Intel: Spectacular September Comeback

Summary:

• Intel Corporation’s turnaround, led by CEO Pat Gelsinger, focuses on restoring technological competitiveness, with 18A products expected by H2 ’25, despite recent revenue declines and layoffs.
• The reason the turnaround hasn’t succeeded is that it only really starts with 18A, which is on track, but ultimately only has an indirect effect on its financial performance/revenue.
• Intel’s stock trades below book value, presenting an investment opportunity based on potential gross and profit margin growth, with a possible return to $30-40 per share.
• Risks include aligning cost structure with current revenue, but improved economics and cost savings from the reorganization offer a path to profitability.
• Intel delivered a stacked September with three major leadership product launches. While a Qualcomm merger might secure its financial stability, it is not required.

Investment Thesis

Intel Corporation (NASDAQ:INTC) (NEOE:INTC:CA) has been called a turnaround since the company announced the 7nm delay in mid-2020 and subsequent hiring of Pat Gelsinger as CEO in early 2021. However, four years after that delay, Intel’s revenue has only decreased, culminating with the recent reorganization that will result in a reduction of over 15k employees.

Hence, at first sight, it seems to have failed. This may require an explanation to justify a bullish investment thesis going forward.

In short, the turnaround is actually on schedule. Intel was getting overtaken technologically, which meant it was losing pricing power and market share. This is the side Intel has been fixing. However, it hasn’t been completed yet, with the first 18A products scheduled for H2 ’25. Nevertheless, Intel already delivered no less than three strong product launches in September.

With the stock currently trading below book value, given its technological progress, this seems like an investment opportunity.

Background

Prior coverage

Previously, Intel and NVIDIA (NVDA) were compared. Intel is probably the riskier investment, but also likely has larger potential upside, whereas Nvidia might be the more solid investment for continued AI growth.

Discussing Intel’s Q2 earnings results, it was remarked that while the technology roadmap remains on track, the revenue growth thesis seemed to be fading, and hence the stock was given a neutral rating.

Thesis

In recent coverage, comparing Intel to Nvidia, it was remarked that the Intel turnaround thesis does not seem to have panned out. This might warrant an answer to inform the forward-looking thesis.

To recap, when the 7nm process node (since renamed to Intel 4) delay was announced in mid-2020, this pertained to a “defect mode” which resulted in the node’s development progress having fallen behind by 12 months.

This meant that it was known Intel would fall further behind, which happened from late 2021 (when the node was originally scheduled for launch) to late 2023 (when it actually launched). (Intel was one year behind TSMC’s N7 node, so the two-year delay of Intel 4 resulted in being three years behind.)

So, while it has been known since 2020 that Intel is in a turnaround (and the company has been investing as such), the bottom in terms of technology and financials was only in 2023 (and 2024 for Foundry specifically).

Technology

Meanwhile, TSMC (TSM) pulled ahead with its own N5 and N3 nodes. The first product to reverse this process of falling behind was Meteor Lake in the PC market, where Intel’s market share has remained quite stable. In the data center, Intel is only just getting more competitive, with the Sierra Forest launch in Q2 (but this marks a new product category) and Granite Rapids in September.

Financials

Financials are mostly constrained by end-market demand. Here, it had been Intel’s goal (as laid out in the early 2022 investor meeting) to keep growing revenue (from a combination of a steady PC, but growing data center, network, and edge, as well as its new emerging businesses with foundry, GPU, and Mobileye), and then to start growing even faster from 2025 onwards. However, in practice, a broad semiconductor downturn started in Q2 ’22, with no full recovery.

With the change in the reporting structure in Q1, the impact of Intel’s loss of process leadership has only in the last two quarters started to become quantitatively clear. This was reflected in a 66% operating margin loss in Q2 of the Intel Foundry business. Nevertheless, this does show the potential for a large improvement in earnings if this business would return to profitability.

Growth opportunities

To recap.

1. DCAI: Intel is expecting that the price per core will remain quite stable going forward. Hence, the increase in core count over time due to Moore’s Law should result in higher prices, likely boosting revenue. In addition, there is some potential for market share and gross margin gains from more competitive products.
2. NEX: While Intel remains bullish, it remains to be seen if this segment can return to growth (which in the previous decade was mostly based on market share gains).
3. Mobileye Global (MBLY): Its potential is visible, for example in design win dollars for the last two years being 3x revenue. Its revenue has upside as the market moves towards more advanced/autonomous systems.
4. IFS: Intel’s goal for IFS is to become the second-largest foundry by 2030 with $15B revenue, which Intel claimed (as recently as the Q2 earnings call) it remains on track to. But with logic revenue (which seems mostly focused on 18A and beyond) still a few years from ramping, this is not a near-term catalyst.
5. AI: Intel has not been successful yet in taking market share (also compared to Advanced Micro Devices (AMD)), with a guide for $500M Gaudi 3 revenue in H2’24. Therefore, this also does not seem like a near-term catalyst, although there might be potential for a stronger ramp in 2024 as Gaudi 3 launched in late September, which means it is mostly Q4 product.

