March 14, 2016

Statistics and Warren Buffet's returns

The media loves Warren Buffett. Everyone does. And I’m a big fan too, not least because he’s a staunch advocate of indexing for ordinary investors.
He’s also idolised by those who believe in active management. He is, simply, the Sage of Omaha — the most famous active investor in the world.
But how good is he? Don’t get me wrong, Buffett’s long-term performance has been exceptional. But some of the old magic appears to have worn off over the years.
I’ve been speaking to statistician, author and blogger Salil Mehta, who has studied Buffett’s returns in detail. Based at Harvard, Salil runs the website Statistical Ideas, and serves on the Editorial Board of the American Statistical Association and as a consultant to BlackRock.
If you put to one side the media adulation and focus on the value that Buffett has actually delivered, Professor Mehta says his record is not as extraordinary as most people think. 


You’ve tracked the performance of Buffett’s company, Berkshire Hathaway, very closely over the years. How has it done lately?
Last year the S&P 500 returned a lousy 1.4%, but Berkshire Hathaway was down 12.5% on a market-value basis. Over a two-year period, Buffett has returned 11.1% while the S&P has returned 15.3%. Per year, that’s 5.4% for Buffett and 7.4% for the S&P.
I suppose you can divide Warren Buffett’s career into two. Broadly speaking, he was very successful until the turn of the Millennium; since then he’s done well, but he hasn’t shot the lights out. Why do you think that is?
Luck finally ran out on the great investor. He had a good run from the start, and arguably if he stopped part way through his career there would be no way to disprove that he was exceptionally skilled. But over time we can see the particular trading strategy he used stopped working and his performance dissipated. 
You haven’t exactly ingratiated yourself with Buffett fans with what you’ve written in recent years, have you?
Berkshire Hathaway’s performance on either a book-value or a market-value basis has been quite mediocre in recent years. I first suggested this was the case years ago, and there was a firestorm from Buffett devotees — some of them in the financial press — who incorrectly assumed that his performance would continue to be as strong as ever.
You’re suggesting that Buffett’s success in the first part of his career can partly be explained by luck. Is it also possible that bad luck has played a part in his failure to replicate that success latterly?
No, it’s not really possible that the recent underperformance is down to bad luck.  The performance is not terrible for any random investor — it’s just bad relative to the great returns Buffett once had. He’s continuously underperforming the S&P by a small amount and, relative to everyone else, there is no shame or bad luck in that.
Do you think Buffett will ever be able to produce double-digit annual returns again?
The S&P has produced about a 5% annual return over the past two decades, and I don’t think the benchmark will produce double-digit returns for any sustained period. A recent article of mine shows that  most managers perform close to the benchmark, and that none really outperform by skill. I expect Buffett to be no different.
What is the one eye takeaway from your research into Buffett’s performance?
It’s nearly impossible to beat the market by a statistically significant (or risk-adjusted) amount over the long run. Warren Buffett used to do it, but even he stopped doing that a while back.

No comments:

Post a Comment

Share this...