How to Configure Computers in the AI Era? Long-term Potential + Economic Cycles + Competitive Landscape Provide the Answer [Master Class with Wang Zijing 7.2]

```

Content of This Issue

Hello, Wallstreetcn users and friends. I am Wang Zijing, Chief Analyst of the Computer Industry at Dongwu Securities. Welcome to my master class, "China-U.S. AI Investment Paradigms."

Computer Industry Track Research: The Underlying Logic of Long-term Potential, Prosperity, and Competitive Structure

For research on the computer industry track, we mainly conduct in-depth analysis from three dimensions: long-term potential, track prosperity, and competitive structure. Among these, the assessment of track prosperity can be characterized from the perspectives of technological cycle, policy drivers, and the spending cycle of downstream clients.

Long-term Potential: The Decisive Factor for Valuation Anchoring

First, we focus on long-term potential. The computer industry is a highly elastic field, and its market potential directly determines the potential for upward valuation. Its valuation methods are unique—current valuation approaches are often difficult to apply, and current performance can hardly directly explain a company's stock price. The valuation foundation is usually projected to a forward valuation three or five years out. Therefore, the industry ceiling determines future heights, while industry scale ensures sustained high growth for computer companies. The core factors of PE fluctuations or valuation improvement include growth potential and marginal changes. Marginal changes determine short-term direction, while growth potential sets the long-term ceiling. The core logic of PE uplift for computer companies is precisely a structural shift in the long-term potential of the track in which the company operates. Take cloud computing as an example. It's not just a technical form; its core value lies in significantly expanding long-term potential. For example, the subscription model after migrating to the cloud lowers the payment threshold, effectively reducing the number of pirated users and encouraging more customers to become paying users. In addition, the platformization of cloud computing continues to expand the boundaries of traditional software products and, by providing value-added services, creates new incremental markets. Taking the ERP industry cloudification as an example, its business scope has expanded from a single enterprise back-end management product to middle-office PaaS, and has extended further to front-office businesses such as marketing, customer relationship management, and supplier management. These businesses were not traditionally within the narrow ERP scope but have become part of company core operations after cloudification. Middle-office PaaS has built a cloud ecosystem, making it easier for third-parties to profit-share on the platform. These factors together have increased both the number of customers and the value per customer for ERP vendors. Meanwhile, the subscription-based payment model also raises customer lifetime value, as annual payments ensure a monotonically increasing cumulative amount from each customer. It is precisely this growth in long-term potential brought by cloud computing that has lifted company valuations.

Track Prosperity Cycle: Resonance Effect of Three Superimposed Cycles

For the computer industry, prosperity cycles do not stem from a single factor but are the superposition and resonance of technology cycles, policy cycles, and downstream prosperity cycles. Let’s explain these one by one. Technology Cycle and Investment Window. This is a classic life cycle—emergence, growth, maturity, and decline stages. For computer industry investment, the best entry point is the growth stage. Looking back at 2014, Microsoft’s move to the cloud from 0 to 1 remains a classic case of several dozen-fold return. Investing during the emergence stage means the company is still in heavy R&D, and it’s hard to forecast return—opportunity costs are high, and few new technologies deliver economic value in the end. Investing at maturity faces valuation decline and a lack of growth potential—not much opportunity there. For computer companies in the growth stage with rising penetration, market share is more important than revenue and profit—we pay particular attention to market share and the competitive landscape. Computer companies have extremely low marginal expansion costs (when a user buys a product, the company’s marginal cost is almost zero), but once users are on board, switching software carries high cost and risk—customer stickiness is strong. Once the market share is lost, it’s very hard to regain. This is especially true in foundational software, cloud SaaS, and other heavyweight application domains. Therefore, in the early stages of the industry, companies generally prioritize capturing market share over pursuing revenue or profit first. Policy Drivers: The Key Force Determining Major Market Trends. The logic driving the computer industry isn't "speculating on performance," but rather "event-driven." What matters is not absolute performance but the marginal change. Policy and technology are the most important drivers for surges in the computer industry. Historically, all major market trends in this sector have been due to policy or technology. Compared to technology, policy-driven marginal change is more direct, because most downstream clients in this industry are semi-governmental sectors, which are highly sensitive to policy changes; their IT demands are also easily affected by policy shifts. Looking back over the past decade: in 2014, the Snowden incident sparked China’s cybersecurity market; in 2015, "Internet Plus" was included in the government work report, leading to a super bull market; from 2018–2019, the trade war gave opportunities for self-controllable (Xinchuang) initiatives; at the end of 2022, the "20 Data Measures" fueled the data