AI's "Money-Making Magic" in the First Half: Hardware Reigns Supreme, High Dividends for Defense! New Logic for Global Allocation
$EFUND A SEMICON ETF(03486.HK) $EFUND APAC HD(03483.HK) $Nokia Oyj(NOK.US) $Taiwan Semiconductor(TSM.US) $TENCENT(00700.HK) $XIAOMI-W(01810.HK)
Good afternoon, everyone! Today, our "Bùduàn Talks ETF Special" has invited two heavyweight guests—Simon from E Fund Hong Kong and Kimmy from Everbright Securities International.
Before we sat down to start the livestream, everyone was feeling a bit anxious about Hong Kong stocks, thinking that while this round of global AI hype (especially in the hardware sector) is shining brightly, Hong Kong stocks seem to be lagging behind. But if you broaden your perspective to global allocation, you'll find a completely different landscape! Kimmy mentioned that the South Korean stock market (KOSPI index) in the Asia-Pacific region alone has already achieved a year-to-date return exceeding 70% of last year's full-year performance!
So, how should we view and invest in this astonishing AI supercycle? Below, we've summarized the "core insights" from our two guests:
💡 How far can the AI rally go? Two key indicators
Simon shared E Fund Hong Kong's core view: For the AI rally to sustain, two conditions must be met:
1. AI must genuinely be helpful: Making life and work more comfortable, enabling what was previously impossible (e.g., AIGC, Sora video models).
2. Companies must make money: Stock prices, in the long run, depend on profitability.
The current situation is: Upstream companies making GPUs, CPUs, optical modules, and storage (e.g., NVIDIA, SK Hynix, Chinese optical module manufacturers) have all reported earnings far exceeding expectations, reaping huge profits; while downstream cloud providers and software companies, even after making money, must continue to invest heavily to expand production. That's why capital is now relentlessly crowding into "hardware"!
🛠️ Stock Selection Logic: Finding the "Pearls of the Human World"
Many ask: Will the semiconductor pullback after the surge be deep? After hardware fabs are built in two years, will there be overcapacity?
Simon used a vivid analogy: "Now the whole world wants to build houses (develop AI), so everyone is scrambling for steel and cement. In 1.5 to 2 years, supply will catch up, but will demand then skyrocket to the heavens?" For example, video calls took over a decade to become widespread due to network and bandwidth limitations; when computing power costs (Token) come down in the future, new demands like AI producing TV series and interacting with virtual stars will be created.
Therefore, investing in the AI era should follow the logic of "the strong get stronger, winners take all," seeking the most formidable "pearls" in each country or industry, and avoiding small-cap stocks easily eliminated in internal competition:
🌍 The Core "Pearls" of the Global AI Industry:
• Chip & Computing Power Master: NVIDIA NVDA
• Advanced Storage (HBM) King: SK Hynix 000660.KS
• Advanced Chip Foundry: TSMC TSM / 2330.TW
• Core Lithography Machine Monopoly: ASML
• Traditional Chip Giant Revival: Intel INTC
• Full-Stack Integration Leader: Google/Alphabet GOOGL (Both Simon and Kimmy are optimistic about its full-stack integration capabilities from chips (TPU), large models (Gemini) to applications)
• Dark Horse Potential Stocks: Nokia NOK, BlackBerry BB (Kimmy pointed out their layout in optical communication/fiber-related businesses, worth noting for transformation opportunities)
💡 Extended Thinking: Where is the future super hardware? Simon mentioned that AI has huge potential for integration with three major tech hardware areas: AI + autonomous driving (e.g., Tesla TSLA's FSD layout), AI + drones, AI + robots. ToC application software companies (e.g., Tencent 0700.HK, Alibaba 9988.HK), while still searching for the "indispensable" Super App, possess strong ToC genes and remain significant weights not to be ignored.
🎯 Two Recommended Allocation ETFs: One Offensive, One Defensive
If you don't want to bother picking individual stocks or worry about buying the wrong one and getting "stuck at the peak," our two guests suggest using ETFs for global/Asia-Pacific cross-market allocation, balancing offense and defense.
1. Offensive 🚀 E Fund Asia Semiconductor ETF (03486.HK)
• Features: This semiconductor ETF carefully selects the most competitive transistor industry chain companies from Hong Kong, Mainland China, the US, South Korea, Japan, etc.
• Component Highlights: Accurately captures recent high-flyers like SK Hynix, TSMC, Samsung, as well as Mainland Chinese semiconductor leaders with massive mature process capacity. It also includes giants like Tencent (0700.HK) and Xiaomi (1810.HK) with forward-looking layouts in AI Agents and smart hardware. Its performance clearly outpaces traditional blue-chip indices.
2. Defensive 🛡️ E Fund Asia Pacific High Dividend ETF (03483.HK)
• Features: If you're afraid of AI's high valuations, this is the perfect defensive choice. It bundles high-dividend companies from Hong Kong, Japan, and Australia together.
• Why include Australia? Simon pointed out that Australia is far from geopolitical conflicts, and its mining industry (e.g., iron ore) is highly monopolistic and doesn't blindly overexpand in internal competition, generously distributing dividends when profits are made! Kimmy added that Australia has already raised interest rates three times this year due to strong economic resilience, and the Australian dollar is performing strongly. With the support of Australian and Japanese stocks, this ETF can still "continuously hit new highs" even when Hong Kong stocks are volatile.
• Holding Suggestion: Buying the wrong AI stock might result in a 50% loss in 3 years, but high-dividend products are suitable for Buy and Hold, holding for 3-5 years to earn steady returns.
💬 Conclusion
The current AI wave is undoubtedly still in its first half. The ideal investment strategy is to hold both the strongest AI (offense) and the strongest high dividends (defense) simultaneously, for example, allocating them in a 60-40 ratio.
The copyright of this article belongs to the original author/organization.
The views expressed herein are solely those of the author and do not reflect the stance of the platform. The content is intended for investment reference purposes only and shall not be considered as investment advice. Please contact us if you have any questions or suggestions regarding the content services provided by the platform.
