History and Evolution (Author: Jen Heinzen Krueger)

This chapter will serve to explore the development of the information markets over the past few decades. Such markets have been available online to the general public since the mid-1990s, in both real-money (gambling) and play-money (game) formats, and a few have developed large communities of regular traders (Servan-Schreiber, Wolfers, Pennock, & Galeback, 2004). These markets appear to be very efficient and any inefficiencies appear minimal and go away with publicity, this could be a result of the profit opportunity (Hanson, 1999). In the past, the cost of handling trades was sufficient to limit the concept speculative markets since they could not be made widely available (Hanson, 1999). However, these markets are gaining increased popularity. These markets are making it possible to collect feedback from customers, voters and employees in a way that is fun, engaging, and, based on the recent growth, effective.

There are two types of information markets we will look at from a historical perspective. We will look at real-money exchanges, some that are popular with the public include the Iowa Electronic Markets (http://www.biz.uiowa.edu/iem) and Trade Sports (http://www.tradesports.com), a betting exchange headquartered in Ireland (Servan-Schreiber et al., 2004). As we explore the history of the information markets, we will be looking at an information market, as transitionally constructed, it is a market created for the purpose of extrapolating predictions and information from share prices (Abramowicz, 2004). We will also examine play-money markets some of which include: The Hollywood Stock Exchange (http://www.hsx.com), which focuses on movie box-office returns, News Futures’ World News Exchange (http://us.newsfutures.com) whose sports and financial markets are operated jointly with USA Today, and the Foresight Exchange (http://www.ideosphere.com), which focuses on long-term scientific discoveries and some current events (Servan-Schreiber et al., 2004).

The Origin of Information Markets

The origin of Information Markets may be traced back to the Iowa Presidential Stock Market. The Iowa Presidential Stock Market was initiated in April of 1988 and opened on June 1, 1988 for trading (Forsythe, Nelson, Neumann, & Wright, 1992). This original experiment, allowed traders to pay 2 1⁄2 cents for each percentage point of the popular vote in the presidential election won by Bush, Dukakis or others (Wolfers & Zitzewitz, 2004). Interestingly, the results of this “play” market, yielded results that were very similar to the actual election. The candidates in the election included George Bush, Michael Dukakis, Jesse Jackson, as well as independents. The Iowa Presidential Stock Market included four candidates. The market included George Bush, Michael Dukakis, Jesse Jackson, and a candidate labeled "Rest of Field" that would cover all other candidate’s votes in the election (Forsythe et al., 1992). The Iowa Presidential Stock Market was operated as a computerized double auction market. To enter the market, traders would be charged a portfolio investment fee to obtain their individual account. That individual account enabled them to participate in the market. Participants traded in this market by issuing bids (offering a price to buy stock) and asks (asking a price to sell stock). This particular market included 192 participants as traders and those traders held a collective 1462 shares (Forsythe et al., 1992). This market had a very exciting outcome. The market forecasted that 53.2% of voters would vote for Bush, 45.2% of voters would vote for Dukakis and 2% would vote for third-party. These results were very close the election outcomes (Forsythe et al., 1992). This is a fantastic realization for this initial run of the IEM. One may attribute its future success to the realistic outcome this first market yielded.

The faculty at the University of Iowa developed the IEM to be an Internet-based tool for research and for teaching (IEM Media Kit). The IEM allows students to invest real money ($5.00-$500.00) and to trade their money in a variety of contracts (IEM Media Kit). Students have the opportunity to trade in many contracts whose eventual payoff is dependant on a future event. Some of the future event include: corporation's stock returns quarterly earnings, economic indicators, or a movie's box office receipts (IEM Media Kit). Since its inception over ten years ago, over 100 universities from several countries have enrolled in the IEM (mainly large research-oriented institutions such as Harvard, MIT, Michigan, and Northwestern) (IEM Media Kit). The IEM is used to teach several subjects including business, economics, and technology (IEM Media Kit).

