Opportunity

Intel’s stock is trading near a ten-year low and, for the first time since at least 1990, below its book value per share. That’s left the chipmaker looking worse than very cheap (this feels like left-for-dead territory). But, the thing is, Intel’s important: it’s a too-essential-to-fail kind of business. I conducted a sum-of-the-parts valuation on the company, and even with conservative assumptions, it pointed to a fair value that’s 60% higher than Intel’s current share price. That suggests you may want to consider buying a small position in the stock and gradually adding to it as the company meets key milestones – that is, if you believe it will.

Thesis

• Intel is in the grips of a much-needed turnaround and is taking bold steps to improve shareholder value. That includes restructuring, cutting costs, and suspending its dividend to free up cash for its expansion plans.
• The firm’s market share in personal computers (PCs) has stabilized. With the PC market expected to return to growth next year and Intel's strong positioning in AI-powered PCs, where its chips command a price premium, its computer business appears poised to expand at a low single-digit rate.
• Intel’s recent market share losses in data centers are most likely permanent. But it could recoup some of that lost revenue by manufacturing chips designed by big tech firms for their own cloud data centers. Intel’s newly launched AI chip shows early promise, and could do well in specific segments of the AI data center market.
• Intel’s foundry business is also a reason for optimism: it’s strategically important to the US, it’s made strides in attracting high-profile customers, it’s eligible for cheap funding from the government, and it’s been an early adopter of the latest technology for making the most advanced chips.
• Even with conservative assumptions, a sum-of-the-parts valuation on the company points to a fair value that’s 60% higher than Intel’s current share price.

Risks

• Intel could continue to lose market share in PCs and data centers.
• The firm’s newly launched AI chip could flop, foiling its plans to gain a foothold in the booming AI data center market.
• Operating profit margins could fail to improve because of declining sales, rising costs, and huge investments in manufacturing.
• The foundry business could struggle to secure customers.
• Cost overruns and delays could hit the firm’s new manufacturing projects.
• Issues could come up with the new technology Intel adopted to make the most advanced chips.
• Balance sheet and cash flows could deteriorate, forcing Intel to issue equity at a depressed valuation.

This is a new type of analysis: detailed research into a single company’s stock, aimed at helping you evaluate a potential opportunity. Let us know your thoughts, and what you’d like to see next.

PART 1: THE PAST

What’s happened with Intel?

Intel has missed out on the AI boom, while its old-school businesses have been losing market share. Its revenues and profit margins have slumped as a result, leaving its stock price near a ten-year low.

For almost 30 years, starting back in the 1990s, Intel was the top dog among chipmakers, riding high on the PC boom. But about a decade ago, cracks began to show. The company got a little too comfortable and a little too cautious – and missed out on huge trends like smartphones and AI. And as if that weren’t enough, the whole semiconductor industry started shifting from the US to Asia, making it even harder for Intel to keep its grip on the top spot.

Making matters worse, while many competitors adopted a “fabless” model, designing their own chips but outsourcing the manufacturing (or “fabrication”) to foundries like TSMC, Intel doubled down on both. It proved a losing proposition: production delays and problems in its foundries left the company trailing in the race to make the fastest chips with the smallest transistors, while the fabless Advanced Micro Devices (AMD) sprinted ahead. Distracted by its troubles, Intel failed to see the extent to which graphics processing units (GPUs) would come to dominate the explosive AI market, and its focus on central processing units (CPUs) left it far behind.

Intel is set to wrap up a third straight year of shrinking sales, estimated to end 2024 with just $52 billion in revenue – some 30% less than it made in 2021. Its investors have all but given up on the firm: a disastrous second-quarter update sent its stock plunging 26% on August 2nd, its worst single-day fall in 50 years. That’s left Intel’s share price languishing near a ten-year low – a huge fall from grace for a firm that was once the most valuable US semiconductor company. By contrast, Nvidia is estimated to end the year with $126 billion in revenue, some 370% higher than in 2021. Its market value, meanwhile, is currently 34 times Intel's ($3.4 trillion versus $100 billion).

On the bright side, Intel’s massive drop has left its shares looking cheap. The firm is trading below its book value for the first time in at least 34 years. That means investors are now pricing one of the world’s biggest chipmakers for less than the value of its facilities and other balance sheet assets. So that has to make you wonder whether its stock is a timely opportunity – or a go-nowhere value trap.