Progress update

Technology

Intel has touted 18A as the finish line of its turnaround, restoring its technology competitiveness. At a recent investor event, it said its 18A defect density was below 0.4 per cm², which for this stage of development (several quarters before production) seems on track.

Next year, Intel will launch Panther Lake for the PC and Clearwater Forest for the data center, followed by Diamond Rapids in 2026.

Financials

The result of the technology progress should be for the manufacturing business (Foundry) to reach breakeven (from the 66% operating loss in Q2), targeted around 2027.

However, there is at best an indirect correlation between technology competitiveness and revenue (which is limited by the number of PCs and servers sold, market share and pricing). While increasing its spending in R&D and capex, its revenue has declined. This is what Intel has needed to adjust to, resulting in the reorganization.

Stock/Thesis

A question mark would be its ability to grow its revenue over time. Without a clear path to revenue growth in the near-term, the main investment opportunity would be the improvement in profitability over time.

Foundry Split

There has been a recent report, although subsequently rebutted, that Intel would be planning to split the company. Intel has since announced Foundry would become an independent subsidiary. In practice, too, Intel had already been in the process of getting both parts of the company to operate largely independently, with separate P&L.

Such a split would likely not have had any advantages. Given that Intel has been able to regain competitiveness in terms of technology against TSMC, it would not have made sense to divest its manufacturing assets, which are differentiated with technologies such as PowerVia, providing a path towards breakeven. With 18A starting commercial production in just several months, which Intel claims should be competitive both in terms of technology and cost, it is then just a matter of ramping this node over the coming years (to become the biggest contributor to volume).

Conclusion

Technologically, Intel remains on track to the roadmap it announced in 2021. Therefore, the turnaround might be called successful. On the other hand, Intel has not achieved its financial goals, with revenue falling mostly due to market conditions (in addition to its competitiveness issues). Since Intel had ramped up its spending to fund the turnaround, this has meant a decline in profit, free cash flow and gross margins.

Going forward, it probably takes on the order of two years to fully ramp a new process node and products. This means there is quite a gap between the technological and financial side. As such, it likely won’t be before (late) 2026 that 18A will start to show up in the financials.

Hence, the reason the turnaround hasn’t worked out (yet), is because it simply hasn’t even started yet in the first place. At least not in the market.

Similarly, AMD had also announced Zen on its roadmap years before its actual launch. Yet, its stock price kept declining, mostly just reflecting its current financials. But, it turned out, competitive technology did matter.

Hence, the same case could be made for Intel with 18A. Even then, as AMD’s case shows, the turnaround is just the start of a multiyear process of improving financials and stock price.

Stacked September

After announcing subpar earnings and guidance, rumors/speculation about Intel spinning off the fabs began reappearing (for the first time since mid-2020). However, with the finish line of the 5 nodes in 4 years strategy in sight, this made little sense. Instead, Intel delivered a September stacked with announcements, reaffirming the multi-year investment thesis.

Full portfolio refresh: reclaiming data center leadership

The most significant development occurred late in September with the launch of updates to nearly its complete product portfolio.

On the PC side, Intel launched Lunar Lake, its second-gen AI PC CPU for laptops. Initial reviews have shown decent CPU performance, leadership GPU performance and leadership battery life rivaling Qualcomm, confirming Intel’s claims of challenging Qualcomm in that area. In October, Intel will launch Arrow Lake for the desktop.

On the data center side, Intel launched Granite Rapids with double the core count of its predecessor, and overall 2x to 3x gen-on-gen performance. With its 128 cores, it also beats AMD’s Genoa (96 cores) and Bergamo (128 cores) CPUs in performance, as confirmed by initial independent reviews: Welcome Back Intel Xeon 6900P Reasserts Intel Server Leadership.

Although AMD will soon launch its own next-gen CPU (with 128 cores based on Zen 5 or 192 cores based on Zen 5c), the significance of Granite Rapids is that for now it is the highest performing data center CPU. Depending on whether one considers the first-gen Epyc (Naples) a leadership CPU, this marks the first time since 2018 or 2019 that Intel reclaims data center leadership.

This was the goal Intel/CEO Pat Gelsinger set in early 2021, and Intel has delivered exactly that.

Thirdly, on the accelerator side, Intel launched Gaudi 3. Intel’s claim is it delivers higher performance than Nvidia’s H100, but at a lower price. On paper (and from Intel’s own benchmarking), although Nvidia is launching Blackwell soon, it seems a solid product.

Financial developments

On the financial side, Intel has said its own halfway through its 15k employee reorganization. Several business units have also been reorganized. In addition, it is pausing its Germany and Poland fabs by two years. It was further awarded $3B for the U.S. Secure Enclave project.

However, addressing the speculation about a potential foundry spin-off, Intel instead announced it would make Foundry an independent subsidiary with its own independent board. Given the progress on the 5 nodes in 4 years roadmap, with 18A just months from ramping production (on schedule), this rumor/speculation did not make much sense, since this roadmap was announced as the alternative to divesting the manufacturing business.