element concept; last year’s focus was on the low-altitude economy. Major trends in computers are inseparable from policy boosts, and since most downstream clients are central state-owned enterprises or government, the importance of policy-driven forces must not be underestimated. Many investors wonder: there are many catalysts in the computer industry, so why do some trends fade quickly, while others last? How can we judge the strength of policy drivers? We divide policies into four levels, corresponding to the scale of the trend: top-level policies can be repeatedly traded on, bringing super-large opportunities—for example, Xinchuang (IT application innovation), low-altitude economy, and other “number-one projects.” The second level is the General Offices of the Party Central Committee and the State Council (Two Offices), which typically support a trend lasting more than half a year. The third level is ministries and commissions, which usually drive trends lasting half a month, e.g. previous trends led by the Ministry of Industry and Information Technology or the Ministry of Transport around connected vehicles. The last level is local policies, which can nearly be ignored. Therefore, the very level of a policy determines the level of the market trend it brings. Technology Catalysts: Random but Disruptive. This force, aside from policy, acts as a technological catalyst. It is rare and unpredictable—often abrupt. Major historic technological inflection points, such as cloud computing, indeed brought booming times for the computer industry in China, fueling the rise of the "Five Flowers" (key companies). The recent case of artificial intelligence was not entirely policy-driven but the outcome of global industrial resonance. No one foresaw ChatGPT’s sudden emergence at the end of 2023; the market underappreciated it at first, and it only truly took off two or three months later. In sum, big opportunities in the computer industry almost always come from policy or major technological change, with policy effects more direct. Downstream Client Spending Cycle: Fiscal Sensitivity. Downstream clients’ spending directly determines computer company revenues, as many rely on clients from semi-governmental sectors. This means the computer industry is highly sensitive to fiscal spending, and industry slumps usually relate to tight fiscal budgets. In IT informatization, high downstream prosperity boosts IT investments; conversely, as the sector cools, IT budgets are the first to be cut. Therefore, when researching informatization companies, it is critical to pay attention to downstream client prosperity: no matter how good the products or services, if clients have no budget, companies will struggle to profit.

Competitive Structure: Oligopoly Is the Industry's Long-term Fate

Finally, let's analyze the industry's competitive structure. More often than not, the computer industry ends up forming a single or dual oligopoly, and the dual oligopoly is frequently the product of administrative intervention. Usually, a single oligarch prevails; enterprises after the first and second find it very difficult to survive. We can verify this by observing the list of top computer companies in China and abroad—compared to a decade or even twenty years ago, there have been few structural changes. For example: PC operating systems are now a duopoly (Windows and macOS); mobile operating systems are a duopoly (iOS and Android); CPU is a duopoly (Intel and AMD); GPUs are now a duopoly (Nvidia and AMD). Nearly all segments, without exception, are monopolies or duopolies. This returns to our initial argument: As long as an IT story in information technology is highly realizable, it is a story that can change the world, precisely because realizing it is extremely hard. Innovation in IT often means going head-to-head with U.S. tech giants—a challenge that is on another level compared to what Chinese manufacturing faces. Therefore, it’s not that domestic companies aren’t working hard, but the ones "above" have deep-rooted foundations (mostly American firms). The computer industry attaches great emphasis to historical accumulation, technology and ecosystem, first-mover advantage, and user stickiness. Its characteristic is that "the more it’s used, the better it works; the less it’s used, the worse." The Matthew effect is especially strong; it is extremely difficult for small firms to break through. Thus, within any given segment, the destinies of different companies may be worlds apart. In addition, there is another kind of competition structure in the computer industry: fully competitive sub-sectors. In China, some areas exhibit this, such as cybersecurity, medical IT, and financial IT. Their features include high product and service homogeneity, so their competitiveness mainly comes from the ability to win orders, i.e. channel capability. The industry concentration is low and there are often regional/sectoral advantages. For these sub-sectors, we focus on trends for the whole sector since they tend to rise and fall together—so look at the sector beta, especially during policy drivers and improved downstream prosperity. Also, continuously observe whether a fully competitive segment is evolving toward oligopoly. That's all for this lesson. Feel free to leave comments and discuss, see you next time. Risk Disclaimer: Master classes select third-party compliant institutions’ professionals to teach investment research theory courses on this platform. The content does not constitute any buy/sell or investment advice for specific products. The opinions expressed in platform courses are for study and reference only, do not represent Wallstreetcn’s views, and do not address users’ special investment goals, financial situation, or needs. The market is volatile and uncertain, and the platform is not liable for losses resulting from relying on course viewpoints or information. Investments are risky; please decide with caution. ```