User participation in a corporate prediction market

• Develops theory-based hypothesis generated in open financial markets.
• Investigates patterns of corporate prediction market use over time.
• Isolates particular characteristics of use of a corporate prediction market.
• Compares the behavior of two different groups in a corporate prediction market.
• Finds day of the week, and other market effects.

Corporate prediction markets allow companies to use external market concepts to facilitate and support corporate decision making. Recently, Google, Microsoft, GE, Best Buy, and other firms have generated and used prediction markets as a means of gathering the “collective” intelligence of their employees. Since these markets capture and aggregate information from employees and ultimately provide information for decision making, some researchers have referred to them as decision support systems or group decision support systems.Unfortunately, there has been limited theory development and empirical investigation of participation in corporate prediction markets. Accordingly, the purpose of this paper is to use theory generated about external investment markets to investigate participation behavior in an internal corporate market.Analysis of the number of unique traders by date and market leads to a number of findings, including that market traders apparently trade on specific information, there is a day-of-the-week effect of their participation, and participation is decreasing over time. Understanding the existence of such effects is important because they can influence the ability of the market to provide sufficient, timely, and quality decision support information.