--- title: "After Zhongji Innolight Breaks the Trillion-Mark Threshold" type: "News" locale: "en" url: "https://longbridge.com/en/news/283950811.md" description: "Zhongji Innolight surpassing a trillion-dollar market cap marks the peak of the AI technology optical module industry. Analysts believe that the combination of AI supply-demand gaps and overseas new energy investments will drive market development. Historical data shows that leading stocks breaking the trillion mark typically have further upside potential, contingent on earnings growth, profit improvement, and industry competition dynamics. By 2026, AI pricing is expected to shift towards supply-demand gaps, making technology and overseas investment the mainstream directions" datetime: "2026-04-24T07:20:33.000Z" locales: - [zh-CN](https://longbridge.com/zh-CN/news/283950811.md) - [en](https://longbridge.com/en/news/283950811.md) - [zh-HK](https://longbridge.com/zh-HK/news/283950811.md) --- # After Zhongji Innolight Breaks the Trillion-Mark Threshold **Core Conclusion:** Standing at the moment when domestic AI technology optical module leader Zhongji Innolight breaks the trillion mark, we once again emphasize that the probability of "AI Supply-Demand Gap (Price Hikes) + Overseas New Energy" becoming the new "Ning Combination" is increasing. Even with high-level volatility, this round of collective holding is unlikely to end easily. In fact, for any industrial trend, a leading stock breaking the trillion mark is a very important milestone event, often indicating that the industry is in the peak pricing stage. Here, we briefly review A-share industrial trend stocks that broke the trillion market cap in the past decade. We find that while the stock prices of these core leaders become more volatile after breaking the threshold, there is basically no case where they begin a trend downward immediately after breaking the trillion mark; subsequent significant upside usually depends on the following three points: 1. Whether high earnings growth possesses scarcity compared to other industries during the same period; 2. Whether the second derivative of profits can continue to improve, particularly for those with sustainable earnings realization; 3. Whether the industry competition landscape remains stable. Further breaking down the gains, before reaching the trillion mark, these core industrial leaders mostly profited from earnings explosions (Davis double play), whereas significant gains after breaking the trillion mark were often driven by substantial valuation increases under high earnings growth (looking back). More importantly, after industrial trend leaders break the trillion mark, secondary-tier products and supply-demand gap segments in the industrial chain (price hikes) outperform core leaders significantly. For example, in mid-to-late 2020, Jiugui Liquor clearly began to outperform Kweichow Moutai; similarly, in 2021, Tianci Materials and Lithium Carbonate outperformed CATL. Based on our exclusive four-stage theory of tech investment, we tend to believe that by 2026, AI pricing will transition from infrastructure segment pricing to supply-demand gap segments, corresponding to a gradual shift from "Large Optics" to "Optical Aperture." Based on Q1 2026 institutional holdings observation, the proportion of holdings in Technology + Overseas has broken through 70%, becoming the main base for institutional investment, fully validating our clear judgment over the past few years that "Technology + Overseas are the two main lines of industrial fundamentals." Combining quarterly report views from fund managers we have long tracked, core investment directions focus on: 1) Core AI computing power directions (optical modules, fiber optic cables, supply-demand gaps, etc.). 2) Directions in the overseas sector expected to achieve continuous ROE improvement (construction machinery, grid equipment, wind power, lithium battery energy storage, etc.). 3) Some resource products with commodity attributes that have price hike expectations. Regarding AI technology pricing, institutional TMT holdings in the first quarter remained at a high level near 40%. On April 23rd, Zhongji Innolight, the leader in optical modules, breaking the trillion market cap triggered wider market discussion. Based on the "Trillion Moment" of industrial trend leaders and current market observations, we offer the following insights: 1. In fact, for any industrial trend, a leading stock breaking the trillion mark is a very important milestone event, often indicating that the industry is in the peak pricing stage. As the AI industry wave enters its mid-term, one must understand that all trends involve collective holding, and the final frenzy is the craziest. From the Mao Index (peak holdings 45%) in 2019-2021 → Pan-New Energy (peak holdings approaching 40%) in 2020-2022 → AI Technology (current holdings around 37%) from 2023 to present, the essence is the same. Objectively, it is difficult to conclude that all industrial waves start to decline upon reaching a specific portfolio level, as each wave has its unique characteristics. Based on our previous research conclusions, generally, if high stock prices overdraw earnings growth by more than three years, caution is needed; the strongest industrial trend for the best company exceeds five years as an extreme limit. In this calculation process, the reasonable PE valuation center is generally 20x, not exceeding 30x at the highest. Additionally, the end of industrial trend stocks often presents a double-top pattern (M top), where the second peak is the key point to grasp. 