Barclays Lifts Samsung, SK Hynix Targets on $137.55 HBM Demand
1h ago|4 min readStandard
Fazen Markets Editorial Desk
Collective editorial team · methodology
samsungsk-hynix
Fazen Markets Editorial Desk
Collective editorial team · methodology
Trades XAUUSD 24/5 on autopilot. Verified Myfxbook performance. Free forever.
Risk warning: CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. The majority of retail investor accounts lose money when trading CFDs. Vortex HFT is informational software — not investment advice. Past performance does not guarantee future results.
Analysts at Barclays raised their price targets on shares of South Korean memory chip leaders Samsung Electronics and SK Hynix, citing exceptionally strong demand for high-bandwidth memory (HBM) used in artificial intelligence systems. The firm announced the revised targets on June 29, 2026, reflecting a structural shift in the semiconductor market toward AI-optimized hardware. The upgrade arrives as the broader market shows mixed performance, with the SPDR S&P Semiconductor ETF trading near its daily low of $137.22 as of 14:51 UTC today.
This analyst action underscores a multi-year transformation within the memory sector. Historically, memory chipmakers were viewed as cyclical commodity producers, with profits tied to volatile pricing for standard DRAM and NAND flash. The surge in generative AI model development, which began accelerating in late 2022, fundamentally altered that dynamic. AI training requires specialized, high-performance memory stacks with vastly superior data transfer rates compared to conventional chips.
High-bandwidth memory (HBM) has emerged as the critical bottleneck and enabling technology for advanced AI accelerators. Each new generation of HBM offers significantly higher bandwidth and density, directly impacting the performance of AI server clusters. Chip designers Nvidia and AMD now design their flagship GPUs around specific HBM specifications, locking in long-term supply agreements with a limited number of qualified manufacturers.
The current macro backdrop features elevated but stabilizing interest rates, with the 10-year Treasury yield holding near 4.3%. This environment pressures growth stocks but favors companies with demonstrable, near-term earnings power from tangible technological shifts. The Barclays upgrade signals confidence that the HBM super-cycle has sufficient runway to override broader economic concerns for the sector's leaders.
Barclays' revised targets imply significant upside for both Korean chip giants, though the firm did not disclose specific new price levels in the initial report. The upgrade is grounded in concrete supply and demand metrics. The global HBM market, valued at approximately $14 billion in 2025, is projected to grow at a compound annual growth rate exceeding 35% through 2030.
Samsung and SK Hynix collectively control over 90% of the HBM market, with SK Hynix holding an estimated 50% share due to its early partnership with Nvidia. This duopoly commands premium pricing; HBM chips sell for multiples of the price of equivalent-capacity commodity DRAM. The financial impact is clear in forward earnings estimates. Consensus analyst forecasts for SK Hynix's 2027 earnings per share have risen by over 22% in the last quarter alone.
| Metric | SK Hynix (Est.) | Samsung (Est.) |
|---|---|---|
| 2027 HBM Revenue Growth | ~45% Year-over-Year | ~40% Year-over-Year |
| HBM as % of Total DRAM Sales | >50% | ~35% |
For comparison, the Philadelphia Semiconductor Index (SOX) has gained 12% year-to-date, while the broader S&P 500 is up 8%. The targeted upgrade for Samsung and SK Hynix suggests their growth trajectory will continue to outpace the sector.
The immediate second-order effect is capital rotation within the tech sector. Funds may flow out of pure-play AI software names with stretched valuations and into the foundational hardware layer. Primary beneficiaries include semiconductor equipment makers like ASML and Lam Research, which supply the advanced lithography and etching tools needed for HBM production. Companies in the advanced packaging supply chain, such as Taiwan's ASE Group, also stand to gain from increased demand for the complex process of stacking HBM dies.
Conversely, laggards in the HBM race face clear risks. U.S.-based Micron Technology, while investing heavily, remains a distant third in market share and technology cadence. Any execution misstep could widen the gap. Traditional data center players reliant on older memory architectures may face cost inflation and supply constraints for new server builds.
A key acknowledged risk is customer concentration. Both Korean firms are heavily reliant on orders from Nvidia and, to a lesser extent, AMD. A significant design win shift or a slowdown in AI infrastructure spending by major cloud providers like Microsoft Azure or Amazon AWS could rapidly alter the demand picture. Current positioning data from futures markets shows institutional investors maintaining net long positions in semiconductor ETFs, with recent options flow indicating increased hedging activity via puts on broader indices, suggesting a view that chip strength may be isolated.
The next major catalyst for the thesis is Samsung's and SK Hynix's Q2 2026 earnings releases, scheduled for late July. Investors will scrutinize HBM revenue figures, yield improvements, and commentary on supply agreements for 2027. Samsung's roadmap for its next-generation HBM4 technology, expected in detail at its annual Memory Tech Day in October, will be critical for assessing its competitive stance against SK Hynix.
Market levels to watch include the $135 support zone for the SPDR S&P Semiconductor ETF (XSD), which aligns with its 200-day moving average. A sustained break above its recent high of $139.56 would confirm the bullish momentum signaled by analyst actions. For SK Hynix, the key threshold is its all-time high share price, set earlier this year; a breakout could trigger further algorithmic buying.
Investor attention should also focus on the U.S. Commerce Department's next round of export control updates, expected by September 2026. Stricter limits on AI chip sales to China could indirectly affect HBM demand forecasts, though most analysts believe the impact on leading-edge HBM used in restricted chips would be minimal.
High-bandwidth memory (HBM) is a type of DRAM where memory chips are stacked vertically and connected using silicon vias, creating a much wider data pathway than traditional side-by-side layouts. This architecture is crucial for AI because training large language models requires moving massive datasets between the processor and memory at extreme speeds. Standard memory acts as a bottleneck; HBM's high bandwidth allows AI accelerators like GPUs to function at full capacity, drastically reducing training times for complex models.
Vortex HFT is our free MT4/MT5 Expert Advisor. Verified Myfxbook performance. No subscription. No fees. Trades 24/5.
Position yourself for the macro moves discussed above
Start TradingSponsored
Open a demo account in 30 seconds. No deposit required.
CFDs are complex instruments and come with a high risk of losing money rapidly due to leverage. You should consider whether you understand how CFDs work and whether you can afford to take the high risk of losing your money.