DRAM ETF Issuer: Samsung, SK Hynix, Micron All Surpass $1 Trillion, the AI Era of Memory Chips Has Only Just Begun

marsbitPublished on 2026-06-24Last updated on 2026-06-24

Abstract

Authors: Dave Mazza, Thomas DiFazio | Source: Deep Tide TechFlow The article, written by Roundhill Investments (issuer of the DRAM ETF), responds to Morningstar's caution about investing in memory chip stocks. Morningstar warns of the sector's history of boom-bust cycles, a lack of economic moats, and potential momentum-driven overvaluation. Roundhill argues the current situation is structurally different due to AI. Key points in Roundhill's rebuttal include: * **Changed Demand & Supply Dynamics:** AI infrastructure, not consumer electronics, is now the primary growth driver for memory demand. New, strict long-term supply agreements with hyperscalers reflect the high capital intensity of advanced manufacturing. * **Existence of a Moat:** High-Bandwidth Memory (HBM), essential for AI, has extremely high manufacturing barriers. The market is dominated by Samsung, SK Hynix, and Micron, with new entrants blocked by technological complexity and long lead times for equipment like ASML's EUV machines. * **Strong Fundamental Outlook:** Analyst consensus projects the three companies will rank among the world's most profitable by 2027, with combined profits of $704 billion on over $1 trillion in revenue. Their operating margins have already reached record highs. * **Valuation Re-rating:** Despite significant stock price gains, memory stocks trade at attractive valuations (e.g., a median NTM P/E of 8.37x for the DRAM ETF) relative to projected explosive EPS growth. Roundhill su...

Authors:Dave Mazza, Thomas DiFazio

Compiled by: Deep Chao TechFlow

Deep Chao Guide: The market capitalizations of the world's three largest memory chip manufacturers have all surpassed $1 trillion. Morningstar promptly published an article reminding investors not to forget fundamentals. Roundhill Investments (the issuer of the DRAM ETF) countered point by point: AI infrastructure has reshaped the supply and demand structure of the memory industry; the manufacturing barriers of HBM make it impossible for new players to enter; and the combined projected profits of the three giants for 2027 are expected to reach $704 billion. Please note that the authors of this article are the managers of the DRAM ETF, a naturally bullish position.

The world's three largest memory chip manufacturers—Samsung Electronics (005930 KS), SK Hynix (000660 KS), and Micron (MU)—have all surpassed a market capitalization of $1 trillion, joining an exceedingly rare club. However, this milestone has also invited scrutiny.

Morningstar recently published a blog post reminding memory ETF investors not to overlook fundamentals, raising several pointed questions:

  • A Cautionary Tale from History: The memory industry has repeatedly experienced boom-bust cycles. Investors might be ignoring this history.
  • Memory Companies Lack Moats: Memory is essentially a commodity business. New capacity can always enter the market and erode pricing power; companies lack true barriers to protect profit margins.
  • The Rally Might Be Momentum-Driven, Not Fundamentals-Driven: The enthusiasm surrounding memory stocks reflects more excitement about AI than a sober analysis of earnings, profit margins, and supply-demand dynamics.
  • Valuations Have Skyrocketed: Memory stock prices have surged significantly, possibly running ahead of fundamentals.

Caption: Overview of the Memory Chip Industry

Roundhill's position is: this time is different. To understand the future of the memory industry, one must first look back at its past.

History is Indeed a Warning, But Is History Still Relevant?

The boom-bust cycles of memory chips are a fact. A classic cycle occurred in the mid-1990s. Microsoft released Windows 95 in August 1995, transforming personal computers from business-exclusive to consumer goods. The DRAM capacity per PC quadrupled from 1-2 megabits to 4-8 megabits. Manufacturers were caught off guard by the sudden demand, leading to a frenzy of factory construction and expansion, ultimately resulting in oversupply and a price collapse.

A similar story unfolded in the mid-2010s. When Apple released the iPhone 7 and upgraded the base storage from 16GB to 32GB, it seemed like a small change. But scaled up, demand surged dramatically. Manufacturers again invested heavily, followed by oversupply and price drops.

