As markets eye the situation in the Middle East, Nomura states that the &#34;long-term memory bull market&#34; will far exceed the &#34;short-term oil price surge.&#34;
According to the Speed Trading Desk, on March 23, Nomura Securities released its global market research report, indicating that the bull market cycle for the memory industry will be longer and more sustainable than oil price increases driven by geopolitics. The core variables are the continuous expansion of AI demand and the lag in supply expansion.
Nomura expects that the price increases for memory in the second quarter of 2026 will &#34;significantly exceed previous expectations,&#34; with quarter-on-quarter price increases for commodity DRAM and NAND reaching 51% and 50%, respectively. This figure represents a &#34;step-change&#34; leap from their previous forecasts of 6% and 20%.
Based on strong price trend expectations, Nomura has raised its target price for SK hynix by 24% from KRW 1.56 million to KRW 1.93 million, reiterating its &#34;Buy&#34; rating. Nomura believes that the recent stock pullback influenced by the Middle East situation presents an excellent opportunity for investors to buy on dips.
<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/36de350b-0ad3-43c1-b2b8-e76dbd3ba488.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/36de350b-0ad3-43c1-b2b8-e76dbd3ba488.png"/>
<h2>AI Demand Reshapes Industry Logic</h2>
In its report, Nomura delved into why the memory cycle can &#34;transcend&#34; macroeconomic fluctuations. A key logic is AI capital expenditure driving a &#34;long-term memory bull market.&#34;
Nomura believes that driven by the AI boom, the investment cycles of major tech companies will become the dominant force in the AI industry. This robust demand growth, stemming from the capital expenditures of tech giants on AI, makes its cycle &#34;longer and more sustainable than short-term oil price upcycles.&#34; In contrast, demand related to consumer devices (PCs and smartphones), which are sensitive to the macroeconomic environment, has significantly decreased its share, dropping from 60% a decade ago to less than 30% currently. This provides a solid foundation for the long-term stability of the memory industry.
Nomura describes the AI development trend as a &#34;black hole&#34; for memory demand. As AI demand evolves from simple text chatbots to enterprise-grade agents, and from text-driven to multimodal, memory requirements are becoming more specialized, shifting from a focus on HBM/LPDDR to DDR for RAG and next-generation SSDs for larger context processing.
<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/70cb9f0c-cc60-483d-a64c-1eb54ce8940e.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/70cb9f0c-cc60-483d-a64c-1eb54ce8940e.png"/>
<h2>Commercial Model &#34;Long-Term Agreement&#34; Enhances Stability</h2>
Nomura believes another highlight in the memory industry is the ongoing transformation of its commercial model. To cope with persistent supply shortages, memory suppliers and customers are actively shifting towards long-term agreements (LTAs).
Nomura wrote in the report: &#34;As supply shortages are likely to persist long-term, we believe memory manufacturers are seeking sustainable long-term agreements (LTAs) rather than simple price increases. Customers are also actively requesting to sign LTAs with suppliers to meet the continuously growing AI memory demand.&#34;
It is understood that these agreements include terms such as prepayments (as deposits), capacity guarantees, and pricing schemes. Nomura believes that although it is uncertain whether LTAs can perfectly ward off cycles, the memory industry is transitioning from &#34;spot-driven&#34; to &#34;long-term planning-driven,&#34; which will significantly enhance earnings stability.
<h2>Slow Supply Growth: Shortage May Continue Until 2028</h2>
On the supply side, Nomura's conclusion is relatively pessimistic. The report emphasizes that despite surging demand, supply growth will remain insufficient until at least early 2028.
Nomura highlights that memory (especially HBM for processing complex inference and training) remains a key bottleneck for AI. The report states: &#34;Increasing memory supply to keep up with high demand growth still requires physical time... At least until early 2028, supply growth will be insufficient to catch up with high memory demand growth.&#34;
Even with strong cash flows from giants like SK hynix, capacity expansion is difficult in the short term due to production cycle and technical bottlenecks.
Nomura forecasts that SK hynix's operating profit for 2026 and 2027 will be revised upward by 36% and 37% respectively, reaching KRW 256 trillion and KRW 365 trillion. For cash returns that investors are concerned about, Nomura expects SK hynix's free cash flow (FCF) yield for fiscal year 2026 to reach 19%, showing significant valuation attractiveness.
<img src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/e62bdc5f-ad91-4fc7-b1d7-6610f0878188.png?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://wpimg-wscn.awtmt.com/e62bdc5f-ad91-4fc7-b1d7-6610f0878188.png"/>

588780

512760

561980

Nomura believes that driven by the AI demand boom and lagging supply increases, the long-term bull market for memory chips will far exceed the short-term oil price upcycle. It is expected that by Q2 2026, prices for commodity DRAM and NAND will surge by 51% and 50% quarter-on-quarter, respectively, significantly exceeding previous forecasts of 6% and 20%. The memory industry is accelerating its shift towards long-term agreements, with supply shortages anticipated to persist into early 2028

- Nomura forecasts that the storage industry's bull market will significantly outlast temporary oil price increases due to ongoing AI-driven demand expansion.  
- They predict storage price growth in Q2 2026 will notably exceed prior estimates, with DRAM and NAND prices rising by 51% and 50% respectively, prompting an upgrade in SK Hynix's target price by 24%.  
- Despite expected supply shortages until at least 2028, the transition to long-term agreements among storage suppliers offers increased stability amidst fluctuating demand.

Wallstreetcn

GTJA Allianz SSE STAR Chip Design Thematic ETF

Guotai CES Semiconductor Chip Industry ETF

China Merchants CSI Semiconductor Industry ETF

VanEck Semiconductor ETF

Harvest SSE STAR Chip ETF

ChinaAMC SSE STAR Semiconductor Material & Equipment Thematic ETF

E Fund CSI Semiconductor Material & Equipment Thematic ETF

GTJA Allianz CSI All Share Semiconductors & Semiconductor Equipment ETF

SPDR S&P Semicon

Guotai CSI All Share Integrated Circuit ETF

First Trust Nasdaq Food & Semicon

VanEck Vectors Semiconductor UCITS ETF Accum A USD

Guotai CSI Semiconductor Material Equipment Theme ETF

iShares Semiconductor ETF

Invesco Semiconductors ETF

Harvest CSI All Share Integrated Circuit ETF

Direxion Semicon Bull 3X

ChinaAMC Guozheng Semiconductor Chip ETF

Spdr Select Tech

XL2CSOPHYNIX

Q2 Memory Price Surge "Far Exceeds Expectations," Nomura: "Long-Term Memory Bull Market" Far Outweighs "Short-Term Oil Price Rise"