The Iowa Electronic Markets is the only continuously operating information market that uses real currency in the United States (Abramowicz, 2004). The currency traded in this market does not serve as a claim to ownership; instead, it is a claim to payoffs that are dependant on the occurrence of a future event that the trader is bidding on. Most of the information markets run by the Iowa Electronic Market are related to the results of political elections, however, there are a few nonpolitical markets (Abramowicz, 2004). One distinguishing feature of the Iowa electronic markets is that the value of all securities in any given market collectively add up to one dollar (Abramowicz, 2004). There are two types of election markets: vote-share markets and winner-take-all markets (Abramowicz, 2004). In a vote-share market, a security corresponds to a particular candidate or political party and pays off the number of cents equal to the percentage of votes obtained. In any winner-take-all market, a security corresponding to a candidate or party pays off if and only if that candidate or party receives the most votes in the election (Abramowicz, 2004).

The Iowa Electronic Market is presently still in operation and demonstrating a high level of accuracy. For example, with the 2000 presidential election information markets, the winner-take-all market's prediction on election eve that Bush had a 70% chance of winning the popular vote, which roughly corresponds to what we might have expected based on the evidence at the time (Abramowicz, 2004). Additional political markets include: the 2003 California gubernatorial election, the 2004 presidential election, the 2004 Democratic presidential nomination and how the Federal Reserve will alter the federal funds interest rate (Wolfers & Zitzewitz, 2004).

Around the same time the Iowa Presidential Stock Market was launched, the Ideas Futures (IF), now known as the Foresight Exchange was beginning. The Ideas Futures is an implementation of Robin Hanson's Idea Futures concept; a market in which people invest for and against claims about future events (Kittlitz, 1999). The history of the Foresight Exchange can be traced back to the late summer of 1994. It is in 1994 that the members of Daemon Maxwell reportedly became interested in Idea Futures (Kittlitz, 1999). It was Sean Morgan and Mark James that developed a prototype if an information market, or an Idea Futures market, for the internet (Kittlitz, 1999). The market designed by Sean Morgan and Mark James used HTML pages for its interface and the system let people complete trades in three different categories (Kittlitz, 1999). Those categories included: Science, Math and Miscellaneous (Kittlitz, 1999). The first trades occurred in late September of 1994 and not long after the market included mailing lists, discussion groups and was experiencing increased user traffic (Kittlitz, 1999). The increased usage of the sight began to pose a problem for the infrastructure of the Idea Futures market which was leading to wait times and frustration (Kittlitz, 1999). It was as a result of those challenges that the development of the Foresight Exchange began in the fall of 1995 (Kittlitz, 1999).

The FX (Foresight Exchange) is a market that does not utilize real money. The Foresight Exchange uses a form of play-money referred to as the “credibill"(Kittlitz, 1999). For each player in the FX, a score is recorded to track progress of the player. The concept grew quickly and by June of 1996, the game had over one thousand users (Kittlitz, 1999). Some even consider the FX to be the first exchange.

Another market utilizing a fake currency is the Hollywood Stock Exchange (Abramowicz, 2004). The Hollywood Exchange was founded in 1996. The company was acquired by Cantor Holdings in May of 2001. The Exchange allows for trading on security is corresponding to movies, including both those in production and those in theaters, with each movie security liquidated four weeks after the movie release for one dollar per million dollars in the box office gross (Abramowicz, 2004). This market appeals to a different audience that the previously mentioned markets. The Hollywood Stock Exchange allows people to use virtual currency to speculate on movie-related questions like opening weekend performance, total box office returns and who will win Oscars (Wolfers & Zitzewitz, 2004). This idea is very appealing to those interested in the entertainment industry. The benefits for the industry itself are paying off through the ability to forecast sales at the box office before a film is released and based on the participation, movie goers enjoy the opportunity to voice their opinions.

Many of the information market companies are now appealing to wide audiences. For example, The Trade Exchange Network owns and operates TradeSports.com, Intrade.com, and TradeBetX.com (Executive Statement).

