywhere between 48-60 months. This is nowhere near the company’s projected 2.5-year cadence, and could allow competitors like Samsung, GlobalFoundries and Taiwan Semiconductor (NYSE:TSM) to catch-up (and maybe even surpass) Intel in terms of chip fabrication technologies. So the first issue here is that Intel’s manufacturing lead is at risk with this delay.
- We have to also reflect on the fact that Intel has been ahead of AMD for the better part of the past decade in the x86 segment due to its advanced chip architectures and chip fabrication tech. But this might change going forward. AMD is already coming up with its ambitious Zen architecture next year, which has shown impressive performance improvements over its predecessors from the past. If GlobalFoundries manages to narrow-in on Intel’s lead over the next few years, as we just discussed, it would be able to provide AMD with cutting edge chip fabrication tech that would allow the latter to fiercely compete with Intel in the x86 and ARM-based server space. So the delay of the 7nm node not only impacts Intel’s manufacturing lead, but also promotes the growth of AMD.
- Next factor at play here is Intel’s fab business for ARM-based server chip manufacturers. Fabless companies such as Cavium, Qualcomm and AppliedMicro rely on their chip manufacturers to produce their final products. This market belonged to GlobalFoundries and Taiwan Semiconductor until Intel decided to enter the space and compete in the segment. But Intel won’t necessarily be able to make a dent in the segment if its manufacturing lead is exhausted. ARM-based chipmakers such as Qualcomm or AMD won’t have any compelling reason to chop off their ties with their manufacturers if they are able to get their chips manufactured on the same node as Intel’s. So Intel stands to also lose out on this ARM-based server chip outsourcing business as well.
- Intel has implied that its next-gen EUV lithography technology would be used at the 7nm node. I have discussed about this in detail in my last article so I won’t go over it again. But the crux of it all is that the delay of 7nm launch also means the deployment of Intel’s EUV tech gets delayed as well. Please read my last article to have a better understanding of how this impacts Intel’s business.
Now that we have taken a look at the impact of this delay on Intel from an operational standpoint, let’s take a closer look at its financial aspects as well.
- First of all, Intel has been able to price its products at premium due to the lack of credible competitors in the x86 space; it’s desktop and server offerings are priced significantly higher compared to rival offerings from AMD. Intel has been able to command this premium mainly because its chips fare better than AMD offerings. But if Intel’s lead is exhausted over the next few years, and AMD manages to close-in on it in terms of chip performance with the use of similar nodes, the chipzilla might lose its pricing power. This could mean that it’s 7nm chips won’t necessarily be as profitable as their predecessors.
- Intel might also get hurt in the server space. Once all the chipmakers round up on the same node, which I estimate to be around 2019-2020, all ARM-based chipmakers would have access to similar chip fabrication technology. These chipmakers would end up competing in the ARM and x86 space primarily based on their chip architectures. So, in order to remain competitive and relevant in the market, Intel might be forced to reduce the average selling prices of its Xeons in its server segment. This could have a serious financial blowback for the chipzilla as it generates the bulk of its operating profits, around 56% in FY15, from the server segment alone.
- Thirdly, Intel had started buying EUV lithography equipment back in 2015 for the 7nm node. I noted in my last article that the value of these purchases stood around $1 billion last year alone. So the delay of the 7nm launch means this equipment would be lying idle until the year 2022. It would be prone to wear and tear, whilst facing the risk of obsolescence, without generating any returns in the form of revenue over the next 6 years. Intel might be forced to scrap this equipment and repurchase the same, again, if the wear and tear is significant and hurts its yields. Meanwhile, it would be booking non-cash depreciation expenses over the period, which would cause a drag on its bottom-line. So the delay in the launch of 7nm chips could have a significant impact on Intel, financially.
Putting it all together
Slowly and steadily, its getting clear that Intel’s manufacturing lead will erode over the next few years. We’re heading towards a phase of ultra competitiveness, where Intel won’t be the dominant chip manufacturer anymore. This could hurt its financials and take away its pricing power. The exhaustion of its manufacturing lead might also encourage analysts to rerate the stock, with lower price targets. And as far as the 7nm delay is concerned, I expect reduced profitability out of the node’s entire life cycle due the aforementioned reasons.