The global semiconductor industry is currently experiencing a historic supply imbalance driven by generative artificial intelligence. In the first half of 2026, Micron (MU), a leading memory manufacturer, released financial results that pushed the intensity of this supercycle into the spotlight. While the market debates whether AI can sustain a longer cycle, investors are more concerned with a fundamental question: In a year when industry-wide revenue has doubled year-over-year, how can we distinguish between a natural cyclical peak and the onset of structural growth?
The Underlying Logic of the Semiconductor Cycle: Four Key Indicators of a Supercycle Peak
Historically, the semiconductor industry operates on a complete cycle of roughly three to four years. The year 2026 marks the peak of the current cycle. However, this cycle stands out because AI computing investments have surged in a short period, significantly boosting the price elasticity of DRAM and high-bandwidth memory (HBM), while demand for traditional PCs and smartphones has yet to fully recover, resulting in clear internal divergence within the industry.
A supercycle peak typically exhibits the following four characteristics:
First, capacity utilization approaches its physical limits, with price elasticity driving revenue growth while output growth slows. Currently, DRAM contract prices have risen 58% to 63% over the past quarter—a historically high range for peak periods. Second, top customers—namely hyperscale data center operators—reach short-term highs in capital expenditures. Industry forecasts show that in 2026, the combined capex of the world’s top five hyperscalers will rise from about $443 billion in 2025 to roughly $602 billion, a 36% increase, with approximately 75% allocated to AI-related semiconductors. Third, major industry players are simultaneously launching large-scale capacity expansions as supply begins to catch up with demand. In Micron’s fiscal Q2 2026, net capital expenditures reached $5 billion, while Samsung Electronics and SK Hynix are making similar investments in capacity growth. Fourth, after a significant increase in performance benchmarks, the marginal slowdown in year-over-year growth will trigger a valuation reset in a given quarter.
It’s important to note a key difference between the current cycle and past "silicon cycles." AI data centers consume memory at much higher rates than the traditional general-purpose server era. Each new generation of NVIDIA GPUs carries more HBM, creating a structural increase in demand. This means that even if the growth rate of AI training investments slows at some point, long-term memory consumption driven by inference could extend the plateau phase of this cycle.
The Limitations of P/E Valuation for Cyclical Stocks: Why Low Multiples Can Be a "Trap"
When analyzing cyclical industries, the traditional price-to-earnings (P/E) valuation method has systemic biases—especially dangerous at the peak of a cycle.
Take Micron’s 2026 valuation as an example: Based on fiscal Q3 guidance and full-year consensus earnings forecasts, the company’s adjusted EPS is expected to exceed $19, translating to a 12-month forward P/E of about 10x at prevailing share prices. This figure is central to recent discussions around Micron’s valuation.
At first glance, a 10x forward P/E is far below the Nasdaq 100’s average of roughly 22–25x, seemingly implying upside potential. However, this "low multiple" is based on peak earnings. Once the cycle turns, memory prices retreat, capacity utilization falls, and per-unit profits shrink, causing EPS to drop sharply over several quarters. Even if the share price holds steady or declines, the P/E ratio can expand dramatically relative to the new earnings base—leading to the phenomenon of "buying low P/E stocks at the cycle top and facing high P/E stocks at the bottom." Some analyses note that in the previous memory downturn, Micron’s share price fell by over 70%.
Morningstar’s valuation framework for commodity-based businesses highlights a critical logic: In a rising commodity price environment, producers benefit from the combined effect of fixed costs and financial leverage, resulting in enterprise value and share price elasticity exceeding that of the underlying commodity. Conversely, when prices fall, earnings contract much more than commodity prices themselves. This explains why cyclical stocks require a different analytical approach from stable growth companies.
Applying Morningstar’s "Peak P/E for Commodity Businesses" Framework to MU Risk Analysis
Applying Morningstar’s framework directly to Micron’s current situation yields the following insights:
As a memory manufacturer, Micron’s DRAM and NAND products are highly standardized commodities, with pricing power dictated by supply and demand. The current surge in performance is not due to product differentiation and market share gains, but rather historic premiums for HBM amid AI accelerator shortages. This means the company’s earnings structure is highly "price-dependent." Once supply constraints ease, mean reversion in profit margins becomes highly likely. The industry expects the HBM market to grow at a 41% CAGR to around $100 billion by 2028, but this also reveals another reality—all major manufacturers see the same massive opportunity and are investing billions from 2025 onward to secure their share.
