Morgan Stanley Raises Q3 2026 Memory Price Forecasts, Warns of Near-Term Momentum Peak

• Morgan Stanley raised its price forecasts for multiple memory types in Q3 2026, while cautioning that short-term momentum for memory stocks may weaken.
• Under this reporting framework, the price increase expectation for PC DRAM has risen to 15–20%, and the breadth of DRAM profit revisions approaches 89%.
• Samsung and SK Hynix remain supported by AI demand, but earnings guidance, LTA terms, and capital expenditure statements will influence whether the rally continues.

Morgan Stanley significantly raised its third-quarter memory price forecast in a research report on July 7, but also cautioned that memory stocks may face short-term downward pressure.

This is not a memory cycle reset. The report maintains its "attractive" view on Korea's tech sector, continuing to favor Samsung Electronics and SK Hynix, with its model projecting over 35-40% earnings growth for related companies by 2027. The real caution is that memory prices, earnings expectations, and investor positioning are already at elevated levels, meaning stock prices may not continue rising at the same pace as in recent months.

The most direct figure is the price forecast. The report raised its expected quarter-over-quarter ASP increase for 3Q26 PC DRAM from the previous range of 3-8% to 15-20%, server DRAM to 13-18%, GDDR6 and GDDR7 to 15-20%, and enterprise SSDs to 18-23%.

The directional outlook from public pricing agencies is also intensifying. According to a TrendForce article on July 3, the DRAM market in Q3 2026 will remain extremely tight, with contract prices expected to rise 13%-18% quarter-over-quarter, and NAND Flash contract prices forecast to increase 10%-15%. However, TrendForce also noted that while server DRAM continues to face supply shortages, long-term supply agreements will moderate the pace of price increases.

Prices are still rising, but trading memory stocks is becoming more challenging. According to Morgan Stanley’s metrics, the breadth of DRAM earnings revisions has recently approached 89%, nearing historical highs. The memory market rally over the past two years, driven by AI capital expenditures, HBM demand, and server demand, has already priced in many of these positive factors.

This price adjustment covers a broad range. Price expectations for server DRAM, graphics DRAM, standard DRAM, and enterprise SSDs in Q3 2026 have all been significantly raised. Among these, the 15-20% increases for PC DRAM and graphics DRAM are the most noticeable price signals for the market. The 18-23% rise in enterprise SSD prices indicates that storage price increases are not limited solely to AI server-related products.

The issue is here as well. The faster the price expectation is raised, the more likely the stock is to enter the "good news priced in" phase.

Memory stocks have not gone without corrections over the past two years. According to Morgan Stanley, since the generative AI wave began in November 2022, DRAM-related stocks have experienced three significant pullbacks, including profit-taking, specific event-driven shocks, and prolonged adjustments lasting several weeks. Each correction did not disrupt the long-term rally driven by AI capital expenditures, but each served as a reminder that even in strong cycles, deep阶段性 declines can occur.

DRAM stock experienced three major pullbacks of approximately -25%, -25%, and -35%, but overall still rose from its 2022 low to a new high in 2026.

Spot prices also support the price increase rationale. The report’s price tracking shows that DRAM spot prices have risen sharply since early 2025, and NAND spot prices have recently rebounded clearly from their lows. Although contract prices lag behind, they are also moving upward. In other words, the short-term pullback signal is not due to deteriorating prices, but rather because prices and expectations have risen too quickly.

DDR5 16GB spot prices have risen to around $47, and MLC 64Gb NAND spot prices have recently increased to $31.10.

For stocks, what matters more than the price itself is how much upward potential remains.

The breadth of DRAM earnings revisions has recently reached approximately 89%, nearing historical highs. This metric indicates that an increasing number of analysts are raising their earnings forecasts. When most analysts have already raised their estimates, the difficulty of further exceeding expectations increases.

DRAM profit revision breadth rose to approximately 89% after 2025, nearing historical peak levels.

