Is Crowdsourced Investment Research Here to Stay?

Crowdsourced equity research platforms continue to grow as investors rely on and have access to online information. On the surface, websites like Estimize, Seeking Alpha and StockTwits appear to compete with Wall Street analysts, potentially unhinging an incumbent model. New research by SMU Cox Accounting Professor Stanimir Markov and co-authors reveals that one online venue, Estimize, has positive predictive powers, over and above Wall Street, when it comes to forecasting corporate earnings.

In their research, the authors also suggest that Estimize earnings forecasts are less biased and bolder (further from the consensus) and a better source for short-term information. According to Markov, "Typical sell-side Wall Street analysts are believed to be more biased because they serve too many masters." He continues, "Estimize forecasts are less so because of the many different people with various backgrounds who are only interested in estimating earnings. This crowd is not worried about offending corporate management or upsetting investment banking colleagues."

Online social media sites are delivering investment advice to about one out of three U.S. individuals, the research notes. The Forbes-recognized start up Estimize was founded in 2011. During 2012 and 2013, 3,255 individuals submitted 51,012 quarterly earnings forecasts for 1,874 firms. Estimize crowdsources earnings forecasts from a diverse group of contributors such as analysts, portfolio managers, independent investors, corporate finance professionals and students. Estimize forecasts are available on estimize.com and Bloomberg terminals, and also sold as a data feed.

Findings

Estimize forecasts are incrementally useful, according to Markov and the study. Said another way, "The Estimize consensus is informative over and above the sell-side consensus," Markov surmises. The combination of IBES (Wall Street) and Estimize forecasts 30 days prior to an earnings announcement "is more accurate than the IBES consensus 60% of the time," the authors note. Further, that measure increases to 64% on the day prior to the earnings announcement.

The authors note that the significant difference in forecast timing suggests a complementary relation between IBES analysts and Estimize contributors. Approximately half of Estimize forecasts are issued in the two days prior to announcement; less than 2% of IBES (Wall Street) forecasts are issued in the same period. According to Markov, “Estimize is inactive over longer-horizons (e.g. forecasting one or two-year ahead earnings). This is where the sell-side rules — they are then the only game in town. But closer to earnings announcements, the sell-side goes dark, creating a niche for Estimize to fill in."

Markov’s research shows that Estimize forecasts are useful because they are less biased than IBES forecasts and occur later (or closer to the date of an earnings announcement). This allows the crowd to incorporate more public information and information missed by the sell-side.

Additionally, as the size of the crowd increases, the relative importance of Estimize as a measure of the market’s expectations increases. The research finds that if there are five or more Estimize contributors, the Estimize consensus fully encompasses the IBES consensus. On an individual basis, however, the average individual Wall Street analyst issues more accurate forecasts than the individual Estimize contributor. Thus, Estimize's value comes from attracting a large crowd with diverse and independent information.

Meaning of the crowd

For many individual investors or boutique investment firms that lack access to the sell-side analysts of Wall Street, online investment sites such as Estimize and Seeking Alpha help level the playing field. "If you are a small investor wondering about earnings, you are essentially cut off from Wall Street's sell-side analyst consensus," says Markov. "This is a way for small investors collectively to see what everyone is thinking about corporate earnings. They can then interpret any earnings surprise. Wall Street has this knowledge and does not have to disclose their information publicly. Small investors do not have access to this knowledge."

About the research, Markov describes it as a first academic research study comparing apples to apples, that is, earnings forecasts provided by the crowd to earnings forecasts provided by the sell-side. He adds: "I was surprised that Estimize even exists, and I am surprised at how well the amateurs do versus the professionals. It is amazing to observe people taking the time to forecast earnings and contribute their earnings estimates to Estimize."

Investors now have a useful resource, courtesy of founder and former quantitative hedge fund analyst Leigh Drogen, concludes Markov. The best analyst-contributors are recognized with prizes and featured in podcasts. The longevity of the crowdsourced investment phenomenon is yet to be known — the crowd could grow weary if not properly incentivized. At the same, the recent growth of Estimize contributor base, a key value driver for Estimize, suggests that the usefulness of Estimize may continue to grow.

The paper, "The Value of Crowdsourced Earnings Forecasts" by Stanimir Markov, Cox School of Business, Southern Methodist University; Russell Jame of University of Kentucky; Rick Johnston of University of Alabama at Birmingham; and Michael Wolfe of Virginia Tech is under review.

Written by Jennifer Warren.