With its widely anticipated initial public offering, Facebook is seeking a market value of as much as $100 billion. If successful, the IPO would yield the company a higher price-to-earnings multiple than virtually every other company in the Standard & Poor’s 500 index. So why such investor enthusiasm for a company with fewer than 2,500 employees worldwide? According to Brian Uzzi, professor of leadership at Northwestern University, the answer comes down to the emerging science of networks. In a session at the 65th CFA Institute Annual Conference in Chicago, Uzzi dissected the network effect: he explained how personal networks can be not only key assets for investment professionals but also a source of collective intelligence about financial markets, which, if properly harnessed, can help money managers make better investment decisions.
Uzzi began his talk by differentiating between two types — one akin to a modern-day Paul Revere, and one more like William Dawes. Revere, of course, was the American patriot who traveled from Boston, Massachusetts, by horseback on April 18, 1775, to alert colonists that the British were coming. His message spread far and wide, and Revere is credited with helping to raise the Continental Army that ultimately defeated British forces. His role in the American Revolution is memorialized in a Henry Longfellow poem titled “Paul Revere’s Ride.”
Dawes, by contrast, is the other messenger that fateful night — the one who rode into oblivion. Even though he galloped on horseback all night and into the next morning, longer than Revere, he was wholly ineffective in spreading his message. It wasn’t because Revere was better educated or more economically advantaged than Dawes. What accounted for the difference in their effectiveness, Uzzi argued, was the quality of their networks. And what was true of the midnight ride is also true in the contemporary world of business: there is something crucial about “how networks take your knowledge and skills and raise them to the power of who you know,” said Uzzi, who has studied network effects through such varied lenses as Broadway musicals, email communication, financial performance, and scholarly collaboration.
So are you a Paul Revere or a William Dawes? You can figure this out by analyzing your network. What you need to know is how many of your professional contacts are also contacts of each other. The more your contacts consult each other for professional advice, as a fraction of your overall contacts, the more likely you are to be a William Dawes. Uzzi cited a study suggesting that the average U.S. business person has a network in which 8.2 out of 10 individuals are contacts of each other. In other words, most people are more like William Dawes.
The problem with William Dawes-type networks, Uzzi went on to explain, is that they exhibit a high level of clustering. These clusters are in effect “echo chambers,” which can be detrimental to leadership because they provide a myopic view. When you’re inside the echo chamber, he explained, you’ve most likely confirmed your existing point of view — but you haven’t improved the accuracy of your information.
This observation leads to a key takeaway from network analysis: To get out of the echo chamber, you must aspire to be someone who bridges otherwise disconnected networks. Uzzi calls these individuals “brokers,” and they exist in all types of contexts, from corporate boards to high school dating networks to maps showing the transmission of typhoid and SARS. Brokers are the ones who spread ideas (or diseases) and make products successful. They are not necessarily the most charismatic individuals, as one might expect, but rather the people who come into contact with a wide cross-section of different types of people, typically through shared activities, and thus see the world in “much more of a panoramic view,” said Uzzi. Facebook’s lofty valuation can be explained in part by the fact that the site can identify who the brokers are, he said.
In the second part of his presentation, Uzzi turned to sentiment analysis of trader networks, a budding area of research that has attracted a lot of attention from financial professionals — and has even inspired the launch of a hedge fund. Uzzi argued that it is possible to “decode” the networks formed by traders within a firm to derive predictive utility from words contained in their instant messaging chats. In one such study, Uzzi and three colleagues analyzed the content of 3 million IMs within a trading firm in one-hour blocks, scanning for what he called “arousal words” — those likely to trigger an action. They found that arousal was positively associated with trading volatility net of market variables.
In another study cited by Uzzi, he and his colleagues looked at instant messages from traders at a firm over a 40-month period. They found that certain word occurrences conveyed traders’ understanding of both same-day and next-day volatility. One network of words predicted short-term (same-day) volatility; the other predicted future (next-day) volatility. With investors continuing to grapple with heightened stock market gyrations, the quest to turn such insights into money-making strategies is sure to continue.