This is also one reason why short-term momentum may weaken. Memory stocks are not lacking fundamental support; rather, price increases, upward earnings revisions, capital inflows, and AI-related sentiment have already piled up. Once earnings guidance is weaker than expected, capital expenditure comments are less optimistic, or major cloud providers face downward pressure on their stock prices, the memory sector is more likely to experience amplified volatility.

Morgan Stanley still favors DRAM and traditional memory, where the flow of funds is clearer and supply bottlenecks are more evident, viewing them as more attractive than NAND, and is relatively least optimistic about memory module manufacturers. This ranking indicates that the market is not simply betting on "all storage prices rising," but rather assessing whether price increases can truly translate into profits.

The expected price increase for PC DRAM in Q3 2026 has been raised from 3-8% to 15-20%, and enterprise SSD prices have been raised to 18-23%, with simultaneous upward adjustments across multiple DRAM categories.

The earnings reports from South Korea's two leading memory chip manufacturers will serve as a key indicator for validating the price increase rationale.

Samsung Electronics released its Q2 2026 earnings guidance on July 7. The company expects sales of approximately KRW 171 trillion and an operating profit of approximately KRW 89.4 trillion, with an operating profit range of KRW 89.3 trillion to KRW 89.5 trillion. For Samsung, the market does not focus solely on quarterly profits but also on the strength of the memory business recovery, progress in AI-related products, and whether the upward trend in traditional storage prices can continue to improve profitability.

SK Hynix’s earnings date is still pending. Public market calendars indicate the company is expected to release its next earnings report on July 29. Given SK Hynix’s more prominent position in HBM and AI server memory, the market will be more sensitive to its statements on Q3 commodity memory prices, long-term supply agreements (LTAs), and capital expenditures.

If management confirms that commodity DRAM demand remains strong in Q3, LTA commitments increase, and capital expenditures are only modestly raised, the short-term pullback may resemble a healthy correction. However, if guidance is weak or capital expenditures are interpreted as signaling too-rapid supply growth, the market may reassess how long DRAM price increases can be sustained.

LTA itself is not a risk-free signal. Historically, long-term agreements have not necessarily led to stock price increases; some agreements have been renegotiated or turned into binding obligations for customers when demand shifted. The market does not look solely at "how much was signed," but also considers pricing, term lengths, customer quality, and execution flexibility.

The long-term bullish thesis still stems from AI, particularly AI agents capable of performing tasks, invoking tools, and engaging in continuous interaction. Morgan Stanley expects in its report that related companies’ earnings growth could still exceed 35–40% by 2027, which is why it does not equate short-term pullbacks with the end of the cycle.

However, the debate around AI demand is shifting. Previously, the market largely bet on continued expansion in model training and inference scale, driving upward pressure on compute and memory demand. Now, some investors are concerned that after the third quarter, cloud providers may place greater emphasis on token savings, inference efficiency, open-source low-cost models, and the pressure on margins from chip inflation.

Another more sensitive issue: Do the largest AI spenders actually have sellable computing power, suggesting that early-stage infrastructure investments may have been temporarily excessive? This claim has not yet been conclusively proven, but it is enough to make the market more cautious ahead of major cloud providers’ earnings reports.

The current divergence in the memory market is not about whether AI demand will disappear immediately, but rather whether price increases, upward profit revisions, and customer capital expenditures can continue to reinforce each other. If AI supply chain earnings in Q2 remain favorable but guidance for Q3 begins to weaken, memory stocks may first undergo a round of valuation and positioning adjustments.

The memory cycle still unfolds against the backdrop of massive AI capital expenditures, but short-term trading has shifted from "how much prices have risen" to "how long the market will continue to believe in these price increases." Samsung and SK Hynix’s upcoming earnings guidance, cloud providers’ statements on capital spending, and the execution of long-term agreements will determine whether this correction is merely a pause in a bull market or the true beginning of a slowdown in the pace of gains.