TradeSports.com, a widely-known market offers contracts and number of areas including sports, politics, finance, Law, entertainment, and even the weather (Hahn & Tetlock, 2005). The Trade Exchange Network was established in 2001 (Executive Statement). The Trade Exchange Network advertises that they launched the first real-time, trading platform that was person to person and incorporated trade matching, account management, and settlements in 2001 (Executive Statement). According to the Trade Exchange Network’s homepage, they have members participating from over 127 countries worldwide (Executive Statement). Not only do they have participation from a wide range of countries, they have a very large volume of trading. The Trade Exchange Network processes over 750,000 orders per day (Executive Statement).

Another new market entry with wide appeal is from Goldman Sachs. Goldman Sachs exchange is called economicderivatives.com, which hosts call auctions for securities based on economic indices (Hahn & Tetlock, 2005). Goldman Sachs and Deutsche Bank have launched markets on the likely outcome of future readings of economic statistics, including employment, retail sales, industrial production, and business confidence (Wolfers & Zitzewitz, 2004).

Another application of the concept of information markets has been adopted by companies for internal decisions and information. Hewlett-Packard employees information markets for forecasting sales in various financial and accounting results (Ostrover, 2005). Hewlett-Packard commissioned Charles Plott, and the economist from the California Institute of Technology, to set up a software trading platform (Kiviat, 2004). HP’s market started with their own employees, mostly product managers and finance managers. They provided each employee with about $50 in a trading account to bet on what they thought computer sales would be at the end of the month (Kiviat, 2004). The results of this test were astounding. In the firm's first pilot, the market beat official sales forecasts 75% of the time (Ostrover, 2005). The results from this market where only 13% off the official forecast and only 6% off in the stock market forecast (Kiviat, 2004).

Another company utilizing these markets internally is Eli Lilly, the pharmaceutical firm. Eli Lilly used information markets to identify the drugs in early stages of development most likely to win regulatory approval (Ostrover, 2005). The first pilot correctly identified the three most successful drugs and bid data was analyzed to provide insights about the perceptions of participating employees including, chemists, biologists, and project managers.

Both of the internal applications demonstrate the value of information markets in the private setting. While they do not use their markets outside of the company’s employees, there has been a large return on their efforts.

Conclusion

In most prediction markets, the mechanism that matches buyers to sellers is a continuous double auction, with buyers submitting bids and sellers submitting asking prices, and with the mechanism executing a trade whenever the two sides of the market reach a mutually agreeable price (Wolfers & Zitzewitz, 2004). These markets have a wide application to our ability to understand business trends and forecast success of products, outcomes of elections and many other decisions.

Based on the current growth of prediction markets, it is clear that their uses are being further explored by companies for internal and external use, by political parties to understand potential outcomes, and by colleges and universities for research purposes. The volume of trades being placed in these markets on a daily basis is a clear indication that the growth history could very well continue over the next few years.

Abramowicz, M. (2004). Information Markets, Administrative Decisionmaking, and Predictive Cost-Benefit Analysis. The University of Chicago Law Review, 71(3), 933-1020.

Executive Statement. Retrieved March 25, 2006, 2006, from http://tradesports.com/aav2/TEN/TENstatement.html

Forsythe, R., Nelson, F., Neumann, G. R., & Wright, J. (1992). Anatomy of an experimental political stock market. The American Economic Review, 82(5), 1142-1161.

Hahn, R. W., & Tetlock, P. (2005). Using information markets to improve public decision making. Harvard Journal of Law and Public Policy, 29(1), 213-288.

Hanson, R. D. (1999). IEEE Intelligent Systems & Their Applications. Los Alamitos, 14(3), 16-20.

IEM Media Kit. Retrieved March 19, 2006, 2006, from http://www.biz.uiowa.edu/iem/media/summary.html

Kiviat, B. (2004, July 12, 2004). The end of management? Time, 164.

Ostrover, S. (2005). Employing information markets to achieve truly collaborative sales forecasting. The Journal of Business Forecasting, 24(1), 9-12.

Servan-Schreiber, E., Wolfers, J., Pennock, D. M., & Galeback, B. (2004). Does Money Matter? Electronic Markets, 14(3).

Wolfers, J., & Zitzewitz, E. (2004). Prediction Markets. Journal of Economic Perspectives—, 18(2), 107-126.