The AI Chip Bubble: Why South Korea & Taiwan Are In the Danger Zone | Michael Fritzell | Asian Century Stocks

Episode Summary

Executive Summary: The episode argues that Asia’s returns are being driven by sharp thematic dispersion: AI and memory-chip speculation in Korea and Taiwan, governance reforms in Japan and Korea, and deep value opportunities in Southeast Asia. Michael Fritzel says South Korea looks bubble-like in memory chips, but still sees compelling long-term value in small caps, especially those with strong governance and insider buying. China remains unattractive due to policy risk, property weakness, and capital repression.

Main Topics: AI-driven dispersion and market speculation in Asia (Priority: 5/5): The discussion opens with how AI has overwhelmed Asian equity markets, lifting Korea and Taiwan while leaving China and India behind. The guests compare current enthusiasm in semiconductors and related supply chains to past bubbles, especially in South Korea. South Korea memory chips: bubble risk versus cyclical opportunity (Priority: 5/5): Fritzel argues that SK Hynix and Samsung’s earnings expectations look extreme and speculative, especially given likely future supply from Chinese competitors. He distinguishes commodity memory from more defensible logic-chip businesses. Japan’s bull market and governance reform (Priority: 4/5): Japan’s rally is attributed to a weak yen, export strength, and corporate governance reforms that pressure firms to improve ROE, reduce cross-shareholdings, and increase buybacks. Fritzel thinks Japan has largely run its course. Korea’s reform agenda and small-cap value (Priority: 5/5): He sees Korea as increasingly attractive because of inheritance-tax changes, lower dividend tax burdens, stronger shareholder protections, and rising activism. He argues value has shifted from large-cap tech to small-cap domestic businesses. China’s policy risk and structural slowdown (Priority: 5/5): China is described as a market to avoid because of property-sector collapse, US tariff pressure, and repeated state interventions that undermine private-sector confidence. The party’s control over business is presented as a central investment risk. Southeast Asia as a value and governance opportunity (Priority: 4/5): Thailand, the Philippines, Singapore, and parts of Malaysia are portrayed as cheap relative to the rest of Asia, with some countries benefiting from reforms and supply-chain shifts. Indonesia is viewed as the weakest governance story. Consumer/culture exports and stock-picking ideas (Priority: 3/5): The conversation explores Korean beauty, noodles, liquor, and entertainment spillovers from K-pop/K-drama, but notes that many of the best ideas are not the obvious consumer winners. Stock selection is emphasized over broad thematic exposure.

Key Arguments: Asian equity dispersion is unusually wide because country cycles are not synchronized and the AI theme is concentrating speculation in a few hardware-linked markets. South Korea’s large memory-chip names look bubble-like because forward profit expectations imply absurdly high profits for commodity producers by 2028. Commodity DRAM is easier to model and more likely to face a supply response from China than advanced logic chips like TSMC’s foundry business. Japan’s market strength is real but increasingly mature; weak-yen tailwinds and governance reforms have already been priced in. Korea is more attractive in small caps than in the headline chip names because reforms, insider buying, and cheap valuations remain underappreciated. China is structurally harder to own because the state can and does intervene unpredictably across sectors, which weakens the investability of private companies. Southeast Asia offers pockets of value, but catalyst quality varies widely by country, with Thailand and the Philippines better than Indonesia. For long-term investors, cheap valuations and improving governance matter more than popularity or theme-chasing. Western investors often only buy the most obvious names, such as TSMC or Samsung, leaving many lower-profile Asian opportunities underowned. Cultural exports like K-pop can help consumer brands overseas, but many of the best investments are still in enabling businesses, not the visible brands themselves.

Data Points: iShares MSCI Korea ETF 1-year return: 255% - Used to illustrate the extraordinary surge in Korean equities, especially memory-chip heavy indices. Taiwan market 1-year return: 112% - Compared with Korea to show broad AI-related strength in Taiwan as well. SK Hynix and Samsung Electronics forward net profit estimates for 2028: ~$200B and ~$250B - Cited as evidence of bubble-like expectations for commodity memory producers. South Korea index weight in memory chips: 50-60% - Speaker says the KOSPI has become heavily concentrated in memory chip stocks. Japan yen exchange rate: ~160 JPY/USD - Weak yen cited as a major driver of Japanese stock performance. Korean inheritance tax: 50% - Identified as a structural reason families historically suppressed share prices. Korean dividend tax: Reduced for high-dividend payers - Presented as a reform that improves shareholder returns and valuation support. Korean governance liability change: Since 2025 - Directors now can be personally liable for related-party transactions harming minorities. China residential new starts: Down 70%+ since 2021 - Used to show how severe the property downturn has been. Japanese SaaS valuations: 2-3x EV/sales - Example of domestic Japanese growth stocks trading at low multiples. Korean credit score provider valuation: ~9-10x P/E - NICE Information Service cited as a cheap, stable compounder. Dividend yield for NICE Information Service: ~5-6% - Presented as part of the stock’s appeal. FNGuide valuation at sale: ~11x P/E to ~30x P/E - Example of a stock Fritzel sold too early after a rerating. Toa Paint valuation: ~8x P/E - Used as an example of cheap Thai exposure. General Korea/Asia market comparison: EM index is ~50% Korea/Taiwan - Illustrates how concentrated emerging-market benchmarks have become.

Pivotal Quotes: "I think it is because specifically South Korea, I do think it is a bubble because the estimates for forward profits for SK Hynix and Samsung Electronics, we're now talking 200 billion US dollars in net profit estimates for 2028." — Michael Fritzel: On why the Korea memory-chip rally looks speculative and potentially overheated. "If you can track inventories and track prices daily, that means they're standardized enough that... we have three or five companies competing." — Michael Fritzel: Explaining why commodity DRAM is more vulnerable to cyclical supply responses than advanced semiconductors. "If you're a stock picker, I feel like it's time to move on from Japan and look towards Korea." — Michael Fritzel: Summing up his view that the most attractive opportunities have shifted from Japan to Korea.

Implications: Listeners should distinguish between momentum-driven index rallies and durable stock-picking opportunities. The biggest risks are crowded AI trades, policy intervention in China, and late-cycle Japan; the biggest opportunities may be in Korean and Southeast Asian small caps where governance and valuation are improving.