H-Yamaguchi.net

In the article, I tried to address an issue on General Election Hatena, the first political prediction market in Japan. Concerning the market, I wrote an article "First Political Prediction Market in Japan" before. Also Japan Media Review wrote articles on it (this and this).

The issue is the relationship between prediction markets and the Election Law. Since it was the first attempt, some people raised a question "Does it violate the Election Law?"

The article 138-2 of the law prohibits publishing the results of opinion polls on election results (who is going to win the seat, become the Prime Minister, and how many seats are occupied by the party X, and so on) during the election period. The article does not provide any explanation on the definition of "opinion poll," and some people were anxious whether Hatena violated the law or not.

I analyzed the minutes of a government-sponcered research group that studied the issue several years ago, and found that the key point discussed there was that opinion polls tended to be biased unless they were conducted under strict statistical procedure, i.e., random sampling. Politicians are anxious that people with some political interest might conduct intentionally biased opinion polls to affect public opinion.

This logic is hardly persuasive to many people including experts, but politicians many years ago introduced the article and it remains until now.

But this concern does not apply to prediction markets because prediction markets does not rely on random sampling for the following reasons:

(i) Predictions rather than opinions
In prediction markets, participants are motivated to predict the election result rather than expressing their own opinions, thus it does not rely on unbiasedness of the samples. Biasedness of participants does not bias the results, which is verified by the preceding experience at Iowa Electronic Markets.

(ii) Difficult to bias the market
In opinion polls, biases are more likely to remain because biased opinions are not affected by others. In contrast, in prediction markets the effect of trades based on biased opinions can easily be nullified by the following trades by other traders. Existence of liquidity traders, and even that of manipulators in general is rather helpful for the market mechanism to be functional.

Participants of prediction markets are more similar to the election experts (who predict the result) than respondents of opinion polls (who express their own opinions). The former is not prohibited by law, then the latter should be treated in the same way.

Therefore, prediction market is totally different from opinion polls. The concern on prediction market as above is based on the lack of knowledge on prediction markets. This technique is new and the theory has not fully developed, but there are increasing number of academic research and cases that support the idea (there is a short list of selected literature below.).

Despite the recognition above, there may be some way to bypass the potential disputes with respect to the Election Law because such disputes are totally counter-productive. For example, predicting poll result, rather than election result, may be bypass the law. Or, questions like "Is Party X be the government party after election?" or "Is the postal reform accomplished?" would be OK.

Suggested Readings
Berg, Joyce E., Robert Forsythe, F. Nelson, and Thomas A. Rietz (2000). "Results from a Dozen Years of Election Futures Markets Research." Handbook of Experimental Economics Results, C.A. Plott and Vernon Smith (eds.).
Forsythe, Robert, Forrest Nelson, George Neumann, and Jack Wright (1992). "Anatomy of an Experimental Political Stock Market." American Economic Review 82, 5: 1142-1161.
Forsythe, Robert, Thomas Rietz, and T.W. Ross (1999). "Wishes, Expectations and Actions: A Survey on Price Formation in Election Stock Markets." Journal of Economic Behavior and Organization 39, 1: 83-110.
Hanson, Robin D. (1999). “Decision Markets.” IEEE Intelligent Systems 14, 3: 16-19.
Hayek, Friedrich A. (1945). "The Use of Knowledge in Society." American Economic Review 35, 4: 519-530.
Malone, Thomas W. (2004). The Future of Work. Boston, MA: Harvard Business School Press.
Pennock, David M., Steve Lawrence, C. Lee Giles, and Finn Årup Nielsen (2001). "The Real Power of Artificial Markets." Science 291 (5506): 987-988.
Spann, Martin, and Bernd Skiera (2003). “Internet-Based Virtual Stock Markets for Business Forecasting.” Management Science 49, 10: 1310-1326.
Surowiecki, James (2004). The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business,Economies, Societies and Nations. Doubleday.
Wit, Jorgen, Charles R. Plott, and Winston C. Yang (2003). “Parimutuel Betting Markets as Information Aggregation Devices: Experimental Results.” Economic Theory 22, 2: 311-351.
Wolfers, Justin and Andrew Leigh (2002). “Three Tools for Forecasting Federal Elections: Lessons from 2001.” Australian Journal of Political Science 37, 2: 223-240.
Wolfers, Justin and Eric Zitzenwitz (2004). “Prediction Markets.” Journal of Economic Perspectives , Spring 2004.