Digging Deeper: Conversations in Fundamental Investing

Episode Overview

• This episode explores the challenges and structural dynamics of executing deep value investing in a modern financial market dominated by momentum-driven, quantitative, and multi-manager strategies.
• The discussion highlights highly lucrative, structural arbitrage opportunities in the South Korean stock market, where corporate holding companies trade at massive discounts to their underlying technology assets.
• The guests dissect the true, nuanced impact of generative AI on business models, contrasting vulnerable asset-light digital marketplaces with high-potential asset-heavy operators and premium human data platforms.
• It provides a practical framework for identifying deep valuation anomalies, evaluating delayed customer acquisition flywheels, and navigating structural market shifts to capture asymmetric investment returns.

Key Concepts

• The Challenge of Value Investing in a Momentum-Driven Market: Modern market microstructure has shifted capital toward passive indexing, quantitative models, and multi-manager "pod" strategies that rely on price momentum for risk management. When an asset falls out of favor, systematic selling triggers a massive liquidity vacuum where there are virtually no fundamental buyers. For value investors, this means they can purchase significant volume while the stock continues to depreciate, resulting in high short-term residual volatility despite strong fundamental value.
• The "Korea Discount" and Conglomerate Arbitrage: South Korean equities historically trade at lower valuation multiples relative to global peers due to complex governance, cross-holdings, and a historical lack of shareholder-return focus among family-controlled conglomerates (chaebols). This creates stark mispricings where investors can buy parent holding companies or non-voting preferred shares at 40-50% discounts relative to the market value of their highly profitable underlying operating subsidiaries (e.g., Samsung Electronics or SK Hynix).
• Regulatory Catalysts and the "Value-Up" Program: To bridge the valuation gap of South Korean equities, the Korean government has introduced the "Value-Up" program, mirroring Japan’s highly successful corporate governance reforms. This initiative actively incentivizes historically cash-hoarding conglomerates to improve capital efficiency, adopt shareholder-friendly corporate governance protocols, increase dividend payouts, and initiate share buybacks.
• The "Asset-Light" vs. "Asset-Heavy" AI Paradox: Conventional market wisdom historically favored "asset-light" compounding businesses (such as online classifieds, directories, or basic matchmaking marketplaces). However, advanced agentic AI threatens to completely bypass and disintermediate these platforms, rendering their search and curation moats obsolete. Conversely, traditional, capital-intensive, "asset-heavy" businesses are structurally insulated from AI displacement; instead, they can integrate AI internally to drastically reduce operating costs and expand margins.
• The Premium on Authentic Human Data: As the internet becomes saturated with synthetic, AI-generated text and content, authentic human-to-human interaction data is becoming highly scarce and exponentially more valuable. Platforms like Reddit, which host massive repositories of organic human dialogue, are uniquely positioned to secure highly lucrative, recurring data licensing agreements with major Large Language Model (LLM) developers who require clean data to train their systems.
• Delayed Cohort Flywheels in E-Commerce: For businesses operating in industries with long customer replacement cycles (such as automotive retail), the financial benefits of exceptional customer experiences and word-of-mouth referrals are highly lagged. Because the average consumer purchases a vehicle only once every several years, current financial metrics and near-term revenue projections systematically understate the long-term compound growth of the company’s organic customer acquisition engine.

Quotes

• At 1:25 - "It is a lonely thing to be a deep value investor... If you just think about the world we live in right now, most of the volume and the capital has gone to quants, macro funds, and pods. Almost all of those strategies require momentum as a component." - This quote illustrates why traditional value investing struggles for near-term support in a modern market dominated by algorithmic, trend-following capital.
• At 10:44 - "Typically, a company will see referrals hit their revenue in the year in which the referral happened... But the average consumer in the US buys a used car every seven years. So 2026 revenue, the seeds were sown for that in 2021, 2022, and 2023." - This explains the highly lagged cohort effects in big-ticket e-commerce, showing why near-term metrics often understate a business's long-term organic customer acquisition engine.
• At 15:20 - "SK Square is the parent company to SK Hynix, and that’s a very clear setup whereby you can effectively buy SK Hynix at a 47% discount, and the holding company is actually promising to close that discount through buybacks and dividend payments. This is something that doesn't exist elsewhere in the world." - This details the unique and highly asymmetric structural arbitrage opportunities that exist within the South Korean conglomerate landscape.
• At 22:40 - "Everything that is perceived not to benefit from AI is in a factor basket that is getting destroyed... [But] the total posts and comments on Reddit have grown... There is a behavioral inflection that will occur in the back half of this year where the licensing agreements they have with Google and OpenAI will be repriced... our work suggests they could reprice at 5x or 6x." - This highlights how markets overlook true AI beneficiaries by failing to recognize the soaring premium on authentic, human-generated training data amidst a sea of synthetic web content.
• At 23:53 - "If you believe in a future of AI, then you really have to be watching out for service-level deflation... the best short ideas are actually high-quality compounding great businesses [like online classifieds]... Is it possible that agentic AI will be good enough to replace what we consider to be this amazing supply-demand marketplace moat? I think the answer is yes." - This introduces a forward-looking short framework, explaining how advanced AI agent systems can dismantle the pricing power and matchmaking moats of historically bulletproof, high-multiple marketplace businesses.

Takeaways

• Exploit Conglomerate and Holding Company Discounts: Generate alpha by identifying parent holding companies or preferred shares that trade at steep fractional discounts to the market value of their underlying public subsidiaries, particularly in jurisdictions like South Korea where government-led corporate governance initiatives are actively working to unlock this value.
• Audit Portfolio Holdings for AI Disintermediation Risk: Critically assess long-term holdings in "asset-light" compounding businesses (such as online directories, basic marketplaces, and customer service platforms) to evaluate whether emerging agentic AI can bypass their search and matchmaking moats, eroding their historical pricing power.
• Anticipate and Prepare for Liquidity Vacuums: When buying fundamentally undervalued assets that are out of favor, accept high near-term volatility and buy progressively, understanding that modern quantitative and multi-manager risk models will systematically dump these shares without immediate fundamental buyers to stabilize the price.