Mohnish Pabrai on Dhandho Investing, Checklists, and Korea

⬅️ Back to Videos

🎥 Mohnish Pabrai on Dhandho Investing, Checklists, and Korea

Knowledge Inside · Kim Kiho. Duration: ~54 min · Published June 8, 2026

Timestamps

  • 0:00 Introduction
  • 1:42 Lunch with Warren Buffett vs. Eric Schmidt; Charlie Munger
  • 11:46 South Korea demographics; SK Hynix, Samsung & Micron
  • 17:17 KOSPI
  • 17:55 FAA-inspired investment checklist; Buffett’s Dexter Shoes
  • 25:25 Three checklist essentials; IKEA and Amorepacific
  • 31:17 Active vs. passive; hunting risk-free investments
  • 35:08 Turkey; Reysas
  • 39:22 Dhandho: heads I win, tails I don’t lose much
  • 41:18 AI investing; don’t buy shiny items
  • 43:37 Building wealth; Rule of 72; the 1623 Manhattan deal
  • 50:22 Giving back; Dakshana Foundation
  • 53:50 Advice to listeners

Pabrai manages about $1.4 billion and gives long interviews when the venue fits. This hour with Kim Kiho on Knowledge Inside walks through how he actually invests: hold businesses forever, run a crash-driven checklist, and buy what is hated while everyone chases the next shiny object.

  1. The 2007 Buffett charity lunch cost $650,100, well under the $2 million cap Pabrai set as a tuition bill on the $70 million he had already made applying Buffett’s ideas. He co-bid with Guy Spier; his only goal was to thank Buffett. Buffett read everyone’s bio beforehand and set no time limit. Pabrai rates that lunch 15 out of 10. A separate Eric Schmidt lunch at Google’s cafeteria scored 2 out of 10: Schmidt checked his watch every 15 minutes and left at the one-hour mark. When Pabrai’s wife said she preferred Charlie Munger, Buffett arranged a Munger lunch that Pabrai enjoyed more, and a lasting friendship followed with home visits every three or four months.

  2. On Korea, Pabrai praises the country but worries about population decline faster than Japan’s. Shrinking headcount caps total output unless you are an export powerhouse, and even then you face tariffs and rising labor costs. Hyundai is cheaper to build in Alabama than in Korea, he says. He currently holds no Korean stocks. On memory semis, he regrets selling SK Hynix and Micron when he should have held forever. Samsung, SK Hynix, and Micron sit in a protected oligopoly where a fourth entrant would need patents, engineers, and 10 to 20 years of fab work. If you already own them, do not sell. The party has just started, he says. If you do not own them, do not buy now.

  3. KOSPI is lopsided because Samsung and SK Hynix dominate the index. The rest of the market still looks somewhat undervalued to him, but the population headwind is real. Domestic, non-export businesses face the hardest path. Export champions get tailwinds with their own complications.

  4. His 213-question checklist copies the FAA’s crash-first logic. A plane crash triggers investigation; only then does aviation change. In investing, a crash is a stock that lost you money or went to zero. He built the list from great investors’ visible mistakes. Buffett’s Dexter Shoes, wiped out by cheap foreign labor, spawned a question about foreign-competition risk on every new idea. He started 16 or 17 years ago with 70 to 80 questions; now he runs all 213 before any purchase, usually in one or two hours once the research is done. He makes two or three new investments a year and cites the Korean Air peanut girl episode as what happens when you skip the pre-flight list.

  5. For retail investors he boils the checklist to three items. No leverage in the company or your margin account: IKEA’s founder ran 70 years without borrowing a single won or euro. Moat durability: memory chips went from a terrible business with 20 competitors to a great one with three; Amorepacific is a hard business with a shallow moat despite loyal customers. Governance: owners should love the business, not love the money.

  6. More than 99% of investors should buy an index fund, Pabrai says. Active stock-picking only makes sense if you can forecast a company’s cash flows 10 to 20 years out with high certainty. He uses SK Hynix as the test case. The deals he wants look almost risk-free: a regulated Korean power company earning $100 million a year, bought for $300 million, paying out 100% of earnings, pays back your capital in three years while you still own the asset for decades. Those anomalies live in hated, unloved sectors, not in whatever the crowd loves this year.

  7. About 70% of his portfolio sits in Turkey, a market he entered seven years ago because it screened as the world’s cheapest. Reysas, a warehouse landlord to Amazon, IKEA, Mercedes, and others, traded at a $16 million market cap against roughly $800 million of liquidation value, about 2% of asset value. After due diligence he built a large stake and now owns about 40%. Seven years later the market cap is about $1.5 billion and liquidation value about $2.5 billion, and he intends to hold as long as the founding family runs it.

  8. Dhandho, for him, starts with one question: how can this investment lose money? On Reysas the answer had to be it can’t: no debt, Fortune 500 tenants on long leases, earthquake standards and insurance checked after Turkish quakes hit with no warehouse damage. Applied to AI, the framework says stay away. AI is deeply loved, and loved is the wrong side. Bitcoin was loved, then AI replaced it as the shiny object; next, he predicts, the crowd will rotate into the SpaceX IPO. Please don’t buy shiny objects, he tells the audience.

  9. Asked what to do with $1,000, he names Berkshire Hathaway (BRK.B): lots of cash, no leverage, strong management, deep moat, boring, hated, and unloved. Building wealth from scratch is two moves: spend less than you earn, and put the surplus into something equally boring, Berkshire or an index, and keep adding.

  10. He tells the 1623 Manhattan story to teach Rule of 72. Native Americans sold the island for $23; at 7% compounded, money doubles every 10 years (72 divided by 7), and over four centuries that $23 would grow to roughly $28 trillion, far more than Manhattan’s land is worth today. The lesson: the Indians were not cheated; they needed a patient investment officer. Compounding plus savings beats chasing excitement.

  11. He still invests because he treats it as a game, bridge and chess, not team sports. He asked Google’s AI when he will die and got June 11, 2054; he wants $10,000 left on June 10 and everything else given away through the Dakshana Foundation, which educates poor, gifted kids in India. One engine compounds; the other gives, and giving must accelerate as wealth grows. He closes with a German proverb: if wealth is lost, nothing is lost; if health is lost, something is lost; if character is lost, everything is lost.

Related TMFNK Content

Crepi il lupo! 🐺