Go to page 2, 3, Next       

Dougal Williams is a Senior Portfolio Manager with Vista Capital Partners, a fee-only investment advisor based in Portland, Oregon.  Vista specializes in managing globally-diversified portfolios of low-cost, tax-efficient index funds for individual clients with more than $1 million to invest.

Can skilled investors outperform the market?  What’s the best way to achieve long-term investment success?  The answers to these age-old questions become clear after taking a close look at how markets work, exploring the dynamics of group decision-making, and examining how the simple concept behind Google’s search engine relates to personal investing.

Mission: Faster and More Accurate

In 1995, two Stanford University graduate students embarked on a mission to solve one of computing’s biggest challenges:  retrieving relevant information from a massive set of data.  The massive set of data they were trying to make sense of was every piece of information contained on the Internet. 

By now we know how their mission turned out; within a few years of founding Google, Larry Page and Sergey Brin’s creation was the most frequently used search engine on the Internet.  Close to half of all Internet searches performed are now done via Google’s search engine.¹  The company name has become so widely used, it has been added to Merriam-Webster’s roster of official words (as in “to Google” or “search for” information on the Internet).   Since their August 2004 debut, shares of Google stock have surged more than seven-fold, making Page and Brin perhaps the world’s first “Googillionaires.”
 
So how does Google do it?  Simply by doing a better job of finding the right web page more quickly than any other search engine.  With each Internet search, Google is essentially asking the Web to “vote” for the pages containing the most correct and useful information.  This information is sorted, indexed and continuously updated in order to ensure accuracy.  As a result, the web page receiving the most votes tops the list.  And more often than not, that web page, or the one immediately below it, is exactly the one you’re looking for. 

Markets Work

The world’s financial markets, like Google’s search algorithms, are also complex systems designed to aggregate massive amounts of data.  In the stock market, information is communicated through orders to buy and sell securities.  The exchanges—such as the New York Stock Exchange—aggregate this information and match buyers and sellers at the market clearing price.  At any moment, this market price should be the most accurate estimate of a security’s true value, since all known, publicly-available information is embedded in that price.  If anything more were known, someone would take advantage of it and the price would change.

The beauty of financial markets is how efficiently they work.  Markets dynamically and accurately process an enormous amount of information. A group of investors, for example, may have opposing views about the “true” value of a stock’s price, but their collective opinion as a group is most often the Google's success is built on the basic premise that collective wisdom is often very accurate. best estimate of a stock’s value.  Why?  Because each investor’s estimate contains some accurate information and some error.  When the stock market’s thousands upon thousands of transactions are processed and aggregated (thus turning private judgments about a stock’s price into a collective decision by the market) the errors will tend to cancel out.  Strip out the error and only information is left.  And based on studies of the reliability of market-based decisions², this remaining information has been shown to be incredibly accurate.³

Markets work when they possess 1) multiple agents of diverse and independent opinion, 2) an incentive for participation, and 3) some mechanism for aggregating information.  All markets, whether for chewing gum on the school playground, for the future delivery of wheat orfor the information contained on the Internet, operate under the same three conditions.  The greater the number of agents, the more diverse their opinions, the greater the incentive for participation and the more advanced the aggregating mechanism, the better that market will function.  The stock market is the consummate example.  We can think of Google as the New York Stock Exchange of the Internet. 

Collective Wisdom

The remarkable intelligence of groups when it comes to decision-making was first demonstrated in experiments conducted by sociologists and psychologists in the early part of the twentieth century.  In a very simple study, students in a class were asked to estimate the room’s temperature and an average of the guesses was taken.  The group guessed 72.4 degrees, while the actual temperature was 72 degrees.⁴  A follow-on study asked two hundred students to rank various items by weight.  The group’s “estimate” was found to be 94% accurate, better than all but five of the individual guesses.  The most classic test of group intelligence may have been performed by Jack Treynor (co-creator of the Capital Asset Pricing Model with Nobel prize-winner Bill Sharpe).  Well known for his academic rigor and complex thinking, Treynor often used more simple methods to test collective wisdom.  In one finance class, he asked his students to independently guess the number of jelly beans in a glass jar which contained 850 beans.  Of the 56 students in his class, only one made a better guess than the group’s average of 871.

To be sure, these studies would hardly stand up to scientific scrutiny.  The point to be drawn, however, is not that the group always arrives at a better answer than the smartest individual member of the group.  No, some individual members should be expected to do better than the group.  This is a good thing, since the mere opportunity to outperform the group is what provides an incentive for people to do well, particularly in an environment such as the stock market (if the stock market worked perfectly, no one would have any incentive to uncover the information that becomes so quickly reflected in market prices).  The point is this:  while one or two students might outguess the group each time, it is highly unlikely to be the same two students each time.  In fact, the best way to get a reliable estimate of the contents of the jar—time after time—is not to seek out the smartest individual, but to just ask the group.

Display article as PDF for printing.

Would you like to send this article to a friend?