In his book The Intelligent Investor, Benjamin Graham expresses a distinct wariness, if not outright disdain, for the wisdom of crowds. “You are neither right nor wrong because the crowd disagrees with you,” he penned. “You are right because your data and reasoning are right.” These days, the increasingly robust intersection between social media and professional finance seems to be turning Graham’s saying on its head. In finance, the “crowd” is no longer just the other side of a trade or a contrarian indicator but a source of insight, sometimes predictive, that can be harnessed as part of the investment process.
According to a white paper by Gnip, the social data provider acquired by Twitter, 2013 marked a sea change. Three events in particular focused the attention of financial professionals on the power of social media to move markets: in April 2013 the US Securities & Exchange Commission gave the official nod permitting companies to announce key information on social platforms and remain compliant with Regulation Fair Disclosure; that same month, the so-called “Hash Crash” triggered a two-minute, 140-point drop in the Dow after the Associated Press’s Twitter account was hacked and a fake Tweet suggested that the White House had been hit by explosions; and finally, in August, activist investor Carl Icahn triggered a $12.5 billion jump in Apple’s market capitalization after he Tweeted the following message:
We currently have a large position in APPLE. We believe the company to be extremely undervalued. Spoke to Tim Cook today. More to come.
— Carl Icahn (@Carl_C_Icahn) August 13, 2013
Gnip points out that the adoption of social media data analysis in finance, as in the field of brand analytics, where the tools are more developed, depends on the emergence of a “base layer” of high-volume, high-quality social media content relevant to investing — and on which tools and applications can be built and integrated into existing investor workflows. Such activity is being driven in large part on Twitter by the adoption of the “cashtag,” the combination of a “$” and a ticker symbol (example: $GOOG) that users add to tweets about tradable equities as well as a growing range of financial instruments, including currencies, commodities, and futures.
Social finance has reached an inflection point, Gnip contends, “where costs and barriers to use have decreased, allowing new entrants the ability to gain an information advantage.” With new tools for uncovering meaningful social media signals rapidly maturing, the paper explains, “even early startups and small hedge funds are able to start exploring the patterns and insights in the data.”
A number of start-ups are racing to take advantage of heightened investor interest. Quantopian, which serves the algorithmic trading community, launched in April 2013 and rapidly crossed the 10,000 user mark, joining the ranks of other niche buyside communities for investment professionals, including SumZero, which went live in 2008 and now boasts a pre-screened membership of nearly 10,000 hedge fund, mutual fund, and private equity professionals. Thinknum, launched in December, is harnessing online collaboration tools to democratize financial modeling. Estimize, which offers crowdsourced earings estimates that are more accurate than the Wall Street consensus, is now aiming its model at M&A predictions with a new service called Mergerize. Similarly, a start-up called Premise is developing a model for crowdsourcing macroeconomic data.
Even the financial data giants are entering the fray: Earlier this year, Thomson Reuters added Twitter feeds and social media sentiment analysis tools to Eikon, its flagship desktop product. The move followed on the heels of Bloomberg’s own announcement last April that it had become the first financial data platform to integrate real-time Twitter feeds, at least a selection, anyway, “directly into the investment workflows of market professionals.”
In its paper, Gnip highlights three use cases for financial social media — equity sentiment analysis, breaking news discovery, and macroeconomic trend analysis — and two classes of social finance companies that are building analytics derived from the data. The first class focuses on social media monitoring. They “embrace social media as news and have built displays to show filtered and possibly enhanced social media content to keep users informed.” Thomson Reuters and Bloomberg’s offerings fall into this category, Gnip reports, as do companies such as Eagle Alpha, HedgeChatter, and Finmaven.
A second class of companies focuses on social media analytics. They “apply advanced analytics to create scores, signals, and other derived data from Twitter or other social media.” Among the companies cited by Gnip as examples: Social Market Analytics, Dataminr, and PsychSignal.
So, do crowdsourcing platforms and social data analysis tools help deliver alpha? A paper in the Review of Financial Studies suggests there is value to be mined. The authors found that the tone of blog articles published on Seeking Alpha, one of the largest investment-related social media websites in the United States, is a more accurate predictor of stock returns and earnings surprises than sell-side research and financial news articles. The paper builds on a finding, highlighted in a widely publicized 2010 paper, that Twitter can predict the short-term direction of the Dow Jones Industrial Average. Academic researchers are now looking beyond proving such correlations to focus on optimization techniques and new applications of social data to capital markets analysis, according to the Gnip paper.
Of course, social finance is not without risks. Crowdsourcing opens the doors to a diversity of opinion, and that may include recommendations from analysts with conflicts or whose motives are unclear. At the same time, a study of so-called “mirrored trading” on eToro, an online platform that enables investors to follow each other and copy trades, suggests that “social learning” improves investment decision-making only when individuals in a network each have different information. A lack of diverse idea flow can lead to overconfidence and groupthink.
With the base layer of social finance data increasing exponentially, academics and practitioners alike will have plenty of opportunity to mine for insights, fueled by an ever-growing roster of financial start-ups eager to topple the industry’s data giants.
Here are some of the notable papers exploring the impact and value of social finance:
- Wisdom of Crowds: The Value of Stock Opinions Transmitted Through Social Media
- Crowdsourcing Forecasts: Competition for Sell-Side Analysts?
- Exploiting Topic Based Twitter Sentiment for Stock Prediction
- Correlating Financial Time Series with Micro-Blogging Activity
- Predicting Asset Value Through Twitter Buzz
- The Role of Dissemination in Market Liquidity: Evidence From Firms’ Use of Twitter
- Do Fund Managers Identify and Share Profitable Ideas?
- Twitter Mood Predicts the Stock Market
At the 67th CFA Institute Annual Conference, the CEOs of three social finance companies, Leigh Drogen of Estimize; Joseph A. Gits IV, CFA, of Social Market Analytics; and Divya Narendra of SumZero, discussed crowdsourcing, online collaboration, and the risks and rewards of mining social data for investment insights.
If you liked this post, consider subscribing to the CFA Institute Annual Conference blog.
The content of this site should not be construed as investment advice, nor do the opinions expressed necessarily reflect the views of CFA Institute.
Photo credit: ©iStockphoto.com/retrorocket