PART 2: THE PRESENT

What’s Intel doing to turn things around?

The firm is reshuffling itself into two key businesses, cutting more than $10 billion of costs, and suspending its dividend to free up cash for its expansion plans.

1) It’s restructuring the firm into two key businesses.

In April, Intel reorganized itself into two businesses, each operating as an independent subsidiary with its own management and some strict informational barriers. The products group designs and sells chips for PCs, data centers, and networking equipment. And the foundry division runs Intel’s chipmaking factories.

By separating the two units, Intel can let its products group choose the best foundry to meet its needs. The foundry unit, meanwhile, can assure its prospective customers – some of whom are competitors – that they’re dealing with an independent supplier, that their design secrets won’t leak to Intel’s product folks, and that Intel won’t prioritize its own production needs over theirs. Those steps are all key to achieving the CEO’s goal of turning the foundry business into the second-biggest contract chip manufacturer in the world by the end of the decade, after Taiwan’s TSMC.

2) It’s cutting costs by more than $10 billion.

In August, in that disastrous second-quarter update, Intel announced a plan to reduce its short-term spending (operating expenses) and major investments (capital expenditures) by more than $10 billion next year. That includes laying off 15% of its workforce (more than 15,000 jobs), shrinking its office locations by two-thirds, pausing factory projects in Germany and Poland for two years, and putting another – in Malaysia – on indefinite hold.

It’s a long but essential checklist, given the huge decline in Intel’s profitability. Consider this: the firm had an average 28% operating profit margin between 2014 and 2021, but that plummeted to just 3.7% in 2022 and a mere 0.2% in 2023. And much of that slide stemmed from the fact that it just kept pouring billions into expanding its manufacturing capacity while ignoring the dip in its chip sales and the lack of new business for its foundries

3) It’s pausing its dividend.

Intel also announced in August that it’s suspending the dividend it has paid since 1992 – at least until cash flows markedly improve. While this move surely alienated some income-focused investors, it was likely a shrewd one, as the company looks to preserve all the cash it can to fund the foundry division’s expansion plans. It’s more than a drop in the bucket: the firm paid $3 billion in dividends in 2023 and $6 billion the year before.

PART 3: THE FUTURE

Can Intel walk and chew gum at the same time – growing both units?

Analysts are divided. Some say Intel needs to focus on regaining its leadership status in chips, even if that hurts the manufacturing business. Others argue that the company should prioritize attracting major customers for its foundry business instead – and there’s a strong case to be made on that side.

The products group

This arm has three major segments:

• The client computing group (CCG), which designs and sells chips for PCs.
• Data center and AI (DCAI), which focuses on processors designed for high-performance computing tasks.
• Network and edge (NEX), which makes chips for computers and telecommunication networks.

The NEX segment has remained relatively stable over time, with average revenues of $7.2 billion over the past four years. That said, it’s also the least significant segment for Intel’s overall revenue, accounting for just 10% of total sales. For that reason, it makes sense to focus on the other two parts. Here’s how their revenues have looked over the past five years.

Notice how revenues at both have fallen heavily since 2021. For the CCG business, 2020 and 2021 were peak sales years because the pandemic ignited massive new demand for PCs, with the rise in remote work and schooling. From 2015 to 2019, CCG saw average annual revenues of $34.7 billion a year. Compared to that figure (and not 2021’s), last year’s sales of $29.3 billion don’t look too bad. It’s lower, sure, but that coincides with a broader cooldown in the PC market after the big surge.

Even more worrying is the serious decline in revenue for the once-booming DCAI segment. From 2014 to 2019, revenues at this business grew at a 10.3% compound annual growth rate (CAGR), rising to $23.5 billion. But sales this year are projected to be just $12.6 billion, nearly half of the peak hit just four years ago.

Three key things are steering this slump. First, thanks to all the troubles in its manufacturing facilities, Intel has given up market share in this segment to AMD, which uses TSMC to make its chips. Second, data center budgets have shifted squarely toward Nvidia’s industry-leading GPUs, which are essential to powering generative AI applications. And third, major tech customers are increasingly designing their own chips for their cloud data centers instead of using Intel’s processors.

The good news is that the PC market is expected to return to growth next year. Makes sense: customers who bought devices during the pandemic will soon be ready for a newer model since the typical replacement window is just three to five years. And that refresh cycle will coincide with the arrival of some snazzy AI PCs – those that can perform AI tasks offline, like Microsoft’s Copilot+ laptops.

