What Mohnish Pabrai learned in 30 years: don't sell, clone and buy the moat cheap

Mohnish Pabrai started with a single idea thirty years ago: copy what works, don't reinvent it. In a long Texas conversation he traces a practice that began with $1 million and now steers $1.4 billion, opening with his most candid regret — selling a roughly $6 million 1% stake in Ferrari that today would be worth close to a billion. The spine is clear: cloning is a strategy, patience is a technology, and the quietest return driver is spending discipline. From the lesson of never selling Walmart to the Nifty Fifty, from a Turkish warehouse to the American Express moat, the talk keeps returning to one question — how does an ordinary saver find one great business at a reasonable price and keep it?

The quiet math of saving: spending, index and lost decades

Pabrai's first lesson starts with behavior before knowledge: spend less than you earn, put the difference in an index, start early. Compounding feeds on the asymmetry of small, early differences stretched over decades; continuity, not heroic sacrifice, builds wealth. The S&P 500 has averaged more than 9% a year for a century, but the path is not a straight line. According to ChartRow , the annual returns table since 1928 shows that average punctuated by long stalls, and the video's claim of a flat 1965-1982 stretch checks out against the same dataset — even as the economy and profits grew, returns went nowhere when starting valuations were high.

That cyclicality frames the video's boldest call: 17 years of roughly zero from 1965 to 1982, about a 15-fold rise from 1982 to 1999, and near zero again from 1999 to about 2010. Pabrai argues we stand at a similar threshold in 2026 and, on valuation alone, expects minus 2% to plus 2% a year for the S&P 500 over the next decade, preferring Berkshire Hathaway as a kindergarten-grade base carrier until around 2035 when the index may be ready for its next run. ChartRow's long series confirms these lost decades are not exceptions but pattern — patience is rewarded when valuations are stretched, courage when they are depressed.

The second filter is Pabrai's 213-item checklist. Most questions fall into three buckets: leverage , moat , and manager-owner quality. He illustrates the moat with a sushi shop in Austin — success invites copycats, and without a moat competition eats profit. In autos, he puts Hyundai in the too-hard pile as electrification erodes Toyota's old edge and capital intensity forces constant reinvestment. By contrast, Mastercard, Moody's or S&P Global have visible moats but are usually fully priced; the task is to spot greatness before it is universally recognized as great.

That search took Pabrai to Turkey about eight years ago. On a second visit in 2019, a deep-value friend brought him to one of the country's largest warehouse operators — 99% leased to Amazon, IKEA, Carrefour, Mercedes and Toyota. Market capitalization was $15 million; Pabrai appraised the assets near $800 million and bought a third of the company for $8 million. Today the market cap is near $1.8 billion, more than 100 times in seven years. The story shows how wide the gap between screen price and asset value can become when fear is loudest, and how on-the-ground verification — walking the warehouses, reading the tenant roster — separates a value trap from hidden quality.

To explain bubbles he revisits the Nifty Fifty. According to Wikipedia , in the late 1960s and early 1970s the fifty best American companies — IBM, Xerox, Polaroid, Kodak and peers — were bundled as one portfolio with 2% each; as popularity exploded multiples stretched to 50-100 times earnings and in the slow-motion crash of 1973-1974 the market fell about 50% while the Nifty Fifty fell close to 70%. Pabrai's thought experiment starts there: hold the Nifty Fifty untouched from 1970 to 2026, assume 49 names go to zero and only a 2% Walmart survives as it did — the whole portfolio still compounds near 15% a year, beating the S&P 500 by four to five points. The lesson is that index returns come from very few outsized winners and not selling the few before permanent decline is the real job.

From moat to notebook: the retail investor's compass

The Walmart case is clinched by the family. After the 1971-1972 IPO the Waltons owned about 44% of the company; through buybacks and a live-on-dividends model that stake sits near 46% some 54-55 years later, effectively never sold. According to Business Insider (businessinsider.com) , Rob, Jim and Alice Walton were worth a combined $228 billion in 2024, making the family the second wealthiest after Elon Musk, largely on the same core holding never sold. In Pabrai's words, until a business shows clear evidence of permanent decline, the price — expensive or cheap — is secondary; the same wait applies if you own Google or Apple.

The financial counterpart to patience is leaving leverage out. Pabrai recalls IKEA founder Ingvar Kamprad designing the company to last 500 years, never borrowing a dollar and building each store from retained earnings. According to The Guardian (theguardian.com) , the obituary published on his death in 2018 frames that thrift and debt-free growth as the cultural core of IKEA. In Pabrai's checklist leverage also dominates — it amplifies good times and makes bad times fatal, forcing irreversible sales.

For the retail investor he proposes a two-stage curriculum. Kindergarten through third grade is a single lesson: buy Berkshire Hathaway. Once the alphabet is learned, stage two begins — study only businesses whose products you personally use. It is a circle of competence from the fridge: if a company can regularly extract a dollar from choosy you, it likely has a strong value proposition. Write down the Apple phone you keep buying, the Starbucks or Burger King you visit, the Coca-Cola you drink. If valuation looks absurdly cheap — say a trailing P/E of 3 versus 17 — and still looks cheap after deep work, Pabrai entertains a 10% in that franchise and 90% Berkshire mix.

The Burger King vignette makes the sensory edge concrete. Citing two Wall Street Journal pieces, he notes the Whopper's bun, mayonnaise and packaging were changed, quarterly sales rose 10% and the taste difference is felt; no investment decision yet, but a consumer may spot a shift the market has not priced. The same lens applies beyond burgers — a cosmetics formula that visibly reduces wrinkles and is noticed first by its user can be an early alpha signal before it shows up in numbers.

In AI the same filter applies. Pabrai advises staying away from crowded hype, looking at most at memory or a Nvidia-like pick-and-shovel layer in a limited way, yet reminding that the better opportunity is often off-consensus, like the Turkish warehouse operator. It is selectivity, not laziness — where everyone is excited valuation is stretched; where a quiet product improvement is noticed only by its user, mispricing lives. The heart of the work is therefore reading product experience correctly and waiting for price.

The most vivid moat story is American Express. When a large lending loss crushed the stock in 1964, Buffett stood by cash registers in fine Omaha restaurants and watched them accept the card without hesitation; if trust was intact the decline was temporary, so he bet 40% of his fund, later selling when normalized, then re-buying about thirty years later after a golf conversation with the CEO of Hertz — Hertz paid 3% to American Express versus 2% to Visa/Mastercard yet could not stop accepting it, saying it would lose Avis-bound customers and would pay even 4%. According to Trustnet , Buffett's 1964 American Express investment remains the textbook case for a brand-trust moat, while according to CNBC Munger's recent warnings on leverage remind that even such moats turn fragile when combined with debt. In the late 1980s and early 1990s Boston restaurants tried a joint boycott and the company did not blink; its cardholders earn more and spend more, the boycott dissolved, and the ability to charge the highest annual fee and earn return on equity above 30% confirmed pricing power.

He repeats the same moat read on Apple. Friends who own private jets saying they would give up the jet before the iPhone, and grandchildren refusing to switch to Android, were for Buffett field evidence of brand attachment. According to SCMP , Duan Yongping — founder of Oppo and Vivo — holding Apple as his largest position despite being a direct competitor is framed as an exercise in humility, acknowledging that Apple's software and brand feel cannot be replicated. That the Forbes 400 was built in each case by one business, and that letting great businesses sit until they dominate the portfolio is fine, ties these stories together: find one exceptional business at a reasonable price, then do nothing for a very long time.