Buffett's million-dollar bet against hedge funds — and why the boring option keeps winning
In 2007 Warren Buffett offered a wager to the hedge fund industry. Pick any group of at least five hedge funds, he said, and over ten years they would fail to beat a plain, low-cost fund that simply owned the S&P 500. The stakes were a million dollars, payable to charity. The industry, which manages hundreds of billions of dollars on the premise that it can do better than the market, was not exactly queueing up to take him on. One man did: Ted Seides, a co-manager of Protégé Partners, a firm that invested in hedge funds for a living.
The bet ran from the first day of 2008 to the last day of 2017. Buffett chose a Vanguard S&P 500 index fund. Seides chose five funds of funds, each of which spread its money across many underlying hedge funds. On paper, that was a formidable line-up: dozens of professional managers, the freedom to go long or short, to use leverage, to move between markets, and to sidestep a crash. Against them stood a fund that did nothing at all except own five hundred large American companies and charge almost nothing for the privilege.
How the decade actually played out
The timing looked kind to the hedge funds at first. The bet began just before the 2008 financial crisis, and in that first year the index fund fell sharply while the hedged portfolios lost considerably less. That is precisely what hedge funds are supposed to do, and for a while the challenger was ahead.
Then came the long recovery. The index fund, having taken the full blow, also took the full rebound. The funds of funds, carrying their hedges and their fees, did not. By the end of 2016, nine years in, Buffett reported in his annual letter that the index fund had compounded at roughly 7.1% a year, against roughly 2.2% for the average of the five funds of funds. Seides publicly acknowledged during 2017 that the contest was effectively over. When the final year closed, the index fund had more than doubled the money, while the hedge fund selection lagged far behind. The charity Buffett had named, Girls Inc. of Omaha, collected the winnings.
It is worth being fair to the losing side. Seides argued afterwards that the period happened to suit a cheap, fully invested US stock fund unusually well, and that hedge funds are built to offer something different: lower volatility and protection in bad years. There is something to that. But a decade is a long time, it included one of the worst crashes in modern history, and the protective structure still did not come close to paying for itself.
The arithmetic nobody likes to talk about
The real lesson of the bet is not that hedge fund managers are foolish. Many are extremely clever. The lesson is that cleverness has to clear a very high fee hurdle before it does the client any good.
The classic hedge fund charge has long been described as "two and twenty": around 2% of assets every year, plus around 20% of any profits. A fund of funds adds its own layer on top, often around another 1% a year, sometimes with a share of profits too. So before the investor sees a cent, the underlying managers have to earn enough to pay two sets of fees. In a world where stocks might return something like 7% to 10% a year over the long run, a combined drag of three or four percentage points is not a detail. It is most of the race.
There is also a simple piece of logic, often credited to the economist William Sharpe, that explains why the boring option tends to win. Before costs, the average actively managed dollar must earn the market return, because all the active managers together essentially are the market. After costs, the average active dollar must therefore trail the market by roughly the amount it pays in fees and trading costs. Some managers will beat the market in any given decade, of course. The problem is identifying them in advance, and paying them less than the value they add.
Why the boring option is so hard to choose
If the case for low-cost index funds is so clear, why do so many people, including large and sophisticated institutions, keep paying for complexity? A few reasons keep recurring.
- Doing nothing feels like negligence. A trustee who buys an index fund and goes to lunch looks less diligent than one who interviews twenty managers. The second trustee will usually earn less, but will look busier doing it.
- Fees are quiet, returns are loud. A 2% annual charge never arrives as a bill. It is shaved off silently, year after year, while the marketing focuses on last year's star performance.
- Stories beat statistics. "Our manager anticipated the housing crash" is a better story than "we owned everything and kept our costs low." The story gets the meeting; the statistics win the decade.
- Skill exists, which makes it tempting to assume we can find it. The trouble is that luck and skill look nearly identical over a few years, and by the time the difference is clear, the fees have long been paid.
Buffett has been consistent on this for decades. In his letters he has repeatedly argued that most investors, institutional and individual, would do best owning stocks through a low-cost index fund. In his 2013 letter he went further and described the instructions he left for the trustee of his own estate: put a small portion in short-term government bonds and the rest in a very low-cost S&P 500 index fund. This from a man whose own career is the most famous counterexample to the advice.
What the bet does not prove
It would be easy to over-learn the lesson. The bet does not prove that the S&P 500 will beat everything, everywhere, every decade. American large companies had an exceptionally strong run in the years that followed 2008, and there have been long stretches, such as the 2000s, when the index went roughly nowhere for ten years. Nor does it prove that every hedge fund is a bad deal, or that diversification beyond US stocks is pointless.
What it does show is narrower and more durable. Over a long period, the burden of proof sits with anyone charging high fees to do better than a cheap, broad benchmark. Complexity is not the same as value. And the option that requires the least effort, the least cleverness and the least excitement is very often the one left standing at the end. Charlie Munger liked to make a related point: a great deal of long-term success comes not from being brilliant, but from consistently avoiding being stupid. Overpaying for the hope of brilliance is one of the more common ways to be stupid.
Where Sixty-40 comes in
Every game of Sixty-40 ends with the same quiet opponent: the S&P 500, bought with the same money on the same dates, and left alone. It never panics, never chases the hot pick, and never gets excited about a sock puppet. You, meanwhile, have ten seconds and five tempting options. Some rounds you will beat it handsomely, especially if the wheel hands you an early Apple or Nvidia. Play enough rounds, though, and you will start to feel what Ted Seides felt by 2017: the dull benchmark is a far tougher opponent than it looks.