Research firm IDC is predicting that nearly two out of three PCs sold in 2028 will have on-device AI features. That could present a welcome opportunity for Intel, which is producing chips for AI PCs and is on track to sell more than 40 million of them by the end of 2024. Already, Microsoft’s Copilot+ PCs – which initially included Qualcomm chips – house Intel’s latest processors (as well as ones from AMD).

So, with PC sales returning to growth and Intel's positioning for AI PCs (where its chips command a higher price), I would expect the CCG business to expand again at a low single-digit pace. But I’m not as optimistic about the DCAI segment.

For starters, I expect major tech firms to continue to design more and more of their own chips for their cloud data centers, rather than relying on Intel’s processors. But also (and this is perhaps the bigger issue), I see Intel as having missed out on a massive growth opportunity with data centers used to train and run AI applications. Those facilities rely on GPUs made by Nvidia and, to a lesser extent, AMD, and I think Intel is simply too far behind to catch up.

That’s not to say that the entire enterprise is a lost cause for Intel. The firm recently unveiled a GPU optimized for AI applications called “Gaudi 3”, which could appeal to three kinds of customers. First, those building and training less advanced AI models that don’t require the highest-performing chips from Nvidia. Second, those who are transitioning from training AI models to running the models, where, once again, less advanced chips are also sufficient. Third, price-sensitive customers who simply can’t afford Nvidia’s expensive GPUs.

Still, it's important to remember that while a product may look promising on paper, its ability to generate sales volumes is a different thing altogether, and Intel will need to demonstrate its ability to do so. Recently, Intel scored a small win after Inflection AI – a well-funded startup launched by one of DeepMind’s founders – announced that its latest enterprise AI platform would ditch Nvidia’s GPUs for Intel's Gaudi 3 chips.

The foundry division

The foundry unit made $18.9 billion in revenue last year. But here’s the catch: $18 billion came from Intel itself (i.e. from making chips for its products group), which means less than $1 billion in sales came from external customers. Throw in some high fixed costs, underutilization, and big depreciation expenses resulting from the foundry’s major spending projects, and the division reported a $7 billion operating loss in 2023. That’s a very different set of figures compared to TSMC, which pulled in $69 billion in revenue and almost $30 billion in operating profit.

To get a sense of just how much of an impact Intel’s big spending on foundry expansion is having on profitability, consider this: the firm’s depreciation expense, as a proportion of its total revenue, was 14.5% last year. In contrast, that same ratio at AMD – a fabless rival that outsources manufacturing – was just 1.9%.

For Intel, a key question is whether those foundry investments will pay off. Remember: the CEO wants to turn the business into the second-biggest contract chip producer in the world by 2030. (TSMC is, of course, in that top spot, followed by Samsung). Intel estimates the foundry could bring in $15 billion in external revenue a year by the end of the decade, with operating profit margins reaching 25% to 30%.

That’s ambitious, especially considering that many potential customers already have supply agreements with TSMC and Samsung, and might be wary of working with an unproven foundry. But the business has landed some big, multibillion-dollar wins recently. Earlier this year, Microsoft selected Intel to manufacture a new chip that the software maker designed in-house. And in September, Amazon’s cloud services business chose Intel to produce custom AI chips.

Securing these high-profile clients not only serves as a vote of confidence for Intel’s foundry, but also allows the wider firm to recover some lost revenue. See, Microsoft, Amazon, and Alphabet are among the biggest buyers of chips: they use them to power their enormous cloud data centers. But these tech giants have increasingly opted to use their own chip designs of late – a trend that’s hurt Intel in its most lucrative business. Acting as a foundry for these firms could help grab back at least a portion of that lost revenue. And we’re not talking about small beans here: JPMorgan analysts estimated in May that the market for building custom chips for big cloud providers could be worth as much as $30 billion this year, with potential growth of 20% per year for the foreseeable future.

It's worth noting here that those fabless chipmakers have a vested interest in seeing Intel’s foundry business succeed. And, it’s worth saying that this industry has, at times, banded together to support underdogs and squeeze more players onto the playing field.