2. Reviewing the past decade, stock price volatility increases after industrial trend stocks break the trillion mark. If earnings continue to be realized, there is significant upside space (this variance is large, potentially doubling); otherwise, they tend to hover near the trillion mark. However, there is basically no case of a trend-driven decline immediately after breaking the trillion mark. Breaking down the gains: before the trillion mark, profits came from earnings explosions (Davis double play); after breaking the trillion mark, continued significant gains often come from valuation appreciation (looking back). It can be seen that at the moment industrial trend leaders break the trillion mark, their profit levels (TTM) range from 50 billion to hundreds of billions, and valuation levels (PE) vary widely from dozens to over a hundred times, showing significant variance and being heavily influenced by industry and pricing environments. 3. Regarding whether to add positions at high levels after breaking the trillion mark, the essence depends on whether the second derivative of profits accelerates. In this process, whether high earnings growth possesses scarcity compared to other industries during the same period is also crucial. For example, the core of the 2021 new energy rally was strong industrial prosperity, with profit growth rates continuously revised upwards and exceeding expectations in 2021. Against a backdrop of lacking profit elasticity in other sectors, this high growth intensified capital clustering, ultimately leading to bubble-like pricing in Q4 2021. Furthermore, the industry competition landscape is the ultimate source of excess returns. A stable competition landscape is a key driver of excess performance; otherwise, even if an absolute benefit is formed due to the industrial trend's beta, it is difficult to sustain alpha, making it hard to earn higher excess returns. 4. Currently, the collective pricing around AI technology is far from ending. The core lies in the fact that high AI capital expenditures are likely to persist for the next half-year without being falsified. Correspondingly, this round of AI has not seen severe erosion of downstream cash flows by upstream entities, nor have there been signs of slowing AI capital expenditures. From the perspective of free cash flow of major overseas companies, the four major cloud providers (Microsoft + Google + META + Amazon) have a TTM free cash flow of $207.166 billion, which is still far higher than NVIDIA's $96.677 billion free cash flow. Therefore, based on the internal laws of industrial movement, after industrial leaders break the trillion mark, this round of AI does not show obvious bubble characteristics. 5. After industrial trend leaders break the trillion mark, secondary-tier products and supply-demand gap segments in the industrial chain (price hikes) outperform core leaders. This indicates that when the industrial trend enters the peak pricing stage, the market trend tends to diffuse. For example, in mid-to-late 2020, Jiugui Liquor clearly began to outperform Kweichow Moutai; similarly, in 2021, Tianci Materials and Lithium Carbonate outperformed CATL. Generally speaking, a round of industrial trend pricing follows a pattern of Large Market Cap (early stage) → Small Market Cap (peak stage) → Large Market Cap (often the topping stage). For instance, early mobile internet followed Apple → Apple Chain → Apple, while later new energy lithium batteries followed Ning Combination → Lithium Batteries. 6. According to the four-stage theory of tech investment, the sequence of evolution is Giants (ChatGPT, NVIDIA, Microsoft, etc.) → Infrastructure Investment (Computing Power) → Key Industrial Chain Links (AI Chips) → Supply-Demand Gap Application End (Storage, Fiber Optics, Power, Energy Storage, Copper, etc. are upstream supply-demand gaps; Robots, Intelligent Driving, AI Ecosystem Software are downstream supply-demand gaps; price hikes are unavoidable). Analogizing to 2021, where the Ning Combination lithium mining sector's gains and institutional additions far exceeded lithium batteries, we believe that by 2026, AI technology pricing should gradually transition to supply-demand gap segments. Price hikes are unavoidable, a point validated by recent holding performances. In summary: Over the past period, we repeatedly emphasized that 2026 should be analogized to 2021. In 2021, amidst mild inflation, resilient global economy, and a strong dollar macro background, the broad market fell at the beginning of the year then entered wide-range oscillation. The underlying core of the rebound in the second half was that A-shares had fundamental support, especially supported by the pan-new energy represented by the "Ning Combination." Similarly, in 2026, according to the "Three Bulls" narrative, a purely liquidity-driven bull market tends to wane once the Shanghai Composite Index breaks 4,000 points. Subsequent markets with fundamentals represent a bull market continuation, offering opportunities for the broader market. This fundamental support currently relies more on "Technology + Overseas" continuously improving profitability ratios. During the process of achieving a "30%-60%" share (historical experience suggests a breakthrough of 50% triggers a new upward cycle for A-shares), this validates our earlier judgment of "not a bear market transition," further analogizing