It's difficult to compound gain at 19% for 50 years--the S&P only did 10%. The secret of Buffett's success, which he explains in every letter, is that he bought an insurance company. As long as you are diligent in writing non-money-losing policies, you get the premiums up front and pay later. So you have all this float that you can do something with. But he also picks good companies, picks good managers, and is careful to make sure that the financial incentive for the manager is the same as the financial incentive for Berkshire, which is not the case with most executive packages.
Leverage is only minor component of his late career success, and insurance float could never have gotten him broke. It's totally unlikely like a loan from your broker.
His late career has been less impressive vs is early career. Also, float can dry up if insurance sales drop. So in many ways it's worse than a loan from your broker it's got totally random size fluctuations. He just happened to be really lucky / skilled at running an insurance company over time so that this was not an issue. However, that says more about running an insurance company than investing.
If his results declined after he bought insurance companies, how does that say he was lucky/skilled at running them, and how did he gain any benefits from the leverage they afforded?
The reality is his late career is less impressive only in one way, annualized returns, and it's more impressive in virtually every other way. He started out with less than a million in investors money, and his 40% returns with the Buffett Partnership were probably with an average of $10M to $20M total.
Nowadays his public stock portfolio was $122B at year end 2016. That limits the stock market investments he can make to a tiny fraction of stocks. He's not going to buy hundreds of stocks, not even dozens. He wants to focus in his dozen or so best ideas. So he want to be able to put $10B to work in each and there are very few public companies with market caps and trading volumes large enough to allow him to do that. So problem #1 is that he went from a market of 5,000+ stocks he could choose from to maybe a few hundred, reduced opportunities that directly hurts his returns.
Worse, how can he accumulate $10B worth of a company before he's forced to publicly disclose his purchases. If he's forced to disclose prematurely, his future returns dwindle due to free-riders driving the purchase price up before he's done. That's big problem #2.
The fact that he's still beating the market carrying these heavy chains is more impressive to me than the fact he whipped it at 3x higher rates when he could buy anything he wanted.
But, take out those 40% years and his returns are much closer to the stock indexes.
While managing billions at minimal risk he is not making nearly 17% returns. He is doing OK but again your looking at a statistical outlier so you can't separate skill and luck as much as you might think.
His returns from Berkshire Hathaway has averaged 19% over the last 52 years (from 1965), vs a 9.7% market return over the same period.
His 40% returns were in the Buffett Partnerships, which ran for 12 years from mid 50s to late 60s, and aren't counted in Berkshires results.
Doubling the market return rate over 52 years is huge outperformance. For an investor it's roughly doubling every 3.5 years instead of 7.5 years.
Add in the fact he crushed the market by nearly 4x a year for an earlier 12 years and his immense skill is undeniable. He would be statistically implausible if the world had trillions of investors.
