Internet company Google Inc., already provides its employees with free, unlimited chef-prepared food, on-site car washes, lap pools, haircuts and free doctor check-ups, and now the company is encouraging its workers to gamble while at work with the help of Dartmouth economics professor Eric Zitzewitz.
In the last two and a half years, 1,463 Google employees have participated in what the New York Times called "Google's Lunchtime Betting Game," placing bets on the future of the company and related markets in order to generate information for future company decisions.
Google encourages employees to place bets online in prediction markets concerning the company's products and services. For example, employees might be asked: How many users will Gmail, Google's e-mail server, have at a given point? Or, will Google open an office in Russia?
Google also asks employees to place bets on external markets of interest to the company, responding to questions like: Will Apple release an Intel-based Mac? These questions are mixed in with questions on "fun markets," which include queries about movie ratings, gas prices and sports.
Participants trade in the market with Goobles (as in rubles), which they accumulate and can trade in for small prizes. The company distributes a total of $10,000 in prizes per fiscal quarter.
Most participants are the company's programmers, Zitzewitz said, but all employees are able to get involved.
Zitzewitz is one of two outside economists who are using employee's betting decisions to study how information flows through the company.
"It's all about getting the knowledge out of the organization," Zitzewitz said. "Google uses these bets for its own planning. It helps them make decisions."
Zitzewitz's research focuses on how organizations like Google process information, demonstrated by the employees' betting patterns.
The study, which was published in conjunction with University of Pennsylvania professor Justin Wolfers and Google economic analyst Bo Cowgill, revealed distinct trends in the way information courses through Google. For example, employees sitting next to each other tend to bet more similarly than friends or co-workers in other rooms or buildings.
"Only 100 feet away, it's like a different city," Zitzewitz said.
The study also found that newly hired employees show a notably optimistic bias in their bets, particularly on days when Google stock is appreciating.
Other companies are beginning to use similar prediction markets, Zitzewitz said.
"Best Buy asked their employees questions like 'What TV will be the hottest seller this Christmas season?'" he said. "GE has played around with it, and Microsoft is also exploring these markets."
According to Zitzewitz, there seems to be a big future for prediction markets and similar schemes to make use of information employees can provide their companies for future decision-making.
"There is a lot of information down in the company which the manager never hears about," Zitzewitz said. "There are always conversation around the water cooler, but when the boss shows up it stops, and a lot of this knowledge can be used."