When the three leading memory makers expand capacity simultaneously and most hyperscaler orders are locked in through long-term contracts until the end of 2026, bargaining power in subsequent contract negotiations will shift. In this scenario, supply gluts are not "black swan" events but a cyclical inevitability for the memory industry. For investors, the real risk isn’t Micron’s fundamentals deteriorating, but pricing cyclical assets using low P/E multiples during periods of peak sentiment.
It’s important to note that prudent valuation doesn’t mean rejecting investment opportunities. The key is identifying which stage of the cycle suits which strategy. The current cycle is at its strongest in terms of performance, but forward visibility is now highly uncertain. Any bullish thesis based solely on a low P/E must be supplemented by ongoing monitoring of supply-side changes and inventory levels.
Gate Launches Real Stock Trading: Bridging Traditional Markets with Stablecoins
In June 2026, Gate officially launched its stock trading service, enabling users to trade over 10,000 stocks and ETFs listed on the New York Stock Exchange and Nasdaq directly with USDT in their Gate accounts. This feature opens a unified channel for crypto market users to manage both digital and traditional securities assets within a single account system.
Compared to platforms that only support a few hundred tokenized stocks, Gate’s stock trading covers assets from major US exchanges including NYSE, Nasdaq, NYSE Arca, NYSE American, and BATS. Gate is also the first to offer 16×5 pre-market and after-hours trading, allowing users to capture significant price movements outside regular trading hours. This feature is now available on both Android and iOS. After completing KYC verification, users can access the stock trading interface through the TradFi section of the platform.
Structurally, Gate’s stock trading operates through a compliant brokerage model in partnership with Alpaca, with stocks held directly in the user’s name and funds managed in segregated accounts. For users transitioning from crypto to traditional asset allocation, this compliant structure reduces both counterparty and platform credit risk. Additionally, since Gate supports direct spot holdings, there are no overnight holding fees or funding rates common to CFD products, lowering the cost of long-term stock ownership.
For those interested in semiconductor cycle investment opportunities, Gate’s real stock trading offers several advantages: Users can buy shares of leading semiconductor companies like Micron, NVIDIA, and AMD, or related ETFs, directly with USDT—no need to switch platforms or endure lengthy cross-market fund transfers typical of traditional brokers. Gate supports fractional share trading starting from as little as 0.01 shares, with a minimum investment of just $1, breaking down the capital barriers of whole-share trading in the US market.
Additionally, Gate has fully integrated a VIP tier system: Users with holdings of $2,000 or more qualify for VIP status, enjoying the lowest stock trading fees at 0.023% and access to 1-on-1 account manager services.
How to Trade US Stocks on Gate
To get started, users must complete KYC verification on their Gate account and ensure the service is available in their region. Once verified, log in to the Gate App (Android users should update to the latest version; iOS users need version 8.21.5 or higher), navigate to the "TradFi" section, and select "Stocks" to browse real-time quotes and depth data for tradable US stocks and ETFs.
When placing trades, USDT balances in the user’s account are used directly for order margin and settlement. The system supports both market and limit orders. After execution, stock holdings are displayed in the Gate account under the same asset name, alongside digital assets in the portfolio overview.
For dividends, any cash dividends generated during the holding period are automatically credited to the user’s Gate account—no manual action required.
Conclusion
The semiconductor industry is undergoing a complex and highly contested cycle. AI-driven demand is real and robust, as evidenced by Micron’s record-breaking performance. However, cycles inevitably turn, and intuitive beliefs like "low P/E equals cheap" can pose serious risks for cyclical assets. Understanding Morningstar’s analysis of commodity business valuation and earnings elasticity can help investors make more rational decisions at this stage.
At the same time, Gate’s launch of real stock trading provides crypto-native users with a more convenient, low-barrier gateway to traditional securities markets. In a financial landscape where asset classes are increasingly intertwined, cross-market allocation is shifting from an "optional tool" to a "core skill." Whether for long-term value investing or tactical sector plays, managing both digital and traditional stock assets on a single platform is redefining the boundaries of individual market participation.