Qualcomm takeover

Due to the decline in revenue (from nearly $80B to less than $60B) and stock price, Intel seemingly has become a potential acquisition target, although there has been speculation about whether such a deal would be possible and if QUALCOMM (QCOM) would acquire all business units.

For Intel as a business, given its current revenue and profitability, a deal that would be more like a merger (rather than, for example, Qualcomm just acquiring Intel’s products group and divesting the fabs) would perhaps make sense. Intel has few near-term drivers for strong revenue growth, and Foundry breakeven is only targeted for around 2027. Hence, merging with Qualcomm (as both companies have largely complementary businesses) would provide the combined company already decent profitability.

However, for Intel shareholders, the potential for outsized investment returns is exactly due to the unprofitability of the manufacturing business dragging down the overall earnings. The thesis is as Intel transitions to the EUV era and regains process leadership, it will return to breakeven and then profitability. Given the current stock price around $20, if the stock just returns to a range it has already traded for ($40-60), then this implies 2-3x upside. If, instead, Qualcomm would merge with Intel, the upside potential might be lowered.

Conclusion

Although the financial side remains an issue, on the product side at least Intel is delivering on its turnaround thesis with the launch of Lunar Lake, Granite Rapids and Gaudi 3. They are all leadership products at the time of their launch. Granite Rapids in particular stands out, delivering 2-3x gen-on-gen performance, vastly improving its competitiveness.

Valuation

The current market cap implies a valuation below book value, so revenue growth might not even be required to justify an investment.

This would be the base case of little to no revenue growth, but gross and profit margin growth towards its 60/40 target over time. This might lift the share price back towards $30-40 or so. For reference, the current forward P/E for 2025 and 2026 based on estimates is about 20x and 12x.

Any further upside in terms of revenue (such as the $15B foundry revenue) might lift the stock price back towards $60.

With 18A launching in 2025 and ramping to sizeable volume by 2027, the timeframe for such a stock movement would be by the end of the decade, which implies that the stock might deliver market-beating returns (2-3x in about half a decade).

Risks

Although its technology progress is compelling, the financial side has been less so.

Hence, in the near- to medium-term, Intel needs to show it can align its cost structure to its current revenue level. This is what Intel has called the second phase after restoring its technology competitiveness.

This should be feasible due to the improved economics of its EUV nodes, the slowdown in process node cadence resulting in lower start-up costs, and the finishing of the fab construction projects. Combined with the cost savings from the reorganization and efficiency improvements from moving to the internal foundry model, there seems a clear path to improved profitability.

Rating upgrade

Given the lack of financial progress, as the Q2 results and reorganization showed, a neutral rating could be warranted. The comparison with Nvidia also shows it is a riskier investment.

Nevertheless, the speculation about spinning off the fabs shows the opportunity for longer-term investors: this was the same speculation as in mid-2020, which resulted in the 2021 CEO change and 5 nodes in 4 years roadmap. Nearly 4 years later, this roadmap is being completed on schedule, which indicates the stock might be priced too much on near-term financials. Investors just have to be patient for the products and fabs to ramp over the coming years.

In combination with the reorganization, this should result in improved profitability. With the stock trading around $20, the valuation should be low enough that no revenue growth would be required for a bullish thesis.

Investor Takeaway

First, Intel delivered a stacked September, with reviews of both its PC and data center product launches saying Intel is back. For reference, it was Intel’s goal in 2021 to reclaim product leadership in 2024-2025, so this is on schedule. There were also several financial updates, most notably confirming Foundry would become an independent subsidiary, but no spin-off (which makes sense given the process technology progress on schedule). Meanwhile, there has been a rumor about a potential takeover by Qualcomm. While a combined Intel-Qualcomm company might be more resilient financially, it is not required and might decrease potential shareholder returns.

Secondly, to answer the question of why the turnaround hasn’t succeeded, with 18A in 2025 touted as the “finish line,” it simply hasn’t started yet. Furthermore, technological competitiveness ultimately only has an indirect effect on financial performance, mostly on the profitability side.

Overall, with around a $90B market cap and $20 share price, a return of its profitability (and margins) to or close to its historical level (targeted for 2027 and beyond) implies the share price could return to $30-40. A return to revenue growth (such as from Mobileye, foundry, or AI) might add to that upside, although this is not guaranteed, at least in the near-term, but at the current valuation not required either.

Hence, with Intel Corporation stock only pricing in current or near-term financials, despite the technology progress already being visible in the three September launches, with the process roadmap remaining on schedule as well and close to completing the 2021 roadmap, the stock seems undervalued.

Editor’s Note: This article discusses one or more securities that do not trade on a major U.S. exchange. Please be aware of the risks associated with these stocks.

---

Analyst’s Disclosure: I/we have a beneficial long position in the shares of INTC either through stock ownership, options, or other derivatives. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.

Seeking Alpha’s Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.

---