Morgan Stanley aggressively raises Micron’s target price to $450: As long as AI demand remains strong, China’s capacity impact and overheated capital expenditure are not problems at all!

```

Morgan Stanley believes that the market has greatly underestimated the current shortage of memory chips. Driven by the AI super cycle, the traditional cyclical valuation framework is no longer applicable, and Micron is at the sweet spot where both profitability and valuation multiples are expanding.

According to Chase Wind Trading Desk, on February 11, Morgan Stanley's analyst team aggressively raised Micron’s target price from $350 directly to $450 in their research report, maintaining an “overweight” rating. The core logic behind this move is simple and straightforward—so long as AI demand remains strong, the previously feared risks such as HBM4 mass production issues, the impact of Chinese memory chip capacity, and overheated capital expenditure are all inconsequential noise.

For investors, Morgan Stanley’s report reveals a reality overlooked by the market: the shortage of memory chips has spread to every end market, and pricing power is entirely in the hands of sellers. Morgan Stanley forecasts that Micron’s earnings per share (EPS) will surge past $52 in calendar year 2026 (CY26). Driven by the AI super cycle, the traditional cyclical valuation framework is no longer applicable, and Micron is at the sweet spot where both profitability and valuation multiples are expanding.

Widespread shortage ignites pricing power, sharp upward revision in earnings forecast

The market has clearly underestimated the current tightness in the memory chip market.

According to Morgan Stanley's research, prices for both DRAM and NAND continue to rise in the first and second quarters of 2026. Although Micron’s official guidance for the second fiscal quarter suggests a 37% quarter-over-quarter revenue growth, underlying this is about a 30% quarter-over-quarter increase in average selling price (ASP).

This is only a conservative baseline because competitor data is even more astonishing: SanDisk’s guidance shows a 60% quarter-over-quarter surge in NAND ASP, and teams covering Samsung and SK Hynix model Q1 regular DRAM prices rising by 48% and 55%, respectively.

This price surge directly translates into astonishing profitability. Morgan Stanley clearly points out that Micron’s profitability is reaching new heights, with 2026 calendar year EPS expected to hit $52.53.

Even if Micron chooses not to update specific numeric guidance in its financial report, as long as it confirms that market conditions are better than expected, any stock price pullback due to a lack of specific guidance should be seen as a great buying opportunity. The current consensus expectation is that Micron’s profits will peak at around $12 in late 2027; this forecast is considered extremely conservative by Morgan Stanley and, in reality, Micron’s profit levels may consistently exceed consensus models over the next year and a half.

Valuation logic reshaped: AI premium and cross-cycle profitability

Investors’ biggest divergence right now is about valuation multiples.

Morgan Stanley believes it is wrong to assess Micron based on traditional cyclical thinking. Currently, the share price corresponds to an 8x P/E based on a $48 earnings forecast, while the gross margin is already 10–15 percentage points higher than the peak of the previous cycle. If categorized as a cyclical stock, this valuation is not only reasonable—it might even be considered extremely cheap, since it is closer to 5x peak profitability, instead of the 10x at the 2021 cycle peak.

Morgan Stanley has readjusted its valuation model, raising the estimate for “cross-cycle EPS” from $14 to $18. This figure is significantly higher than historical averages, but still less than half of expected earnings over the next 12 months.

Based on the unique tailwind effect brought by AI and the opportunity of HBM (High Bandwidth Memory), Morgan Stanley maintains a 25x P/E multiple for Micron. 18 dollars cross-cycle EPS times 25, leads directly to the new $450 target price. In addition, Micron’s cash generation capacity of about $10 billion per quarter means it can generate cash equivalent to 10% of its current enterprise value (EV) in a year, which will significantly improve its balance sheet.

Unbridgeable supply-demand gap: $200 billion in new demand

The cornerstone driving this super cycle is the structural supply-demand imbalance brought by AI. At present, inventories on manufacturers’ balance sheets are extremely low, and even customers willing to pay premiums cannot secure enough supply. On the supply side, wafer fab output growth is extremely slow. Morgan Stanley expects wafer input volume, including CXMT (Changxin Memory Technologies in Hefei), SK Hynix M15, and Samsung P4L, to grow only 7% year-on-year by the end of 2026.

By contrast, the growth on the demand side is explosive. NVIDIA expects to add $30 billion in quarterly revenue by the end of 2026; AMD’s data center division is expected to double quarterly revenue to $10 billion; Broadcom’s semiconductor business is also expected to double to about $25 billion. Including incremental contributions from Marvell and Intel, the entire memory industry faces nearly $200 billion in annualized new revenue demand over the next 12 months.

This figure is larger than the entire logic semiconductor market (such as CPU and GPU) in 2020, and HBM’s high capital intensity further restricts the pace at which DRAM suppliers can catch up. Therefore, selling stocks out of fear of supply growth in the second half of 2027 would be premature.

Crushing “short seller” narratives: Chinese capacity and HBM4 risk disproved

Regarding market concerns about the impact of Chinese memory chip companies, Morgan Stanley believes these worries are exaggerated.

As for HBM4 progress, there has been noise in the market recently, but Morgan Stanley stands by its view. Micron had already stated in its December financial report that it had obtained qualification and expected to achieve mass production and shipment of HBM4 in the second quarter of 2026 (C2Q26), and this timeline has not changed. Even if Micron encounters unforeseen difficulties ramping up HBM4 capacity, HBM3e remains mainstream, with a broad ASIC customer base, and will not negatively impact profitability. Early major production share of NVIDIA’s HBM4 has always belonged to SK Hynix, and this is within expectations—not a fundamental negative for Micron.

Risk Disclosure and DisclaimerThe market carries risks and investments should be made prudently. This article does not constitute individual investment advice nor does it take into account the particular investment objectives, financial situation, or needs of any individual user. Users should consider whether any opinions, views, or conclusions in this article are appropriate for their individual circumstances. Investment based on this article is done at your own risk. ```