Every three months, Wall Street performs a ritual dance known as quarterly earnings. It goes like this: A company undersells its own expectations. Analysts at brokerage firms, not wanting to alienate a corporate client, follow its lead and give out lowball estimates. Then reported earnings top estimates and the stock price rises, at least temporarily. Since 1994, according to Bianco Research, more than half of the companies in the Standard & Poor’s 500-stock index have topped the median analyst estimate (as collected by Bloomberg LP, the parent company of this publication) in every quarter but four. So far, for the first quarter of this year, 79 percent have “beat the Street.”
Leigh Drogen is out to change the routine. The 26-year-old Chappaqua (N.Y.) native has created an online clearinghouse for earnings estimates called Estimize. The site gathers estimates from all corners of the investment world, from the so-called sell-side analysts at the banks, who have been calling the tune for the past three decades, to their buy-side counterparts at hedge funds and proprietary trading desks, to independent investors. While estimates have been crowdsourced since before there was such a word, Estimize, says Drogen, is a new form of crowd control. “We are able to put forth the guys who are actually the best analysts,” he says, “vs. the guys who are just the loudest or have the biggest names behind their desks.”
Anyone can contribute to Estimize by signing up for a free account. More than 4,300 people have done so since Drogen launched the site in December, with about 77 percent self-identifying as independent, 20 percent as buy-side, and 3 percent as sell-side. Once inside, members can look up a stock and then move toggles to enter their guesses for earnings per share and revenue. There is also a space to include analysis. Estimize aggregates the entries and posts them next to a “Wall Street” number taken from Zacks Investment Research, which uses the “average of all the current estimates made available by brokerage analysts.” So far, says Drogen, his ragtag crew is besting the usual suspects. In the first quarter of Estimize’s life, the crowd’s estimates have been closer to reported results 63 percent of the time. This covers 120 sets of earnings for which Estimize had at least four estimates.
The Estimize crowd anticipated Research In Motion’s (RIMM) underwhelming fourth-quarter revenue, predicting $4.48 billion while Wall Street guessed $4.54 billion. The BlackBerry maker came in at $4.19 billion. And it expected Intel’s (INTC) better-than-expected results for the fourth quarter. Estimize predicted 66 cents in earnings per share; Wall Street, 61 cents. Intel came in at 64 cents. Admittedly, this is cherry-picking from a small sample size, but that, says Drogen, is part of the point. Estimize can not only help correct for corporate spin, or “guidance” as it’s known in the business, but, by opening its doors to all comers, it can help spot unknown talent. The site provides an analyst leaderboard for each stock and a track record for each analyst. Drogen points to Rob DeFrancesco, or TechStockRadar as he is known on Estimize, who has been more accurate than the Wall Street consensus on 11 of his 13 estimates so far and the most accurate analyst on the site for seven of them. “His ability to do this is pretty ridiculous for a guy who just runs his own book,” says Drogen.
Estimize is not the first attempt to bridge the gap between the official consensus and what the financial community really thinks about earnings. For as long as corporations have been gaming the system, analysts (including the very sell-side gang responsible for the deflated numbers) have been circulating another, usually slightly higher, figure called the “whisper number.” Drogen participated in this traffic when he worked at the hedge fund Geller Capital from 2006 to 2009, and later at his own fund Surfview Capital, where he crafted investment strategies around the reliably “skewed set of expectations” produced by the sell-side. Drogen left Geller to work at Howard Lindzon’s social finance pioneer StockTwits, where he noticed that many in that community were already sharing earnings estimates. He offered to build a platform for aggregating the numbers inside StockTwits. Lindzon passed but gave his blessing for Drogen to build it on his own. So Drogen raised $200,000 from a handful of investors and started Estimize.
WhisperNumber has been publishing unofficial estimates since 1998. The New Jersey-based company gathers its numbers from a group of anonymous investors who register with the site, runs these figures through “a proprietary algorithm,” and shares the results for free with its 85,000 users. (The company also sends tips based on its data about how stocks are likely to move on earnings to a smaller set of users who pay about $800 a year.) WhisperNumber’s president, John Scherr, says he has seen plenty of upstarts come and go. “I think he’s a bit late to the party with this,” he says of Drogen’s Estimize. “It’s a nice attempt, but I don’t think it’s going to fly.”
Drogen says he’s not out to topple WhisperNumber or anyone else in the business so much as roll them all into one big, transparent bag of numbers. “What we’re trying to disrupt here is the idea that the whisper number is a whisper number,” he writes in an e-mail. “We want it to be open and transparent where you can see exactly who it’s coming from and how the sausage is created. We actually pull their estimates into our site. We want people to know about them. They are just another data point.” (Scherr has since asked Drogen to remove the WhisperNumber data from Estimize. Drogen says he’ll comply.)
Estimize does allow users to remain pseudonymous, which Scherr suggests undercuts Drogen’s rhetoric of transparency. “If you’re going to go the transparency route, you’ve got to use real names,” he says. “If you’re right, you want to be known, but most people aren’t right. Most people are wrong. And nobody wants to be known as the guy who has a 50 percent accuracy rating.” But the pseudonymous Web, says Drogen, is the future. “We don’t care that you don’t identify yourself as who you are in real life, as long as you create a track record,” he writes by e-mail. “It allows 18-year-old kids who are amazing traders to share their ideas, and be recognized for the quality of what they are sharing,” which is why the site is now hosting a $10,000 challenge for the best estimators.
If all goes according to plan, Drogen will eventually make himself obsolete. As estimates get closer and closer to results, the trading opportunities disappear. “If we’re successful at this, we take a massive amount of volatility out of the market,” he says, “and we’ll be happy if we disrupt ourselves, we’ll be really happy that we succeeded.”
沒有留言:
張貼留言