Without a doubt, semiconductor stocks have been among the biggest winners of the artificial intelligence (AI) revolution. While stars like Nvidia, Semiconductor manufacturing in TaiwanAnd Broadcom While investing in the broader chip sector has received the most attention over the past two years, it has generated market-beating returns.
At the close of the market on December 20, the VanEck Semiconductor ETF had gained 39% in 2024, far exceeding the returns of both S&P500 (SNPINDEX: ^GSPC) and Nasdaq Composite (NASDAQINDEX: ^IXIC).
Still, not all semiconductor stocks performed as well. Take Micron technology (NASDAQ:MU), for example – with shares up 6% in 2024, investors might think this particular chip stock is a bust.
Savvy investors know that looking at a stock’s performance is just one variable when evaluating an opportunity. Below, I’ll examine what influenced Micron’s price action throughout the year and explain why 2025 could be a rebound year for the company.
The chart below illustrates how Micron stock will perform throughout 2024. The highs and lows depicted in the chart make one thing very clear: Micron is quite volatile. In particular, the last six months have been unusually difficult, with shares falling about 38% since June.
My view on what causes Micron stock to experience so much volatility comes down to one thing: expectations. When companies like Nvidia, Taiwan Semiconductor, Broadcom and many others show robust and consistent growth, investors tend to apply these trends to other companies in the same industry.
While I understand the psychological factors behind these parallels, it is imperative that investors understand that such a notion is based on flawed logic. Not all chip companies make the same products or serve the same purpose, and for this reason, each company is going to experience its own set of unique headwinds and catalysts.
Again, I don’t necessarily see this as a reason to sell the stock. Below, I’ll explain why Micron’s latest plunge is unwarranted.
Since AI became the next global megatrend about two years ago, one product in particular has become the holy grail of the tech industry: graphics processing units (GPUs).
Companies like Nvidia and Advanced microdevices developing chipsets called GPUs capable of running complex algorithms at extremely high speeds, and it is this hardware that powers myriad generative AI applications. Going further, Taiwan Semiconductor makes GPUs for Nvidia and AMD while Broadcom supplies a host of network infrastructure equipment needed for power. data centers where these GPUs are hosted.
Given all of this, doesn’t it seem natural that these specific businesses have seen unusually high growth over the past couple of years?
In my eyes, Micron just hasn’t had its moment yet, but I think it’s coming. Given that investments in AI infrastructure are expected to be in the billions of dollars over the next few years, I think it’s safe to say that the demand for GPUs and data center services is not going to to slow down.
At a more granular level, this means that training and inference workloads will become more sophisticated and critical as competition in the AI arms race intensifies. It is this dynamic that is expected to drive an increased need for memory and storage protocols – and Micron is extremely well positioned to meet this demand.
Nor is it just a lofty theory. According to Micron’s first fiscal quarter 2025 (ended November 28), the company’s data center revenue increased 400% year over year and reached an all-time high. Additionally, the company’s data center segment now accounts for more than 50% of the business. To me, these trends underscore the need for Micron’s memory chips and I expect the tailwinds to continue into next year and beyond.
While the company’s short-term outlook may not have lived up to expectations, I think the long-term narrative is still very much there. According to management, the total addressable market for high-bandwidth memory is expected to reach $100 billion by 2030, more than six times what it is today. Given that Micron’s trailing 12 month revenue is in the range of $29 billion, I think the company has significant upside potential.
Image source: Getty Images.
Valuing Micron is a difficult task. Even though the company generates positive earnings, it has only recently become a profitable business and so using the price-to-earnings (P/E) ratio seems a bit out of place, in my opinion.
Instead, I will use the PEG ratio to evaluate an investment in Micron. The PEG ratio looks at analysts’ forecasts for earnings growth over several years. If the PEG ratio is less than 1, the stock could be considered undervalued. Right now, Micron’s PEG ratio is just 0.23.
In my opinion, Micron’s low PEG suggests that investors may be overlooking the need for memory and storage chips as AI workloads become larger and more complex. However, over time, I believe the need for Micron’s services will become more and more evident. To me, buying Micron now is a good opportunity for investors with a long-term time horizon.
Have you ever felt like you missed the boat by buying the best performing stocks? Then you will want to hear this.
On rare occasions, our team of expert analysts issues a “Doubled” actions recommendation for businesses that they believe are on the verge of collapse. If you’re worried that you’ve already missed your chance to invest, now is the best time to buy before it’s too late. And the numbers speak for themselves:
Nvidia: If you invested $1,000 when we doubled down in 2009, you would have $362,166!*
Apple: If you invested $1,000 when we doubled down in 2008, you would have $48,344!*
Netflix: If you invested $1,000 when we doubled down in 2004, you would have $491,537!*
Right now, we’re issuing “Double Down” alerts for three incredible companies, and there may not be another chance like this anytime soon.
Adam Spatacco has positions at Nvidia. The Motley Fool holds positions and recommends Advanced Micro Devices, Nvidia and Taiwan Semiconductor Manufacturing. The Motley Fool recommends Broadcom. The Mad Motley has a disclosure policy.
6 mins read
