Twitter better than Wall Street for earnings predictions?

Photo courtesy of Scott Beale / Laughing Squid on Flickr This photo is licensed under a Creative Commons license. If you use this photo within the terms of the license or make special arrangements to use the photo, please list the photo credit as "Scott Beale / Laughing Squid" and link the credit to

Photo courtesy of Scott Beale / Laughing Squid on Flickr

Rob Swystun, Pristine Advisers

Could Twitter be a better source for earnings estimates than Wall Street research? That’s what a recent study suggests.

Crowd-sourced company earnings estimates and sentiment data generated by tweets may be more accurate than the available Wall Street research and the tweets may also help generate trading profits, according to a new study by members of John Hopkins University in Baltimore.

To come to these conclusions, the team at John Hopkins examined data from Estimize, a service that allows users to make earnings estimates and then processes the crowd-sourced data. The team also looked at data from iSentium, a firm that specializes in collecting company sentiment data from Twitter.

On the surface, it would appear that tweet sentiment has the power to predict excess returns that companies deliver to shareholders after their earnings reports, the John Hopkins team — Jim Kyung-Soo Liew, Shenghan Guo, and Tongli Zhang — write in the paper.

Risk-adjusted excess returns are a measure of the return a stock generates compared to a portfolio of similar stocks, benchmark or index.

“There appears to be some indication that a well formulated strategy incorporating both data sets could be an interesting avenue of future research and may lead to annualized excess gross returns in the 10 percent to 20 percent range,” the study researchers wrote.

As reported by James Saft of Reuters and run in the Sydney Morning Herald, this conclusion comes with many caveats.

A Look at the Twitterverse

Based on earnings announcements between November 2011 to December 2014, the study used iSentium analysis data to create a value for a company. They looked at all the tweets on a given company and gave those tweets a value based on them being either positive toward the company or negative.

These tweet values were then crunched down to a single value given to the company, which was intended to be a reflection of how positive or negative the Twitterverse was toward the company at the time of its earnings release.

They then looked at what happened to company shares after the release of earnings.

For example, Saft said, going short (betting that share prices will fall) on stocks owned by companies that received negative Twitter sentiment which post a negative earnings surprise generates a bit more than half a percent of risk-adjusted excess returns in the following five trading days. That equates to an annual excess return of about 26%, which is huge.

Betting share prices will rise in stocks with positive tweet sentiment and positive earnings surprises also generate excess returns, but on a lower scale

Street Beating

Some drawbacks to the study and its findings are that the results don’t include trading fees and some of the stocks don’t belong to large companies rendering the shorting scenario improbable, at least on a large scale.

Also, the effects in the study are, of course, short-term. Although the market does tend to move on the news, the gains would need to be realized within days.

However, that doesn’t change the apparent fact that the purely crowd sourced Estimize earnings estimates are better than the expensively-generated Wall Street estimates.

“Over the whole period of our data we find that Estimize‘s consensus is 56.4% more accurate than Wall Street’s consensus – similar to earlier findings,” the study authors write in their report.

Fewer estimates in the Estimize data “lowball” earnings while more tend to overestimate. This is highly significant because it’s something that few Wall Street firms do and it’s what makes the crowd sourced estimates more accurate.

Twitter Lacks Conflicts of Interest

A crowd being able to better estimate earnings isn’t actually that surprising, as Saft notes the financial services industry has a well documented history of underestimating earnings going into announcements that is largely driven by nefarious reasons.

Part of the reason so many firms underestimate earnings may be so companies can beat these estimates and therefore look favorable in the eyes of their stakeholders.

Another reason why firms underestimate is because of what Saft calls agency problems due to misaligned incentives of the financial services industry. It is possible — and even probably, according to Saft — that equity analysts aren’t giving their actual opinions on earnings outlooks because they have their own interests to advance and protect.

Analysts can also be swayed by their desire to keep the lines of communication open with firms. As Saft notes, an analyst who keeps executives happy is one that can keep the lines of access open and help smooth the way for lucrative investment banking business for the firm they work for.

“Not being overly influenced by the need to gain access and favor from company executives appears to be a reasonable explanation supported by the data,” the authors write. “Additionally, building relationships over time with conference calls, equity research events, and social gatherings may have led many Wall Street analysts to lose their objectivity when forecasting company’s earnings.”

Whether or not this study accurately reflects social media’s ability to predict earnings better than Wall Street analysts will require more research, but it is indicative of a larger trend, that being the threat the financial industry faces from technology. Investors may not need analysts in the future to play middleman.

Saft predicts that open sourced information like Twitter and other social media tools are going to drive down the margins of middlemen in and outside of the financial industry.

Will this be so? Let’s ask Twitter for a prediction on that.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s