These cycles share a common pattern: technological breakthrough → surge in demand → manufacturer expansion → oversupply → price collapse.

The question is, is this pattern still applicable today?

The memory chip industry has undergone structural changes. Memory demand is no longer tied to consumer electronics upgrade cycles but to the compute power expansion of AI infrastructure. The scale of this market far exceeds a single smartphone upgrade wave, and its growth potential is much larger.

DRAM and NAND prices have increased more than fivefold since January 2024, leading hyperscale customers to begin demanding long-term supply agreements to lock in bandwidth. Historically, long-term supply agreements in the memory industry have been loose frameworks subject to market changes. But this model has changed. During its January 2026 earnings call, SK Hynix stated that current agreements reflect "strong reciprocal commitments" between customers and suppliers, citing the high capital intensity of advanced memory manufacturing. Micron also reported similar long-term agreement conditions.

Caption: DRAM and NAND Price Trends

The Moat of Memory Chips: Manufacturing Complexity

Not all memory chips are created equal. The memory driving today's AI systems is called High Bandwidth Memory (HBM), which is entirely different from the memory in phones and computers. HBM is specifically designed for AI workloads, with extremely demanding manufacturing conditions.

Data from Goldman Sachs shows that SK Hynix, Samsung, and Micron control nearly the entire global supply of HBM. This industry has consolidated over decades, and the accumulated manufacturing expertise cannot be replicated overnight. Manufacturing complexity itself is the moat, and it's the very reason these three companies have reached this point.

Caption: Global HBM Market Share Distribution

This logic is completely different from that of the old cycles. The past was: demand rises → new capacity enters → prices collapse. The current bottleneck is not capital or willingness, but technical capability. SK Hynix currently controls about 58% of the global HBM supply. On June 2, it announced plans to double wafer capacity over the next five years while warning that supply shortages would persist until 2030. Building a new factory takes at least 3 years, or over 5 years for a completely new site.

Furthermore, ASML—the sole manufacturer of extreme ultraviolet (EUV) lithography machines, essential for producing advanced memory chips—entered 2026 with a backlog of €38.8 billion, exceeding its full-year projected sales. The delivery lead time for a single EUV machine exceeds 12 months. This bottleneck cannot be resolved in the short term.

Fundamentals: Memory Manufacturers Poised to Join Ranks of World's Most Profitable Companies

The earnings, revenue, and margin expectations for Samsung, SK Hynix, and Micron reflect the secular wave of AI adoption. Bloomberg consensus estimates show that by 2027, these three companies will rank among the world's top ten most profitable companies.

Caption: Forecast of the World's Most Profitable Companies for 2027 (Bloomberg Consensus)

The combined bottom-line profit forecast for the three companies for 2027 is $704 billion, with total revenue exceeding $1 trillion.

Caption: Revenue Forecast for the Three Major Memory Manufacturers

Caption: Profit Forecast for the Three Major Memory Manufacturers

In terms of margins, the operating gross margins of Samsung, SK Hynix, and Micron have reached historical records, surpassing the previous highs of 2018.

Caption: Historical Gross Margin Trends for the Three Major Memory Manufacturers

Numbers like these have never been seen in the history of the memory industry. Even if growth slows, in the context of generative AI's continued integration into the global economy, the memory industry is expected to settle on an unprecedentedly high baseline.

Valuation Reassessment in the New Era of Profitability

Historic stock price performance combined with substantial upward revisions to fundamentals suggests the industry is undergoing a significant revaluation driven by earnings growth and margin expansion.

SK Hynix and Samsung are two case studies. For nearly a decade, the NTM (Next Twelve Months) price-to-book ratios of both stocks fluctuated within a certain range, constrained by the boom-bust earnings characteristics of the memory industry. But this ceiling may no longer apply. The expected ROE (Return on Equity) for both companies has surged to levels never before seen in the history of the memory industry. The valuation frameworks investors have long used to judge these stocks need to be re-examined.

