We’re entering our year number five of calling Intel (INTC) the “IBM of the 2020s” – our reference to IBM’s struggles in the 2010s during the fast rise of the cloud. This five year span includes our first official CSI YouTube video, at the bottom of the last bear market in September 2022 where we made the call to avoid Intel stock. Yeah, bold talk, but for long-term mindset investors… well, the chart does the talking.
But we like to think we’ve moved beyond such cliche analysis, and instead want to continue to document the current dilemma facing the old semiconductor industry IDM (integrated device manufacturer) business model. Fabless product engineering specialists rule the day, making use of the rising third-party foundry businesses (like TSMC, most notably).
In other words, it remains unclear what future prospects the IDM biz model will have. In our last segment on Monday (over on YouTube, expanded version on Semi Insider), we discussed Intel’s Q1 2025 earnings. Today, let’s outline the long and winding road that new CEO Lip-Bu Tan will have Intel on as it attempts to rebuild.
Time to throw in the towel on the whole “returning Intel to greatness” thing
It’s very early on in Tan’s tenure as CEO and re-appointment to the board of directors (BoD). He was hired in March 2025 in the new role, previously having been seated on the Intel BoD from 2022 to 2024 (more on the importance of the BoD over on Semi Insider). Thus far, it appears Tan’s plan will be a semi-continuation of the previous strategy to rebuild Intel and return it to greatness. This time, though, Tan is emphasizing simply “building a new Intel.”
A big refresh of Intel’s corporate culture appears to be the first priority for Tan. He acknowledges that the company has been flat-footed in decision making and adopting new ideas. (Really, like two decades of flat-footedness.)
But how do you change a corporate culture that is more than half a century in the making, with some 100,000 global employees? Here’s the game plan, which we’ve summarized thus far from Tan’s shareholder letters and Q1 2025 earnings call.
1. To rebuild culture, you start by parting ways with people and their old work processes.
Intel’s workforce is already down from the 2022 peak of more than 130,000, but it’s still too many. Management has maintained it hasn’t finalized the layoff numbers and details, but it could be upwards of 20,000 (~1/5th of the current total global Intel employee count). We expect most of these will be middle management, followed by fab layoffs as it’s clear Intel still has too much fab capacity for current or expected demand.
2. Re-focus on engineering products
With a downsized workforce, Tan wants to rebuild the business workflow around engineering of products – like a fabless company (AMD, Nvidia, Broadcom, Qualcomm, etc etc etc). Why? Intel essentially isn’t participating in the AI data center boom, so that needs to change. And divesting more non-core segments (like the majority stake sale of Altera to private equity firm Silver Lake) also needs to happen. And to foster future good ideas, Tan decided Intel Capital wouldn’t be sold off after all.
3. Make the capital structure less messy
After addressing the culture, Tan is next going to work on Intel’s capital structure. Debt needs to be paid off. And to get fresh cash into the business (via operations, versus utilizing sale of assets in years past), both capital expenditures (CapEx, like on fabs and manufacturing equipment) and operating expenses (OpEx, like employee payroll) need to be cut. Because, seriously, how long can this negative free cash flow (FCF) situation last?
Additionally, later this year after the Altera sale to Silver Lake is finalized, Intel will stop reporting Altera revenue and operating income/loss in its consolidated financial statements. Since Altera stopped reporting after Q3 2024, it’s unclear if this will help the cash flow situation. But essentially, we’d view this as part of the Intel employee layoffs and OpEx reduction (sorry Altera employees, you’re going to become a Silver Lake “problem” later in 2025).
Tan emphasized the need to pay down debt as well. It sounds like proceeds from the final sale of the old memory business to SK hynix last quarter, and future proceeds from Altera, will be used for this purpose – at least in part. Serious bloat crept in under Gelsinger as the Foundry problem just kept getting kicked down the road.
But for the sake of emphasis (because it’s really important): Before more serious debt paydown can occur, here’s the FCF situation again (cash from operations minus CapEx). What an ugly run. No wonder Tan said he’s been discussing some sort of Foundry partnership with TSMC (Foundry likely accounts for most of the CapEx spending). Flag this for future review if a TSMC deal of some sort is ironed out.
Be careful before buying the Intel stock dip
All of this is to say Intel stock has been a value trap for years – and it still could be. Enterprise value (EV) is still over $125 billion, due to all the net debt on balance. And price-to-book value (“book value” essentially just a fancy term for assets minus liabilities, or shareholder equity value) is below 1.0. Many in the value investor crowd would call anything lower than 1.0 P/BV cheap. But “cheap” book values are a classic value trap, especially in the world of tech (remember we called Intel the “IBM of the 2020s”?). See Semi Insider for our explanation of book value, and why it’s a terrible tech stock valuation metric.
Perhaps Tan can begin to right the Intel ship. But there is a ton of work to be done on all fronts. Intel is now under total siege. For most retail investors, we struggle to see why this one is worth your time. Statistics favor a high-performance ETF greatly beating Intel stock, just as has been the case for many years.