See, without Intel’s chipmaking facilities in the mix, the fabless crew is likely to become even more reliant on a single supplier: TSMC. And that would be foolish from a competition standpoint, but also from a risk standpoint. TSMC and most of its foundries are located in Taiwan, so any disruptions to operations there – whether because of geopolitical tensions, natural disasters, or supply chain hiccups – would have a cascading, global impact.

One nice thing about Intel is that it’s got a geographically diverse factory network. And that’s big: pandemic-era disruptions and chip shortages awoke business leaders and politicians to the importance of shoring up domestic production of such a key component of modern life.

That’s why the American government passed the CHIPS Act in 2022, providing nearly $53 billion in funding to chipmakers who want to build or expand US-based factories. Intel has been the biggest beneficiary, grabbing $8.5 billion of that pot, along with $11 billion in loans and a 25% tax credit on foundry investments.

And, look, Intel isn’t just expanding its manufacturing capacity, it’s also equipping its foundries with the latest, most high-tech equipment. The firm let its foundry business fall behind TSMC’s by being late to adopt extreme ultraviolet (EUV) lithography – a key technology used to make the most advanced chips. And it’s determined not to make that mistake again. That’s why it’s making a big bet now on the next generation of EUV – a technology called “high numerical aperture (NA) EUV”, which promises big breakthroughs in fabricating advanced chips. Not only was Intel the first company to secure a high NA EUV machine, but it also has reportedly purchased all the units scheduled to be churned out this year, leaving none for its competitors.

Let me bottom-line this for you: overall my view on Intel’s foundry business is positive. It’s got strategic importance for the US and its developed peers, it’s made progress in attracting some high-profile customers (who have a vested interest in its success), it’s managed to score some cheap government funding, and it’s been quick to adopt high NA EUV technology. Besides all that, the division's operating losses are expected to peak this year as it wraps up some heavy investments, and it looks like there’s a path toward profitability later this decade – albeit one fraught with execution risks.

Would Intel be better off selling its units (or, for that matter, the whole firm)?

A full takeover seems unlikely because of antitrust concerns and Intel’s massive market value. And a partial sale, particularly of its foundry division, doesn’t seem probable either, given the political, financial, and strategic challenges involved.

There’s a ton of talk about a sale – particularly in the past month. Qualcomm was reportedly rebuffed after offering to buy the whole firm. And Arm was sent packing after proposing to buy the products wing.

The thing is, a whole-firm takeover isn’t altogether likely. First, combining Intel with a bigger competitor would almost certainly draw intense scrutiny from antitrust regulators worldwide, because chips have become integral to the digital infrastructure that supports everyday life. Second, given Intel’s market value ($96 billion), a takeover would rank as the biggest acquisition of any tech firm in history. Qualcomm, meanwhile, has only $13 billion of cash on hand, so it would struggle to raise the necessary funds for a deal – even if it intended to sell some parts of Intel to other investors.

A partial acquisition might not be super likely either. Intel made it pretty clear to Softbank-backed Arm that its product business isn’t for sale.

And, sure, that does leave a question about whether Intel might be willing to sell its foundry division. But I think that’s unlikely too, at least right now. For starters, it would be politically difficult: any potential acquisition of Intel’s factory network would face huge scrutiny from US regulators because of the inherent national security risks involved. To gain approval, a new owner also would have to commit to spending the tens of billions of dollars that Intel has promised for new plants. Also, if it sold the foundry now, it would be doing so at a depressed valuation since that business is still extremely unprofitable. Finally, as previously discussed, there are Intel’s feelings about the whole thing: the firm is depending on the foundry unit to recoup at least a portion of the revenue that’s been lost in the products division. Knock that card out of place and the whole house could fall.

PART 4: VALUATION

What is Intel actually worth?

Even with conservative assumptions, a sum-of-the-parts valuation (SOTP) on the company points to a fair value that’s 60% higher than Intel’s current share price.

To answer this question, I built a model that values Intel using a SOTP approach – essentially, assessing its worth by valuing each business segment separately and adding them to determine the total value. You can follow along with my reasoning below, or you can download the model and put in your own assumptions. (Note that you’ll need to make a copy of the template before you can actually fiddle around with the cells).

First, the products division.

I assume sales from CCG will grow at a 3% CAGR through 2030, starting from a low base (Intel’s trailing 12-month revenues from this segment). One percentage point is due to volume, driven by rising PC shipments, and the other two percentage points are attributed to pricing – especially as AI PCs gain ground. My thinking here is that chips for these devices command higher prices.