to 2021's structural adjustment "from Mao Index to Ning Combination." Consequently, the probability of the strongest varieties in Technology + Overseas, namely "AI Price Hikes + Overseas New Energy," becoming the new "Ning Combination" is also increasing. Furthermore, based on dynamic observations of ordinary and equity-oriented funds in Q1 2026, the "Four Rebalancings" proposed at the beginning of the year are being verified. Currently, the old-new rebalancing (Technology/Cyclicals) has been verified; the decrease in financial attributes within resources (Gold/Silver) and the resurgence of commodity attributes (Oil-Chemicals-Coal) have also been verified. Internal rebalancing within Technology (AI infrastructure moving towards AI price hike supply-demand gaps, e.g., CATL moving towards lithium mines and auto parts in 2021) + Overseas (from downstream consumer goods to mid-to-upstream manufacturing centered on European energy security (new energy & power equipment) and Global South industrialization (construction machinery)) is still ongoing. First, Q1 Institutional Addition Logic: 1) AI hardware sectors with the strongest earnings certainty (e.g., optical modules, fiber optic cables, grid equipment, gas turbines, etc.); 2) Mid-to-upstream overseas sectors + prosperous varieties (chemicals, energy storage, etc.); 3) Beneficiaries of rising energy product prices (petrochemicals, coal, etc.). Second, Q1 Institutional Reduction Logic: 1) Certain resource products (Copper, etc.); 2) Weak earnings realization within AI (Computers, Media, certain Electronics, HK Internet, etc.). 1. By Industry, the top five sectors added by institutions in Q1 2026 were: Communications, Basic Chemicals, Electric Power Equipment & New Energy, Pharmaceuticals, Petrochemicals; the top five reduced were: Nonferrous Metals, Electronics, Non-Bank Financials, Computers, Home Appliances. 2. In Q1 2026, institutions showed a trend of "Buying Hardware, Selling Software" regarding the AI industrial chain, primarily seeking earnings certainty within the AI chain, beginning to exhibit certain "collective holding" characteristics. Among them, Fiber Optic Cables (+1.69pct), Semiconductor Equipment (+0.99pct), Grid Equipment (+0.51pct), and Optical Modules (+0.34pct) saw increases, while PCB (-1.80pct), HK Internet (-1.85pct), Games (-0.34pct), and Computer Equipment (-0.71pct) saw significant reductions, with funds clustering in sectors with better earnings realization. 3. In Q1 2026, institutional allocation for overseas moved towards concentrating on upstream Building Materials + Chemicals + New Energy. Major additions concentrated in Chemicals (+0.98pct, among which Potash Fertilizer and Chemical Preparations saw prominent increases) and Lithium Batteries within New Energy (+0.20pct). Downstream Automotive (-0.56pct) and Home Appliances (-0.67pct) were reduced. 4. In Q1 2026, institutional holdings for resource products showed some divergence. Within Nonferrous Metals, Copper (-1.67pct) was reduced, while Shipping (+0.43pct), Coal (+0.27pct), and Petrochemicals (+0.53pct), benefiting from rising oil and gas prices, were all increased. 5. In Q1 2026, institutions reduced holdings in the Big Finance sector. Insurance (+0.63pct) and Securities (+0.30pct) were both reduced, while Banks (+0.23pct) saw a slight increase. 6. In Q1 2026, institutions reduced holdings in the Domestic Consumption sector. Baijiu (-0.18pct), Consumer Services (-0.10pct), and Retail Trade (-0.02pct) were all reduced. Risk Disclosure and Disclaimer Market investments carry risks; proceed with caution. This article does not constitute personal investment advice and does not take into account the special investment objectives, financial status, or needs of individual users. Users should consider whether any opinions, viewpoints, or conclusions herein align with their specific circumstances. Investments made based on this content are the user's own responsibility. ### Related Stocks - [300308.CN](https://longbridge.com/en/quote/300308.CN.md) - [600519.CN](https://longbridge.com/en/quote/600519.CN.md) - [000799.CN](https://longbridge.com/en/quote/000799.CN.md) - [002709.CN](https://longbridge.com/en/quote/002709.CN.md) - [300750.CN](https://longbridge.com/en/quote/300750.CN.md) - [NVDA.US](https://longbridge.com/en/quote/NVDA.US.md) - [MSFT.US](https://longbridge.com/en/quote/MSFT.US.md) - [AAPL.US](https://longbridge.com/en/quote/AAPL.US.md) - [GOOGL.US](https://longbridge.com/en/quote/GOOGL.US.md) - [GOOG.US](https://longbridge.com/en/quote/GOOG.US.md) - [META.US](https://longbridge.com/en/quote/META.US.md) - [AMZN.US](https://longbridge.com/en/quote/AMZN.US.md) - [HCCD.SG](https://longbridge.com/en/quote/HCCD.SG.md) - [03750.HK](https://longbridge.com/en/quote/03750.HK.md) - [NVD.DE](https://longbridge.com/en/quote/NVD.DE.md) ## Related News & Research - [AI face is taking over — and driving plastic surgeons crazy](https://longbridge.com/en/news/286641783.md) - [Jack Antonoff tells people who are making AI art to 'drive right off that cliff'](https://longbridge.com/en/news/286592426.md) - [06:07 ETStandardC Launches AI Platform for Financial Institutions, Where Customer PII Is Never Shared With AI Models (Patent Pending)](https://longbridge.com/en/news/286892045.md) - [11:30 ETDageno Launches Issues Panel and High-volume Prompt Miner to Help Growing Brands Turn AI Search Signals Into Growth Tasks](https://longbridge.com/en/news/286939643.md) - [](https://longbridge.com/en/news/286300823.md)