Is Intel Back? 3 Reasons to Take Intel's Turnaround Seriously.

In an AI-led market, the notable underperformer in the semiconductor space this year has been Intel (INTC -2.93%).

While many AI-oriented chip stocks like Nvidia are up a lot on the year, Intel finds itself down just over 50% for 2024. To add insult to injury, the stock was just kicked out of the Dow Jones Industrial index last Friday, displaced by Nvidia itself.

Amid a tough PC market, revenue declines, and Intel’s plan to become a third-party foundry taking longer than some may have expected, investors have seemingly given up on the turnaround. However, Intel’s recently reported Q3 results and other recent news items indicate news of its death may be premature. Here are three reasons to believe heading into 2025.

1. Q3 results and fourth quarter guidance were reassuring

Intel had somewhat shocked investors back over the summer, after it missed on both revenue and profit expectations while also guiding conservatively for the third quarter.

But third quarter results showed things aren’t falling apart. Revenue came in at $3.3 billion, down 6% year over year but well ahead of analyst expectations for $3.0 billion and also up from $12.9 billion in the second quarter. While Intel’s bottom line was affected by a number of restructuring, accelerated depreciation, and goodwill impairment charges, adding those back would have yielded adjusted (non-GAAP) earnings per share of $0.17, far better than the $0.03 loss per share predicted by analysts.

Better-than-feared revenue and profits were great to see, as was fourth quarter guidance for revenue between $13.3 billion and $14.3 billion and positive adjusted EPS of $0.12. That guidance also topped estimates.

2. Newer products are showing success and higher profitability

Looking under the hood, what was especially encouraging was that Intel did the best where it had just unveiled new products. That seemed to prove out Intel’s technology and production ramp, the most important elements of its turnaround.

Specifically, while the company’s core Client segment declined sequentially from $7.4 billion to $7.3 billion, there was a lot going on under the hood. Older Raptor Lake desktop products built on the older Intel 7 process waned, but notebook PC sales, bolstered by the early September introduction of Intel’s new Lunar Lake chip, increased quarter over quarter. And operating profit for the client segment increased substantially despite the top-line decline, indicating newer notebook products built on the latest technologies are much more profitable.

Data source: Intel 3Q and 2Q 2024 press releases.

The good news on the desktop front is that Intel just launched its new desktop CPUs in October, codenamed Arrow Lake. Like Lunar Lake, the chip is actually produced by Taiwan Semiconductor Manufacturing (TSM -0.72%). But as one can see from the notebook results above, newer technologies are more profitable, even if produced by TSMC. Therefore, look for Intel’s desktop revenue and profitability to increase in the fourth quarter and into 2025.

In addition to Client, Intel also launched new Xeon 6 server products codenamed Granite Rapids and Sierra Forest over the summer on its own Intel 3 process, a newer process using extreme ultraviolet lithography (EUV) technology. Sierra Forest launched in June, but Granite Rapids didn’t launch until the end of September, limiting its impact on the third quarter. Still, Intel’s data center business improved markedly quarter-over-quarter, with third quarter DC revenue up 9.9% and segment operating profit up 25.7% relative to Q2.

3. Intel’s “bet the company” 18A node is on track

In addition to the good results, there was also lots of encouraging commentary on Intel’s 18A node, which is set for mass production in mid-2025. In September, Intel announced it had landed Amazon (AMZN -1.09%) as an 18A customer, and will collaborate on an AI fabric chip as well as custom Xeon processors for Amazon Web Services. On the conference call with analysts, CEO Pat Gelsinger also noted two other customer design wins from unnamed “compute-centric companies” for 18A.

18A is the node when Intel believes it will surpass TSMC in process technology. On the call, Gelsinger noted Intel’s internal 18A products, Clearwater Forest for data centers and Panther Lake for PCs, have “met early 18A milestones.”

While there have been lots of negative headlines around Intel’s efforts, rumors about bad yields, and calls for Intel to break up ins manufacturing and design arms, some may have missed a positive October 25 op-ed in Fortune Magazine by Intel’s former CEO Craig Barrett. Barrett dismissed the calls to break up Intel and praised Gelsinger’s strategy. This passage stood out:

Intel is on the verge of completing an unprecedented pace of node development to catch up to TSMC. It has taken the lead on next-gen technologies that will shape the semiconductor industry for years to come, such as high NA EUV lithography and backside power delivery. Yes, more work is needed–but this is a good start, and they must keep going.

Keep in mind, Barrett was the CEO who succeeded Andy Grove back in 1997, back in Intel’s glory days before the company lost its luster in the mid-2010s. So to see him give the thumbs-up to Gelsinger strategy and Intel’s technology progress was encouraging.

All eyes on Intel into 2025

For sure, it wasn’t all great news in the third quarter. For instance, Intel’s Gaudi 3 accelerator chips saw lower-than-expected uptake, as its new software required more customer effort to learn than anticipated, according to Gelsinger.

But there was much more good than bad, with Intel’s technology and profit improvements being the most salient issues. With shares now only trading at book value, Intel is certainly worth a look for the risk-tolerant value investor heading into 2025.