The one percentage point due to volume assumes some continued market share loss to AMD because PC shipments are actually forecast to grow at a higher rate (researchers at IDC, for example, predict that shipments will grow at a 2.4% CAGR between now and 2028). This assumption could turn out to be conservative because Intel has managed to stabilize its market share over the past three and a half years (see the graph below).

For DCAI, I assume Intel’s market share loss to Nvidia, AMD, and the custom chips designed by big tech firms is permanent. In other words, my starting point for forecasting future revenues is Intel’s DCAI sales over the past 12 months, which totaled just $12.7 billion – 40% less than what the segment generated in 2021. From here, I forecast revenue will grow at a 2% CAGR through 2030, with further market share losses offset by Gaudi (Intel’s GPUs optimized for AI applications), which I expect to do well in specific segments of the AI data center market.

Finally, NEX, as previously mentioned, has remained relatively stable over time. This is a very small part of Intel’s overall revenues and I assume it’ll grow at a measly 1% CAGR through 2030.

Using those growth assumptions, Intel’s forecasted 2030 revenue from its products business comes in at $58.8 billion. That’s only slightly above what it made in 2022, highlighting my conservative assumptions. And sticking with that trend, I assume the business’s operating profit margin in 2030 will be 30% – much less than Intel’s official target of 37% to 40%. That gives $17.6 billion of forecasted operating profit, also known as earnings before interest and tax (EBIT).

We can now value Intel’s products business by applying a commonly used valuation multiple called EV/EBIT – or enterprise-value-to-EBIT. Enterprise value captures the value of a company’s (or a segment’s) equity and its “net debt” (total debt minus its cash holdings).

Now, I have to decide what EV/EBIT multiple to use when valuing Intel’s products business. And I figure I’ll take a common approach here and just see what multiples similar companies are trading at. Remember that Intel reorganized itself into two main businesses, each operating as an independent subsidiary. That means the products business should be compared to other fabless chipmakers that design and sell semiconductors while outsourcing manufacturing. The biggest include Nvidia, AMD, Broadcom, Qualcomm, Marvell Technology, and MediaTek, which trade at an average EV/EBIT of 23.4x based on forecasted EBIT over the next 12 months. (Note: multiples based on forecasted earnings are called “forward” multiples).

Applying a conservative 40% discount to this multiple to account for Intel’s slower growth results in an EV/EBIT multiple of 14x. And when applying that to my 2030 EBIT forecast, I arrive at an enterprise value of $247 billion at the end of 2029 (given that I’m using a forward multiple).

Next, the foundry business.

Here I assume Intel will bring in $10 billion of revenue from external customers – much less than the firm’s goal of $15 billion. This is partly conservatism, and partly because Intel has spearheaded deals with giant pools of money (Apollo Global Management and Brookfield Asset Management) to take equity stakes in some factory projects. Because of those agreements, Intel has to share some of the profit with its partners. So it’s only fair that I value the foundry business based on the revenue (and, in turn, the profit) that Intel gets to keep to itself.

What I really want to know here is whether the business can hit that $10 billion in revenue in 2030. Recall that it made less than $1 billion in sales from external customers last year. But since then, it has signed multi-billion dollar deals with Microsoft and Amazon. Based on that, I think the foundry business has a good shot at hitting at least $10 billion in revenue by 2030, assuming nothing goes wrong during its capacity expansion and its installation of advanced manufacturing equipment. I’m not totally alone here: several research firms expect the overall foundry market to be worth more than $200 billion in 2030. So Intel capturing a 5% share of that shouldn’t be a big ask.

I assume the business’s operating profit margin in 2030 will be 20% – considerably less than Intel’s official target of 25% to 30%. This is partly my own cautiousness and partly the effects of operating leverage. See, given the high level of fixed costs in the foundry business, it’s only reasonable to assume a smaller operating profit margin in light of my lower revenue projection.

Multiplying the two gives me $2 billion of forecasted EBIT in 2030. TSMC – the world’s biggest foundry – trades at a forward EV/EBIT of 16.4x. Applying a premium or discount to this multiple is a tricky call. On one hand, because Intel has a much smaller market share, it arguably has the potential for more growth. And because it has a geographically diverse factory network, it has lower geopolitical risk compared to TSMC, whose manufacturing is clustered in Taiwan. But on the other hand, our assumed 20% operating profit margin pales in comparison to TSMC’s 40%, meaning Intel is less cost-efficient and its earnings could be more volatile.

