University Students Gamble in Presidential Stock Market
Nov. 11, 1988
NEW YORK (AP) _ A presidential ''stock market'' in which professors and students gambled cash to buy and sell shares of the candidates accurately reflected public opinion and correctly predicted the election's outcome, its creators say.
The stock traders' profits and losses had not been tallied by Friday, but one trader claimed to have earned a 30 percent profit and another said he had lost $40 of $150 wagered, said Forrest Nelson, one of four University of Iowa researchers who set up the market.
Most of the 168 traders' profits and losses were probably less than a few dollars, Nelson said.
Nelson and economists Robert Forsythe and George Neumann and political scientist Jack Wright set up the market to see whether it might do a better job of registering public opinion than polls do.
''In theory it's more accurate (than polls) because of the financial incentive people have to put their money where their mouth is,'' said Forsythe.
Prices of the candidates' shares could be converted easily into numbers showing what percentage of the public favored each candidate, Forsythe said.
Since shortly after the second debate on Oct. 13, the stock market showed Republican George Bush with between 51.2 percent and 53.6 percent of the vote, and with a margin of between 5.2 percent and 8 percent over Democrat Michael Dukakis. Bush won with 54 percent of the vote and an 8 percent margin in Tuesday's presidential election.
''Starting Oct. 17 the market has been predicting about this outcome all along,'' said Forsythe. ''If you look at the polls over that period, they've been a heck of a lot more volatile.''
Gallup polls, for example, gave Bush a 6 percent advantage early in October, a 10 percent advantage near the end of October and a 9 percent advantage just before the election, said Coleen McMurray of the Gallup Organization in Princeton, N.J.
The market differs from polls in that it does not reflect the traders' preferences, but rather their impressions of how well each candidate is doing.
Each stock trader is, in effect, a private pollster, surveying friends and following news reports.
The market prices respond to the information traders pick up. Economic theory predicts that because the traders aren't being asked their preferences, they don't have to be a representative sample of the population to accurately reflect public opinion.
The researchers are far from proving that the market is an effective means of surveying public opinion, but they were encouraged when the market did not show the wild swings in public opinion that were registered in polls.
Jack Ludwig, chief methodologist of the Gallup Organization in Princeton, N.J., said an educated group like the Iowa stock traders ''would be less likely to show the kind of swings in popularity that the candidates do on a general population basis.''
Ludwig also said he didn't believe that the small sample of white, educated Midwesterners could be used to estimate public opinion nationwide.
A potential advantage of the market over opinion polls is that it provides a continuous record of public opinion, Forsythe said.
John Eighmey, a senior vice president and market survey expert at the Young and Rubicam advertising agency in New York said he had seen the market demonstrated and found it ''really quite intriguing and impressive.''
''It's one thing to ask a person, 'What do you think about this subject?' It's quite another when they have a personal stake in the answer,'' he said.
Forsythe said the market initially confounded the university's lawyers. ''Our lawyers said, 'You either fall under state gambling laws or state securities laws, and we're not sure which.'''
The lawyers finally decided that gambling laws applied and the university would be exempt if participation was limited to university faculty, students and staff.