A recent academic study revealed that the stock market ideas published on internet investor forum Seeking Alpha outperformed the research written by Wall Street analysts and articles published by financial news services over the past seven years supporting the belief in the “wisdom of the crowds”.
Background of the Study
Last week, the Wall Street Journal wrote an article highlighting a recent academic study called Wisdom of Crowds: The Value of Stock Opinions Transmitted Through Social Media which found that stock market recommendations published on internet-based investor forum SeekingAlpha.com predicted individual stock returns, as well as earnings surprises, in excess of what is provided by Wall Street analyst reports and financial news articles.
The study, which will be published shortly in the Review of Financial Studies, was written by Hailiang Chen, Prabuddha De, Byoung-Hyoun Hwang, and Yu (Jeffrey) Hu of City University of Hong Kong, Purdue University and Georgia Institute of Technology. The researchers analyzed the sentiment of approximately 100,000 articles and related comments posted on SeekingAlpa.com between 2005 and 2012.
The study evaluated the percentage of positive or negative words in SeekingAlpha articles and the associated comments on a particular stock, and tracked the performance of that stock after the article and comments were posted. In order to eliminate the short-term impact of investors immediately responding to published articles, the researchers started tracking the performance of these share prices 48 hours after the article was published.
In addition, the researchers tried to determine whether the articles on SeekingAlpha were primarily moving the market rather than predicting it. They did this by also analyzing whether the sentiment of SeekingAlpha articles or the associated comments were highly correlated with earnings surprises.
Findings of the Study
What the researchers discovered was the more positive the articles and comments were on a specific stock, the more likely that stock was to perform better than similar stocks over the next several months. They also found that negative articles and comments also underperformed similar stocks in the future.
In fact, the study concluded that on average, someone investing according to the sentiment of the articles and comments and holding positions for three months would consistently beat the market. This finding held up even when a number of variables were controlled for, including Wall Street analyst recommendations, upgrade/downgrades, earnings surprises, and the sentiment of traditional financial news articles.
In addition, the study found that the aggregate opinion of SeekingAlpha articles and comments had a strong correlation with earnings surprises, suggesting that the articles did not merely move stock prices, but they had some predictive power.
“We find that the fraction of negative words in SA articles and comments strongly predict subsequent scaled earnings surprises,” says the study. “The earnings-surprise predictability suggests that the opinions expressed in SA articles and comments indeed provide value-relevant information (beyond that provided by financial analysts).”
The study also found that SeekingAlpha articles and comments predicted stock returns over every time-frame examined: three months, six months, one year and three years.
Lastly, the study concluded that in cases where there is broad-based disagreement between the authors of articles and the community’s comments to these articles, the sentiment of the community was more accurate than the authors in predicting future stock price performance and earnings surprises.
Some Issues with Seeking Alpha
Despite the support that SeekingAlpha and crowdsourcing of investment ideas received from this recent academic study, a few have been quick to point out the dangers that sites like SeekingAlpha could create for individual investors.
In an article published this past weekend in Barron’s, called Seeking Alpha Needs to Take Stock of its Policies, John Kimelman wrote that several of SeekingAlpha’s policies could lead to sloppy analysis or even market manipulation by its contributors.
Kimelman wrote, “But Seeking Alpha has also received some troubling press in recent weeks, exposing problems inherent in the site’s policy of allowing anonymous contributors. The problems, I contend, also stem from less than stringent editorial standards that need to be tightened up so that Seeking Alpha can do a better job resisting stock manipulators who see the site as an easy mark.”
In the article, Kimelman discusses a recent instance when SeekingAlpha removed at least a half dozen favorable articles about Galena Biopharma after concluding that the anonymous writers of these articles weren’t being truthful about their identities. After investigating this issue, SeekingAlpha discovered that one individual wrote five articles under different pseudonyms.
Unfortunately, a number of Galena insiders sold shares of the stock which may have gained value, at least in part, due to the favorable articles.
Kimelman concludes that the by protecting the anonymity of the authors, SeekingAlpha is overlooking the dangers that could result from this policy, like enabling a writer to publish sloppy or one-sided analysis, or even allowing a single writer to try and manipulate stock prices by publishing overly bullish or bearish articles by making up phony aliases.
The second troubling policy Kimelman highlights is the fact that SeekingAlpha asks its writers to disclose whether they are long or short the stock when they write an article. Unfortunately, the forum has no way to tell if the writer is telling the truth, an issue which could call into question whether the article is biased or not.
Kimelman concludes his article by explaining, “In response to some of my questions, [Eli] Hoffman [the Editor in Chief of SeekingAlpha] says that his site is taking steps to improve the quality of the site, including a new compensation system that rewards quality rather than just page-views. But no steps are being taken at this time to limit the use of anonymity. I would urge them to reconsider that point.”
In our mind, the recent study clearly points out the value that crowd sourcing of investment ideas from a site such as SeekingAlpha, can provide to individual investors. In fact, Yu Jeffrey Hu, associate professor at Georgia Tech’s Scheller College of Business and one of the authors of the academic study, pointedly explained that, “Seeking Alpha is the only platform to date that we have shown can predict individual stock returns.”
However, the recent Barron’s article also highlights how sites like SeekingAlpha can be abused by bad actors trying to profit from the impact the site has on stock prices – a fact that could put some investors at risk of being manipulated.
Given this, perhaps the most important role that SeekingAlpha, and other crowd sourced investment research sites, should play (at least at present) are they are great sources to find interesting ideas which investors can use to start their research process.