So, for simplicity, I assume the factors cancel out and that Intel’s foundry should trade at the same forward EV/EBIT multiple as TSMC. That gives me an enterprise value of $32.8 billion at the end of 2029.

Finally, total firm value.

I add the two enterprise values and get a total of $280 billion at the end of 2029. To arrive at equity value, I need to subtract forecasted net debt at that point in time. Intel currently has $18 billion in net debt, and I know it’s eligible for $11 billion in super-cheap US government loans under the CHIPS Act, which it will likely use.

With that, I don’t see the firm needing to issue more debt to fund its big spending through 2030, which I estimate will total $105 billion (with the vast majority of that going into building out its foundry network in the US). After all, it’s got much of that covered through $8.5 billion worth of government grants (also under the CHIPS Act), a 25% US tax credit on its investments (worth 25% x $105 billion = $26.3 billion), and five years of $11.5 billion in estimated annual cash flow from operations (CFO). And that $11.5 billion annual CFO estimate is modest because it’s based on Intel’s CFO in 2023, which was the lowest since 2009, and explicitly assumes no growth in cash flow from 2024 to 2029.

So, subtracting $29 billion of net debt ($18 billion on the balance sheet plus $11 billion in future debt) from our total enterprise value of $280 billion gives me an equity value of $251 billion at the end of 2029. Discounting this at a 13.6% cost of equity (derived from a popular technique called the “capital asset pricing model”) gives me an equity value of $132.7 billion at the end of 2024.

But there’s one last set of adjustments I need to make. Intel fully owns programmable chipmaker Altera and has an 88.3% stake in self-driving technology company Mobileye. Intel reports both of these firms’ revenues and profits separately to the products and foundry businesses, so they’re not included in our earlier revenue forecasts and, in turn, our valuation. Now, the value of Intel’s stake in Mobileye is easy to calculate since the firm is publicly listed. When it comes to Altera, though, for the sake of simplicity, I’ll just assume it’s worth what Intel paid for it ($16.7 billion) back in 2015.

Adding all the figures gives me a total equity value for Intel today of $159 billion, or $37 per share (60% above where it’s currently trading). For reference, the average price target from investment bank analysts is $25 – much lower than my $37 estimate. That’s mainly because most analysts apply a valuation multiple to their forecasted earnings over the next 12 months, which, in Intel’s case, are quite depressed. Consider this: before the firm’s discouraging second-quarter update, where it significantly lowered its near-term earnings forecast, analysts’ average price target was around $38. And generally, I’d argue that valuing a company in the midst of a turnaround based on its earnings over the next 12 months isn’t the savviest approach.

CONCLUSION

So what’s the sum and substance here?

To recap: Intel’s shares are lingering near a ten-year low, and the firm is trading below its book value for the first time since at least 1990. And that made me suspect that its stock is cheap – something I confirmed with a SOTP valuation. I found that even with really conservative assumptions, the model spits out a fair value that’s 60% higher than Intel’s current stock price.

Of course, these shares won’t be cheap at all if the firm’s current problems turn out to be existential ones. Fortunately, I don’t believe they are: Intel’s taking the smart, bold actions to turn things around. What’s more, given the sensitivity toward domestic US semiconductor technology and production, it’s unlikely that the American government would ever allow Intel’s troubles to become terminal.

Having said that, Intel’s ambitious transformation plan to restore profitability and significantly expand its foundry network makes this a "show me" story fraught with execution risks. In other words, if you decide to invest in the company’s shares (after conducting your own research of course), you may want to start with a small position and gradually add to it as Intel hits certain milestones in the coming years. Here are a few to watch for:

• Stabilizing or reversing revenue declines in the CCG and DCAI segments
• Reporting strong sales performance for its Gaudi 3 GPU
• Improving operating profit margins
• Securing more high-profile clients for its foundry
• Completing its major manufacturing projects on time and on budget
• Demonstrating the potential of high NA EUV technology on a commercial scale
• Managing cash flows and the balance sheet prudently to avoid being forced to issue equity at a depressed valuation

Intel may indeed be undervalued, but it will likely still take a few successful years before investors reward it with a higher valuation multiple. So before you consider buying into its shares, make sure you’ve got faith in its turnaround strategy and are willing to take the long view.