Caption: SK Hynix NTM Price-to-Book Ratio and ROE Trend

Caption: Samsung Electronics NTM Price-to-Book Ratio and ROE Trend

Despite the recent dramatic stock price gains, the median NTM P/E ratio of the DRAM ETF's holdings is only 8.37x, which is attractive compared to the broader tech sector. Meanwhile, the median current fiscal year EPS growth rate for the portfolio is 632%. Arguing that memory stocks are overvalued is essentially applying old data to a new industry. In Roundhill's view, the gap between historical valuation conventions and current fundamental performance represents opportunity.

Caption: DRAM ETF Holdings Valuation and Earnings Growth Overview

Conclusion: Why Roundhill Isn't Worried

Being skeptical of surging stock prices is reasonable; fundamentals always matter in the long run. But in this case, fundamentals are precisely the reason for the rise in memory stocks.

The old cycles were characterized by: demand explosions without an upper bound, manufacturers over-expanding, and an inevitable price collapse. Today's situation is structurally different: manufacturing barriers limit new entrants; industry leaders themselves say supply shortages will persist until 2030; and the earnings cycle has just begun to reflect the scale of AI infrastructure buildout.

Roundhill believes the market is currently pricing not a bubble, but an industry that has struggled through booms and busts for decades entering a new era.

⚠️ Editor's Note: The authors of this article, Dave Mazza and Thomas DiFazio, are both members of Roundhill Investments, the issuer and manager of the DRAM ETF (Roundhill Memory ETF). The article's position is naturally bullish. Readers should form their judgment by considering third-party views such as those from Morningstar. The ETF risk disclosures and legal disclaimers from the end of the original article have been omitted. Please refer to the original link for complete information.

Trending Cryptos

Related Questions

QWhat is the main reason why Roundhill Investments argues that the current boom in memory chip stocks is different from historical cycles?

ARoundhill Investments argues that the current boom is structurally different because AI infrastructure demand is reshaping supply/demand dynamics. Unlike past cycles tied to consumer electronics, demand is now driven by secular AI infrastructure expansion. High manufacturing barriers for HBM, long-term supply agreements with hyperscalers, and severe capital/technology bottlenecks (e.g., ASML EUV tool backlogs) prevent new competitors from easily entering and causing oversupply.

QWhat are the key arguments made by Morningstar to urge caution among memory ETF investors?

AMorningstar's key arguments are: 1) The memory industry has a history of boom-bust cycles, which investors may be ignoring. 2) Memory is essentially a commodity business where new capacity can enter and erode pricing power, implying a lack of sustainable moats. 3) The recent rally may be driven by AI hype and momentum rather than冷静 (calm) analysis of fundamentals like profits and supply/demand. 4) Valuations have already soared, potentially running ahead of fundamentals.

QWhat is HBM and why does its manufacturing create a moat for the leading memory companies?

AHBM (High Bandwidth Memory) is a specialized memory chip designed for AI workloads. Its manufacturing creates a moat due to extreme complexity and high barriers to entry. The process requires advanced technology (like EUV lithography from ASML), decades of accumulated manufacturing expertise, and is highly capital intensive. This limits production almost entirely to the three incumbents—SK Hynix, Samsung, and Micron—and prevents new players from quickly adding capacity to disrupt the market.

QAccording to the article's projections, what is the combined profit for Samsung, SK Hynix, and Micron expected to be by 2027?

AAccording to Bloomberg consensus estimates cited in the article, the combined bottom-line profit for Samsung, SK Hynix, and Micron is projected to reach $704 billion by 2027.

QHow does Roundhill justify that memory stocks are not overvalued despite their significant price increases?

ARoundhill justifies this by pointing to a disconnect between historical valuation frameworks and new fundamental realities. They note that while prices have risen, the median NTM P/E ratio of the DRAM ETF holdings is only 8.37x, which is attractive compared to broader tech stocks. Meanwhile, the median EPS growth for the current fiscal year is 632%. They argue that soaring expected ROE and profit margins justify a re-rating, meaning old valuation ceilings based on boom-bust cycles no longer apply to the new, more profitable AI-driven era.

