
Goldman Sachs Research Analysis: Samsung Target Price 490,000 KRW, Memory Supply-Demand Remains Tight
Author: Rita
Market concerns suggest the memory cycle may have peaked, but Samsung management estimates that supply and demand conditions in 2027 will be tighter than in 2026. Following an online meeting with Samsung Electronics management, Goldman Sachs reiterated its Buy rating in a report published on September 18, 2026, and added the stock to its Conviction List. The price target for common shares stands at KRW 490,000, representing a 94.1% upside from the current price of KRW 252,500. The price target for preferred shares is KRW 360,000, offering an 86.2% upside.
In the report, Goldman Sachs analyst Giu Lee noted that Samsung management reaffirmed that memory supply-demand conditions will remain tight. Key supporting factors include robust demand momentum, low fill rates, and deferred demand extending from 2026 into 2027. Incremental capacity will be concentrated in DRAM, particularly high-bandwidth memory (HBM), while capacity for other memory products faces squeeze. Supply expansion for traditional DRAM faces structural constraints.
Memory Tightness to Persist Until 2027
Unsatisfied demand from 2026 will be pushed into 2027, making supply-demand conditions there tighter than in 2026. Long-term supply agreements provide visibility beyond 2027, and management anticipates favorable supply-demand dynamics will extend past 2027. Samsung’s P4 fab will primarily produce 1c nm DRAM for HBM4 and HBM4E, while the P5 fab is expected to begin its first wafers around 2028, also focusing mainly on DRAM. For NAND, due to limited space at South Korean fabs, the company will focus on technology node migration at its Xi’an facility.
Goldman Sachs pointed out that bit output growth for both DRAM and NAND faces challenges. Expanding traditional DRAM capacity is difficult as new capacity is primarily allocated to HBM. The trade ratio for HBM rises with each generation, currently standing at 3:1 to 4:1, and increasing chip sizes further exacerbate capacity constraints. This structural shift indicates stronger supply discipline in the memory sector compared to previous cycles, suggesting prices and margins could remain elevated for longer.
Long-Term Agreements Lock in Demand
Current long-term supply agreements differ from previous cycles in duration, coverage, and enforceability, providing the company with financial stability and disciplined capital expenditure planning. Customers pay substantial upfront payments and deposits proportional to scheduled purchase volumes, incentivizing more accurate forecasting.
The remaining 30% to 40% of capacity, unconstrained by long-term agreements, will serve mobile, PC, and non-AI customers, areas that may see upside potential from edge AI and terminal AI. Goldman Sachs believes the proliferation of long-term agreements has shifted the memory industry's business model from spot price volatility toward more predictable contracted revenue. This change benefits leading producers like Samsung, enabling more stable capacity and investment planning.
HBM Squeezes Traditional DRAM Capacity
Foundry capacity at the 4nm node is very tight due to demand for HBM4 base dies, prompting the company to consider expanding 4nm capacity in South Korea. Meanwhile, a new 2nm production line for HBM5 base dies is under evaluation. The company confirmed that equipment installation has begun at its Taylor plant, dedicated to orders from U.S. automotive clients, and it is considering expanding production in the U.S. to meet future potential orders.
Goldman Sachs noted that the capacity squeeze from HBM is structural. HBM chips require larger die sizes and consume more wafers, with the trade ratio increasing each generation. Even if Samsung expands DRAM capacity, most of the incremental output will be absorbed by HBM, leaving traditional DRAM supply growth limited. This trend supports sustained high memory prices.
MX Division Prioritizes Market Share
Despite facing short-term operating losses, Samsung’s MX division continues to prioritize market share in key regions. The company anticipates that financial conditions may deteriorate further before improving, driven by rising storage costs. Regarding foldable smartphones, the company views new entrants positively, believing their entry boosts consumer awareness and validates the form factor.
Goldman Sachs views the MX division's strategy as trading short-term profits for long-term market positioning. Rising storage costs pressure smartphone business margins, but as a memory supplier, Samsung can benefit from memory price increases, partially offsetting this pressure. This vertical integration advantage becomes even more pronounced during memory upcycles.
Shareholder Returns to Be Decided Year-End
Samsung stated it is evaluating share buybacks and dividends, with a final decision expected once year-end results are finalized. Management reiterated that shareholdings by group affiliates will not be a primary factor in determining the shareholder return policy. Goldman Sachs believes clarity on the shareholder return policy could serve as a potential catalyst for the stock price.
Goldman Sachs' price targets are based on an EV/EBITDA segmented valuation for 2026–2027. The common share target is KRW 490,000, and the preferred share target is KRW 360,000, reflecting a 27% discount for preferred equity. Downside risks include severe deterioration in memory supply-demand dynamics, intensified HBM competition, widening foundry segment losses, and others.
If supply-demand conditions tighten further in 2027 relative to 2026, Samsung's memory business will enter a prolonged upcycle. Will the short-term losses in the smartphone segment be fully offset by memory profits?

Disclaimer
This article is a compilation and analysis by TechFlow Research of a third-party brokerage research report (Goldman Sachs, September 18, 2026), combined with publicly available market information. All ratings, price targets, earnings forecasts, and related judgments cited herein represent the views of the brokerage's analysts, reflect only their institutional stance, do not represent the views of TechFlow Research, and do not constitute any investment advice.
The market entails risks, and decisions must be made independently. This article should not be used as the basis for buying or selling any securities.








