In one sentence: Phil explains why he sold Alibaba at a loss: the risk was not Beijing but a US response to a Taiwan conflict that could wipe out a VIE-structured holding overnight, and he argues the right response to a changed story is to leave now, not hope.
Key ideas
- What happened. After a team trip and heavy research on Alibaba and JD.com, Phil bought. Then he sold at a loss. [01:00–03:00]
- The risk he studied, and the one he missed. He studied whether China would stay capitalist, and decided it likely would. The missed risk was the US reaction to a possible conflict over Taiwan, shown by Russian holdings going to zero after Ukraine. [03:00–10:00]
- VIEs explained. Chinese rules block foreigners from owning many companies directly, so US-listed names like Alibaba or Tencent are a Cayman-based shell with contractual claims on profits, not shares in the operating company. The SEC and China were also arguing about audits. A compromise was reported that week. [05:00–09:00]
- A new inversion, then act. Danielle observes this is textbook: the story changed, so he rewrote his inversion and sold. Phil: when the story changes, the investment must change, and promptly. [10:00–15:00]
- Textbook is easy to know, hard to feel. Even if a decision is clear, carrying it out is the hard bit; such losses come, for Phil, perhaps a handful of times in 40 years. [11:30–13:00]
- The hope fallacy. Staying in because the price might rebound is a "strategy" that works sometimes, which reinforces bad habits, and the failures can go to zero (Phil cites a past stock that never recovered). [13:00–18:30]
- A student's options story. A trade that should have lost at most $3,000 grew toward a total loss; he then multiplied his risk repeatedly, reportedly using his daughter's college money. The market recovered and he posted about it. Phil cites a study that most long-option traders lose, and says first losses are best. This is a hearsay anecdote, not a verified case. [18:00–22:30]
- You needn't earn it back in the same stock. Guy Spier's point: look for something better elsewhere. [22:30–24:00]
- The real risk is ourselves. Munger's point: know what you don't know; Phil admits he didn't know what he didn't know about the US response. [24:00–25:00]
- Buffett sells and moves on. Danielle recalls Buffett saying he sold a holding he had owned briefly after learning something new, and the airlines are the example. Phil adds that Buffett would hope for Coca-Cola to fall 50% so he could buy more. Phil's two lessons: do better work, and when it's time to get out, get out. [25:00–32:00]
How it maps to RuleOne
- A holding's written story is what the agent stack should compare against when news arrives; a changed story means review now, not at the next quarterly check.
- /holdings/ is where an exit is recorded; the loss is part of the track record.
- ADRs and VIE structures are a ownership-rights risk the stock page can't show; check the 10-K or 20-F.
Buffett, Munger and Graham links
- Buffett's 2020 airline sales and his 2022 meeting comments on small losses are the likely episodes referred to; Danielle was unsure of the name, so check the letters.
- Munger's "know what you don't know" is the circle of competence; see 001.
- Graham's margin of safety assumes you can be wrong, which is what this loss illustrates (The Intelligent Investor, ch. 20).
Words to know
- VIE (variable interest entity): an offshore shell used to give foreign investors economic exposure to a Chinese company they can't own directly.
- Government (legislative) risk: a law or executive act that changes or ends a business's value.
- Permanent loss of capital: a loss you lock in by selling, or that never recovers.
Try this
Pick one holding or watchlist stock at /stock/TICKER/ and write down its story in two sentences, plus the one fact that would change it. Decide now what you would do if it appeared.
Check yourself
- What risk did Phil miss with Alibaba?
Answer
The US government's reaction to a China-Taiwan conflict, which could block or zero US-listed Chinese holdings. - What is a VIE?
Answer
An offshore structure through which US investors get contractual economic rights, not direct ownership of the Chinese operating company. - Why is "hold and hope" dangerous even if it sometimes works?
Answer
It rewards bad behaviour, and the failures can be total. - Where can you make a loss back?
Answer
Anywhere; not necessarily in the stock where you lost it.
Short quotes
"When the story changes, the investment must change, and do it now." (Phil, ~14:00, auto-transcribed)