He bought insurance companies in order to invest their customers' unpaid premiums — money he holds but does not own — as permanent leverage that cost him less than nothing, then ran it through a concentrated portfolio for sixty years inside a vehicle that never paid a dividend and never sold.
- Started with
- A father who owned a stock brokerage and sat in the US Congress. About $9,800 of his own at twenty. Studied under Benjamin Graham at Columbia, then worked for him — arguably the best apprenticeship that has ever existed in the field, available to roughly twenty people on earth. His first fund opened at $105,100, of which his own contribution was $100; the rest came from his mother, sister, aunt and father-in-law.
- The decisive move
- 1967–69, in two steps. He bought National Indemnity for $8.6m, acquiring his first pool of insurance float. Then he dissolved his partnership at its peak and took Berkshire shares instead of cash — converting himself from a manager taking a cut of other people's money, which can be withdrawn at any time, into the owner of a vehicle that can never be redeemed. Float went from $39m in 1970 to $176bn by the end of 2025.
- How the money was realised
- It never was, and that absence is the mechanism. No sale, no IPO, one dividend in sixty years. Roughly 98% of his net worth has always been Berkshire stock he simply never sold, so the gains stayed unrealised, untaxed, and compounding on the full pre-tax amount. Between 2014 and 2018 his wealth rose $24.3bn while he paid $23.7m in federal income tax — a true rate of about 0.10%.
- Elapsed time
- Thirteen years to never having to work again. Thirty to his first billion. Roughly 98% of the fortune arrived after he turned sixty.
Transfers
- Get paid only on results and put your own money in the same boat. His partnership took no management fee and 25% above a 6% hurdle, with shortfalls carried forward. He could not win unless his backers won first — the cheapest way on earth to be trusted with capital you do not have.
- Look where professionals cannot be bothered, because the opportunity is too small for them. That edge is most available in thin, under-covered markets — which is to say, in exactly the kind of market most readers are standing in.
- Keep fixed costs near zero so you can never be forced to sell. A flat $100,000 salary for decades and the same house since 1958. This is why he held through every crash, and it is buildable at any income.
- Change your method when someone smarter proves you wrong — even while the old one is still working. Munger made him abandon a profitable approach in his forties.
Does not transfer
- The float itself — the actual engine. It requires owning a regulated insurer: approval, enormous statutory capital, decades of underwriting discipline. It is the only leverage that cannot be margin-called, and you cannot have it.
- Being American, in 1930, holding the reserve currency, through the best fifty-year stretch equities have ever had. Compound 20% a year in a currency that halves against the dollar and you are going backwards while winning.
- Sixty years with nothing interrupting the compounding — no war, no seizure, no currency reset, no forced sale. Most people's runway is broken by life.
- Deal flow that exists only because he is Warren Buffett. In 2008 Goldman Sachs sold him preferred stock at terms available to precisely one buyer.
That he got rich picking good stocks and being patient. That describes what he did with the money, not where it came from. He ran roughly 1.6× leverage financed by other people's insurance premiums at a cost below the Treasury bill rate, plus tens of billions in deferred taxes acting as a second interest-free loan. And the deflating part: academic analysis found his alpha becomes statistically insignificant once you control for two known factors — cheap leverage applied to high-quality, low-volatility businesses. The rate of return is the least important variable in his story and the only one anybody copies.