Written by: Rita

Trend Guide

Morgan Stanley released an Asia-Pacific storage update on July 6, providing clear judgments on the three most controversial storage issues: the largest spender in AI is rumored to have excess computing power for sale, raising market concerns that the entire AI infrastructure may be oversupplied; after the announcement of long-term supply agreements (LTA), stock prices have been slow to rise; whether the storage cycle has reached its peak.

Morgan Stanley's conclusion is clear: storage is approaching peak rates of change in three dimensions—year-on-year prices, inventory, and profit adjustments. Short-term stock prices may face pressure, but this AI-driven storage bull market is far from over. Profits in the storage industry are still expected to grow by 35% to 40% in 2027, and the rise of AI entities will continue to drive demand. Morgan Stanley suggests seeking opportunities in DRAM and traditional storage while avoiding module manufacturers.

Debate on "Excess Computing Power": Over-supply or Positive Feedback from AI Construction?

Last week's hottest topic in the market was the rumor that a certain large-scale manufacturer had excess computing power available for sale. Morgan Stanley believes that the market is behaving as it usually does: selling off before the news is released. The pessimistic interpretation is straightforward: if top cloud manufacturers have idle computing power, then the entire AI construction is oversupplied.

However, Morgan Stanley offered another interpretation: this precisely indicates that enterprises are smartly monetizing infrastructure and improving return on capital expenditure, which is a different issue from “over-construction.” The moment that truly decides the direction will be during the second quarter earnings report season, whether large-scale manufacturers maintain or increase capital expenditures. If they maintain, storage is a buying opportunity; if they reduce, the narrative of oversupply will continue to ferment.

Another discussion centered on the token economy. Morgan Stanley observed that enterprises are shifting from “encouraging employees to generate as many tokens as possible” to “seeking cheaper alternatives,” with open-source models rapidly emerging in China. Companies are beginning to layer orchestration on top of cutting-edge models, using open-source for simple queries and cutting-edge for complex queries. This trend of "token minimization" is changing the market's focus; investors are no longer just watching revenue growth but are more concerned about what guidance will indicate.

Morgan Stanley's judgment is that while second-quarter AI supply chain performance is fine, the market has shifted to worry about guidance for the second half of the year, which will have significant implications for storage. Storage is the most direct beneficiary of AI capital expenditure.

Why Did Stock Prices Not Rise After the Announcement of Long-Term Supply Agreements?

Storage companies have consecutively announced long-term supply agreements, but stock prices have not experienced the expected revaluation in the market. Morgan Stanley's explanation is quite straightforward: the market is rational. Investors have learned from the lessons of long-term supply agreements for simulated chips during the pandemic that ultimately became inventory burdens. Those agreements were either renegotiated or customers were forced to accept unwanted inventory.

Morgan Stanley believes that this time, the long-term supply agreements are structurally stronger, and as long as AI remains strong, the logic stands. However, the market's skepticism is reasonable; they need to see actual execution, not just a piece of paper.

Morgan Stanley also acknowledges that the biggest obstacle is the uncertainty of the timeline—how much longer can storage prices rise, by how much, and where will earnings per share reach by 2028. The market's resolution to these questions dictates the upper limit of stock prices. Among AI beneficiaries, the extent of earning revisions for storage stocks is far ahead, but this fact is already well understood by the market.

Cycle Nearing Peak, but Far from Ending

Morgan Stanley repeatedly emphasizes one judgment in the report: storage is still a cyclical industry, but this time it is different, and the rate of price change is approaching its peak.

The report notes that since the emergence of generative AI in November 2022, the storage industry has undergone three cyclical corrections. Each correction is a necessary adjustment within a structural bull market, and the beginning of a new bear market is not valid. The severe corrections caused by crowded positions are actually healthy, clearing space for the next upswing.

Morgan Stanley uses a straightforward analogy: price momentum is beginning to fade, but that does not mean the cycle is over. The recent underperformance of large-scale manufacturers (the core drivers of AI spending) may be a leading indicator that storage is about to enter a period of relatively weak market performance. At the same time, the breadth of earning revisions for storage stocks is approaching historical extremes, which is usually not a good signal.

Morgan Stanley's conclusion is that positions need a correction. The decline in AI beneficiary stocks is not due to high valuations, but because positions are too crowded. These stocks have already risen significantly, and earning revisions confirm these increases, but the breadth of corrections has reached extremes. The market needs a breather, and the fluctuations of earnings season may provide this opportunity.

Morgan Stanley remains bullish long-term, projecting a profit growth of 35% to 40% in 2027, with the rise of AI entities as the next structural driving force. However, in the short term, the possibility of stock prices being under pressure before earnings season is increasing.

Samsung and SK Hynix: Q2 Expectations Overview

Morgan Stanley provided the Q2 outlook for two leading South Korean storage companies:

Samsung Electronics is set to release preliminary results on July 7, with Morgan Stanley expecting operating profits of around 85 trillion won, in line with market expectations. SK Hynix will report its earnings on July 29, with Morgan Stanley expecting operating profits of around 65 trillion won, also aligning with market expectations. The management guidance of both companies is anticipated to align with market expectations, with strong commodity storage continuing in Q3, multiple long-term supply agreements being fulfilled, and capital expenditure only moderately increased.

Trend Perspective

The most valuable point of Morgan Stanley's report lies not in the conclusion itself, but in its acknowledgment of an easily overlooked fact: the breadth of profit revisions for storage stocks is far ahead among AI beneficiaries, but this information is already fully priced into the market, and stock prices may have reflected too much good news.

The title of the report is “Changing Tides.” This metaphor is quite accurate: the tide is receding, but it is not low tide; it has just changed direction. Morgan Stanley emphasizes that while short-term stock prices may be under pressure, the long-term judgment of 35% to 40% profit growth in 2027 remains unchanged. In other words, even if there is a short-term pullback, it is merely a routine correction within a structural bull market, not the end of the cycle.

For investors focused on storage, Morgan Stanley provides a clear framework: crowded positions are a short-term issue, while profit growth is the long-term answer. The key will be to see what large-scale manufacturers say during the second-quarter earnings report season, while the statements from storage companies themselves are less important.

Disclaimer

This article represents the整理与解读 of third-party brokerage research reports by Trend Research. The ratings, target prices, profit forecasts, and related judgments quoted in this article are the views of the brokerage analysts and only represent the positions of their respective organizations, and do not represent the views of Trend Research, nor do they constitute any investment advice.

The market carries risks, and decisions should be made independently. This article should not serve as the basis for buying or selling any securities.