Bloomberg’s weird Buffett spoiler
Bloomberg and Fortune had weirdly competing stories Wednesday on the subject of Warren Buffett’s “million-dollar bet“. The bet’s duration is ten years from January 1, 2008; Buffett is betting a million dollars that the S&P 500 will outperform a fund of hedge funds.
Fortune’s Carol Loomis has unrivalled access to Buffett — she counts herself among his friends, and always helps him write his annual shareholder’s letter. So her report on the status of the bet, time stamped 8:30am, can be taken as definitive. Loomis doesn’t reveal the components of the fund-of-funds that Buffett is betting against, but she does reveal (“you are reading it here first”, she writes) the standings at the end of Year Four — that is, at the end of calendar 2011. The fund-of-funds has not done well over those four years: it’s down 5.89%. But Buffett’s index fund is doing even worse: it’s down 6.27%.
Loomis will always get these scoops, for as long as she’s close to Buffett. That’s fine. But then why did Bloomberg’s Katherine Burton decide to run a story on the exact same bet on the very same morning as Loomis’s scoop, under the headline “Buffett Seizes Lead in Bet on Stocks Beating Hedge Funds”?
The first effect of Burton’s story is simply to confuse everything. Fortune is saying that Buffett is behind; Bloomberg is saying that Buffett is ahead. If you read Fortune or the outlets which picked up Fortune, like the AP, then you’ll believe one thing; if you read Bloomberg or the outlets which picked up Bloomberg, like MSNBC, you’ll believe another thing. And if you read them both, you could be forgiven for thinking that someone is playing with your head.
A close reading of Burton’s story helps to reveal what’s going on here. For one thing, she’s reporting the status of the bet through February 29, rather than the status of the bet after four years. That is peculiar, not least because the fund-of-funds only reports annual results: Loomis was waiting to see what its 2011 returns were before she wrote her story. Burton, by contrast, despite being 25 minutes behind Loomis with her story, only knows what the fund-of-funds returned through the end of 2010. Here’s her explanation for how she calculates the performance of the fund-of-funds:
The hedge funds fell about 4.5 percent, based on Protégé’s index returns for the first three years and results since then for the Dow Jones Credit Suisse Hedge Fund Index, which has roughly tracked the group of unidentified funds when adjustments are made for extra fees.
In other words, when Burton’s story hit the web at 8:55am, it was already out of date: she extrapolated 14 months forwards from Loomis’s 2010 figures, rather than waiting for Loomis’s story to arrive and then extrapolating a mere 2 months forwards from the 2011 figures.
But in any event, Loomis was reporting the facts of the bet; Burton is just taking an educated guess. Hedge funds are by their nature unpredictable things. The fund-of-funds in this bet might have “roughly tracked” some hedge fund index for its first three years, but it can veer far off-index at any time, depending on how it’s put together. Which means that an unambiguous “Buffett Seizes Lead” headline is quite misleading.
That said, Buffett might well have seized the lead at the end of February, if Burton is right about the fund-of-funds tracking the Dow Jones Credit Suisse Hedge Fund Index. That index rose from 92.6 at end-2011 to 95.93 at end-Feb, a rise of 3.6%. We know from Loomis that at end-2011, the fund-of-funds was down 5.89%, which means that it stood at 94.11% of its initial value. If it grew from there by 3.6% in two months, it would have ended February at 97.5% of its initial value, for a loss of 2.5% overall.
The S&P 500, by contrast, rose from 1,257.6 at the end of 2011 to 1,365.7 at the end of February. That’s an increase of 8.6%, fully five percentage points greater than the rise in the hedge-fund index. We know from Loomis that Buffett was down 6.27% at the end of 2011; if his fund rose 8.6% from that level, it would have ended February at 101.8% of its initial level, for an overall gain of 1.8%.
It turns out, however, that even the Buffett side of the bet isn’t particularly easy to calculate: Burton reckons he was up 2.2% as of end-Feb, not 1.8%. But the 40bp difference there pales in comparison to what she’s estimating for the fund-of-funds: she says that it’s down 4.5%, rather than being down 2.5%. That difference, of 200bp, is really substantial.
Ultimately, the only thing we know for sure is that for four years in a row, the fund-of-funds has been in the lead, thanks to substantially outperforming the S&P 500 in 2008, the first year of the bet. At the end of five years, that might have changed: Buffett could well be back in the lead. And it’s even possible that someone with detailed knowledge of how the fund-of-funds is made up could trace the point at which Buffett took the lead back to February 2012. But no one has that information right now, or if they do have it, they’re not telling. The exact make-up of the fund-of-funds is a closely guarded secret.
In any case, the whole point of long bets is that they’re long-dated. This one has a ten-year maturity, and what happens even from year to year is not particularly important, let alone what happens from month to month. Yes, the S&P 500 had a very healthy run in January and February of 2012, and probably outperformed many hedge funds. But there are always going to be two-month periods where hedge funds underperform the index, and obviously the S&P 500 can’t rise by 4.3% a month for any sustained length of time.
So I’m a fan of the way that Loomis is reporting this, deliberately, using hard year-end numbers from the fund-of-funds, even if she has to wait 11 weeks from the end of the year before she gets them. Burton’s piece is significantly less informative even if it’s a couple months more up to date, and it’s also much less in sync with the underlying philosophy of the bet.
As for the decision to release Burton’s story on the exact same day that Loomis’s semi-official report came out, that just looks childish. It’s no secret that Loomis is very close to Buffett; let her have her scoop. It’s a perfectly good story, which in no way requires a Bloomberg spoiler.