Related Reads

IOSG Founder: Ethereum Doesn't Need Another Leap of Technical Faith, It Needs a Musk-style Compromise

Jocy, founder of IOSG Ventures, argues that Ethereum does not need renewed technological faith but a "Musk-like compromise." The recent formation of ETHLabs—funded by major ETH holders like BitMine and Lubin—highlights a market-driven move to fill a gap left by the Ethereum Foundation (EF), signaling a loss of confidence in its decentralized, hands-off approach. The core critique contrasts Vitalik Buterin's (V) idealistic, technology-first vision with Elon Musk's pragmatic, business-driven execution. The author asserts Ethereum's current shortage is not another technical roadmap but a clear, real-world application narrative and a leader willing to engage directly with commercial realities—like Musk. Internal issues are emphasized, citing EF's management problems and talent drain. While the new decentralized model with independent nodes like ETHLabs addresses the single foundation's limitations, it risks fragmentation without cohesive direction. True cohesion, the author suggests, must come from a shared, compelling narrative around ETH's value, not just from aligned financial interests. Independence claims for new entities are seen as aspirational, needing years of transparency to build trust. The ultimate threat is not competitors like Solana, but the broader shift of attention and talent toward AI. Ethereum has a limited window—12 to 18 months—to recapture focus by delivering tangible, real-world applications. The conclusion urges V to shift from abstract ideals to grounded, pragmatic leadership. The time for this crucial pivot is running out.

marsbit1h ago

IOSG Founder: Ethereum Doesn't Need Another Leap of Technical Faith, It Needs a Musk-style Compromise

marsbit1h ago

Google Starts Selling TPUs, Big Tech Aims to Produce "Low-Cost Tokens" with AI Chips

Google has begun selling its proprietary TPU chips and AI computing hardware directly to third-party data centers and clients, marking a strategic shift. Previously only accessible via cloud rentals, TPUs are specialized processors designed for the matrix and tensor operations central to AI models. By combining thousands into supercomputing clusters managed by CPUs, Google achieves high-efficiency AI processing. This move enables Google’s Gemini AI to offer competitive token pricing, challenging rivals like OpenAI. It also signals a broader industry trend where AI compute is becoming a commoditized resource like electricity. While NVIDIA remains dominant with its CUDA ecosystem and high-performance GPUs, the focus is shifting from raw power to cost efficiency and system integration. Google’s approach mirrors NVIDIA’s by selling an entire ecosystem—hardware, software, and data center expertise—rather than just chips. This threatens NVIDIA’s grip on the mid-range inference market, where lower-cost, efficient solutions are increasingly demanded. Similarly, cloud providers like Huawei Cloud and Alibaba Cloud in China are developing their own AI chip ecosystems (e.g., Ascend, Zhenwu), packaging chips, clusters, and tools into full-stack solutions. They aim to reduce token costs and capture market share through integrated systems. In summary, the AI infrastructure race is evolving from a competition for the strongest chips to a contest for the most efficient and cost-effective systems. Google’s TPU sales highlight this transition, emphasizing that future success lies in delivering affordable, scalable AI compute as a foundational service.

marsbit1h ago

Google Starts Selling TPUs, Big Tech Aims to Produce "Low-Cost Tokens" with AI Chips

marsbit1h ago

Trading

Spot
Futures

Hot Articles

Ethena: Building a New Era of Web3‑Native Digital Dollars

Ethena is an Ethereum‑based synthetic dollar protocol that delivers crypto‑native monetary solutions, including USDe, a synthetic dollar, and sUSDe, a globally accessible U.S. dollar savings asset.

53.3k Total ViewsPublished 2026.03.16Updated 2026.03.16

Ethena: Building a New Era of Web3‑Native Digital Dollars

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of ERA (ERA) are presented below.

活动图片