If we look at Micron's valuation purely from a 12M-forward perspective, this puts the EPS at $114/share by 1st July 2027, and against today's price, we are talking about a forward P/E of 9.12x.
FactSet is expecting EPS growth as follows:
FY26 Expected EPS: $73.12E, 782% YoY growth.
FY27 Expected EPS: $154.66E, 112% YoY growth.
FY28 Expected EPS: $166.34E, 8% YoY growth.
This is a meaningful increase from the EPS expected when I wrote my last article in May, driven by stronger Q3 earnings and the more optimistic guidance for Q4. Previously the analysts were also anticipating an EPS decline in FY28, pointing to the end of the cycle, but I already explained earlier why this is unlikely to materialize.
Micron's historical valuation was anywhere between 3 and 50 P/E, underscoring the historical cyclical nature of the business. During periods without excess demand or excess capacity, the valuation is more 14-20x P/E, giving us a sense that the P/E we are paying today isn't extreme at all.
This brings me to my price target and the title of this article. If we assume the $114 EPS in 12M time and apply an average valuation of 15x P/E, which is actually a contraction from today's 16.6x, we would land at $1,725/share by July 2027. Naturally, this is just a price target, and the earnings forecast can move higher or lower, depending on how the overall Hyperscalers CapEx story develops, but I see this as a base case, with 15x P/E as a fair value for Micron.
A bull case would be an increase to P/E of 17-20x, which would put the price at $1,955 - $2,300.
I don't expect any meaningful extra production capacity deployment before 2028, so while the competition is fierce from SK hynix and Samsung, the combination of strong demand and lack of supply will hold the prices of memory chips high for longer than I previously expected.
Investor's Takeaway
The evidence is pointing to a reversal in the memory-chip industry behavior, and what used to be a commodity may be less volatile in the future, supported by long-term contracts, which set a floor on price and revenue for all the peers.
The super cycle is playing into Micron's cards, giving the company an edge to dictate its own terms and negotiate favorable conditions, as memory chip prices remain very high, driven by a lack of capacity. Yes, new capacity will be deployed, but unlikely ahead of 2028, giving a lot of breathing room to the industry. Still, I don't think we've moved away completely from the commodity-like cycles, but it will be delayed, perhaps into 2029-2030.
Micron is in a strong position right now to capitalize on the high prices and limited availability, agreeing to long-term lucrative contracts, and its EPS will grow massively in the next couple of years. I am estimating $115 EPS in 12M time, and if we apply 15x P/E (which is my fair value), Micron's stock price could reach $1,725 by July 2027. If the EPS growth surprises to the upside, it could be well above $2,000.
Disclaimer: Investing carries risk. This is not financial advice. The above content should not be regarded as an offer, recommendation, or solicitation on acquiring or disposing of any financial products, any associated discussions, comments, or posts by author or other users should not be considered as such either. It is solely for general information purpose only, which does not consider your own investment objectives, financial situations or needs. TTM assumes no responsibility or warranty for the accuracy and completeness of the information, investors should do their own research and may seek professional advice before investing.
Despite a blockbuster debut for China’s ChangXin Memory Technologies, which saw its stock pop 466% in Shanghai on its first trading day, U.S.-listed memory stocks are taking a sharp turn lower.
Shares of Micron (MU) closed 2.3% lower on Monday, after falling as much as 7% intraday, while those of Sandisk tumbled 11%. SK Hynix’s American depositary receipts slid 7.5%.
CXMT became China’s most valuable publicly listed company after the strong reception to its initial public offering, according to LSEG data, which showed a market capitalization of about $484 billion. The Hefei-based company makes dynamic random-access memory chips and is the fourth-largest producer of those chips behind SK Hynix, Samsung Electronics and Micron. Without access to ASML’s extreme-ultraviolet lithography technology, however, CXMT is limited from making the advanced high-bandwidth memory chips that have proved lucrative for its competitors.
That said, the Information reported on Monday that China has begun making its own deep-ultraviolet lithography machines, potentially hinting that domestic chip makers in the country could become more competitive. U.S.-listed shares of ASML, which makes both DUV and EUV lithography systems, fell nearly 6% on Monday.
The focus on silicon-intensive HBM for artificial intelligence has contributed to the broader DRAM shortage, and CXMT has been able to capitalize on that industry dynamic, according to Richard Windsor, founder of research firm Radio Free Mobile.
The higher margins from HBM have led the memory leaders to switch capacity toward AI data-center chips from commodity DRAM that is used in consumer electronics, Windsor said, and that “is the gap that CXMT is filling.”
Although Windsor expects “a big correction” for CXMT when DRAM supplies normalize, “between then and now, it looks like good times are ahead,” he said.
Meanwhile, Nvidia said Friday that it and South Korea’s SK Group had signed letters of intent to expand their strategic collaboration with a more than $500 billion initiative to build AI factories with SK Telecom and to co-develop future generations of memory chips for AI, including HBM, with SK Hynix.
Custom chip maker Broadcom also announced a memorandum of understanding with Samsung on Saturday to expand their strategic collaboration for advanced memory chips and foundry services, including advanced packaging for AI and networking chips. The partnership is expected to be worth more than $200 billion through 2030.