Summary of Key Points
On July 16, ChangXin Technology launched its online subscription for funds, raising a total of 29.5 billion yuan (the largest IPO in the A-share market in 2026 and the second-largest on the STAR Market), marking the debut of China's leading DRAM manufacturer in the capital market. The article discusses several key issues: whether the storage industry cycle has disappeared, the opportunities and challenges faced by ChangXin, and the significance of China's storage industry as a whole. The rise in AI demand has led to a surge in the demand for memory chips, but the cycle has not vanished; it is likely to peak in the second half of 2027 to early 2028. Although ChangXin's financial performance (revenue and profit growth) is impressive, there is a technological gap with leading manufacturers, and the funds raised are limited for HBM development. This IPO can be seen as a "coming-of-age ceremony" for China's storage industry, indicating its entry into the capital market, but it also comes with challenges of catching up with technology and managing cyclical fluctuations.
I. ChangXin's IPO: The Largest in the A-share Market in 2026, Yet the Cycle Looms Large
ChangXin's fundraising of 29.5 billion yuan makes it the largest IPO in the A-share market this year and the second-largest on the STAR Market. However, the experience of SK Hynix a week earlier serves as a cautionary tale: SK Hynix went public in the US with a valuation of $24.5 billion, but its stock price plummeted by nearly 20% on the first day, one of the worst daily drops in over two decades. This shows that despite the surge in AI demand, the issue of market cycles (fluctuations in supply and demand) has not been resolved; it has merely been temporarily masked.
Why do these cycles persist? The reason is that it takes more than two years to build new chip fabrication facilities and expand production capacity on the supply side, while demand from sectors like AI changes rapidly. Even if AI demand surges, supply may not keep up in time, and when supply increases, demand could have already shifted, leading to cyclical patterns. The article suggests that the peak of the current cycle might occur in the second half of 2027 to early 2028, with subsequent price fluctuations due to increased supply.
II. The Storage Cycle Has Not Disappeared; It Has Only Been Extended: 2028 Could Be a Critical Turning Point
Many believe that AI demand will eliminate market cycles in the storage industry, but the article argues that these cycles have merely been prolonged:
1. HBM and long-term contracts can help stabilize prices, but they are not a panacea: There is high demand for HBM (high-bandwidth memory essential for AI), and leading manufacturers use long-term contracts to lock in customers and stabilize prices. However, these contracts limit profit margins and reduce price flexibility. Additionally, new HBM production capacity from major players (such as Samsung, SK Hynix, and Micron) is expected to be released around 2028, which could reverse the cycle if AI demand falls short of expectations.
2. DRAM's Dual Benefits and Potential Cycle Turning Point: DRAM (standard memory) is currently benefiting from two factors: servers' need for DDR5 memory (which complements HBM), and leading manufacturers shifting production to HBM, resulting in a shortage of DRAM supply. However, DRAM is heavily influenced by the consumer electronics sector (smartphones, PCs), which is experiencing declining shipments, leading to increased inventory levels among downstream manufacturers. This could result in an earlier cycle reversal for DRAM (by the end of 2027) compared to HBM.
III. ChangXin's Current Situation: Strong Financial Performance, but HBM as a Weakness
ChangXin's financial results are impressive: revenue in the first quarter of 2026 was 50.8 billion yuan (up 719% year-on-year), with net profit at 24.7 billion yuan (up 1688%), and a global DRAM market share of 7.67% (fourth largest, highest in China). However, its HBM technology is a weakness:
1. Technological Gap: ChangXin's HBM3 is still in the sample verification stage, while Samsung and SK Hynix have already moved to HBM4. HBM4 requires different manufacturing processes from DRAM, and ChangXin, having focused on DRAM, lacks the necessary expertise and faces challenges with EUV equipment and packaging technologies.
2. Limited Funds for Investment: Of the 29.5 billion yuan raised, only 9 billion yuan is designated for "forward-looking technology" (including HBM development), with the remaining funds to be allocated to other areas. In contrast, SK Hynix plans to invest over one trillion won (about 50 billion yuan) in HBM, indicating that ChangXin's funding is insufficient for rapid progress.
IV. The Significance of ChangXin's IPO for China's Storage Industry
The success of ChangXin represents a significant milestone for China's storage industry:
1. Embracing the AI Era: AI is reshaping the storage industry, turning HBM from a commodity into a strategic asset. China needs companies like ChangXin to meet domestic AI demand and complete its electronics supply chain.
2. Reasonable Valuation: The IPO price of 8.66 yuan and a valuation of around 580 billion yuan do not overestimate future prospects, leaving room for growth for both the company and investors.
3. A New Beginning for China's Tech Industry: Despite technological gaps, China's chip industry has always overcome challenges. This IPO marks the beginning of a long-term narrative for China's tech sector, indicating that domestic storage solutions are officially entering global competition.
Conclusion
ChangXin's IPO is a pivotal event for China's storage industry, but it must navigate both cyclical fluctuations and technological gaps in HBM technology. Investors need patience, while the industry itself is stepping into the spotlight, marking its transition from a supporting role to a competitive force in the global AI storage market.