Oxford University Press's
Academic Insights for the Thinking World

I miss Intrade

Autumn is high season for American political junkies.

Although the media hype is usually most frenetic during presidential election years, this season’s mid-term elections are generating a great deal of heat, if not much light. By October 13, contestants in 36 gubernatorial races had spent an estimated $379 million on television ads, while hopefuls in the 36 Senate races underway had spent a total of $321 million.

For those addicted to politics, newspapers and magazines have long provided abundant, sometimes even insightful coverage. During the last hundred years, print outlets have been supplemented by radio, then television, then 24/7 cable TV news. And with the growth of the internet, consumers of political news now have access to more analysis than ever.

One analytical tool that the politics-following public will not have access to this year is Intrade, an on-line political prediction market. Political prediction markets work very much like financial markets. Investors “buy” a futures contract on a particular candidate; if that candidate wins, the contract pays a set amount (typically $1); if the candidate loses, the contract becomes worthless. The price of candidates’ contracts vary between zero and $1, rising and falling with their political fortunes—and their probability of winning. You can see a graph of Obama and Romney contracts in the months preceding the 2012 election here.

Organized political betting markets have existed in the United States since the early days of the Republic. According to a 2003 paper by Rhode and Strumpf, during the late 19th and early 20th centuries wagering on political outcomes was common and market prices of contracts were often published in newspapers along with those of more conventional financial investments. Rhode and Strumpf note that at the Curb Exchange in New York, the total sum placed on political contracts sometimes exceeded trading in stocks and bonds.

Political betting markets became less popular around 1940. Betting on election outcomes no doubt continued to take place, but it was a much less high-profile affair.

Modern political prediction markets emerged with the establishment in 1988 of the Iowa Electronic Markets (IEM), a not-for-profit small-scale exchange run by the College of Business at the University of Iowa. The IEM was created as a teaching and research tool to both better understand how markets interpret real-world events and to study individual trading behavior in a laboratory setting. The IEM usually offers only a few contracts at any one time and investors are allowed to invest a relatively small amount of money. As of mid-October, the Iowa markets—like the polls more generally—were predicting that the Republicans will gain seats in the House and gain control of the Senate.

Wooden Ballot Box, by the Smithsonian Institution. Public domain via Wikimedia Commons.
Wooden Ballot Box, by the Smithsonian Institution. Public domain via Wikimedia Commons.

An important feature of political prediction markets—like financial markets—is that they are efficient at processing information: the prices generated in those markets are a distillation of the collective wisdom of market participants. A desire to harness the market’s ability to process information led to an abortive attempt by the Defense Advanced Research Projects Agency in 2003 to create the Policy Analysis Market, which would allow individuals to bet on the likelihood of political and military events—including assassinations and terrorist attacks–taking place in the Middle East. The idea was that by processing information from a variety disparate sources, monitoring the prices of various contracts would help the defense establishment identify hot-spots before they became hot. The project was hastily cancelled after Congress and the public expressed outrage that the government was planning to provide the means (and motive) to speculate on—and possibly profit from–terrorism.

Another, longer-lived—and for a time, quite popular–prediction market was Intrade.com. This Dublin-based company was established in 1999. At first, it specialized in sports betting, but soon expanded to include an extensive menu of political markets. During recent elections, Intrade operated prediction markets on the presidential election outcome at the national level, the contest for each state’s electoral votes, individual Senate races, as well as a number of other political races in the US and overseas. Thus, Intrade offered a far variety of betting options than the IEM.

Intrade was forced to close last year when the US Commodities Futures Trading Commission (CFTC) filed suit against it for illegally allowing Americans to trade options (by contract, the IEM secured written opinions in 1992 and 1993 from the CFTC that it would not take action against IEM, because of that market’s non-profit, educational nature). The CFTS’s threat to Intrade’s largest customer base very quickly led to a dramatic drop-off in visitors to the site, which subsequently closed. Alternative off-shore betting markets have entered the political markets (e.g., Betfair), but their offerings pale by comparison with those formerly offered by Intrade and are probably too small at present to spur the CFTC to action.

I regret the loss of Intrade, but not because I used their services—I didn’t. Given the federal government’s generally hostile view toward internet gambling, I felt it was prudent to abstain. Plus, having placed a two-pound wager on a Parliamentary election with a bookmaker when I lived in England many years ago convinced me that an inclination to bet with the heart, rather than the head, makes for an unsuccessful gambler.

No, I miss Intrade because it provided a nice summary of many different political campaigns. Sure, there are plenty of on-line tools today that provide a wide array of expert opinion and sophisticated polling data. Still, as an economist, I enjoyed the application of the mechanisms usually associated with financial markets to politics and observing how political news generated fluctuations in those markets. No other single source today does that for as many political races as Intrade did.

Feature image credit: Stock market board, by Katrina.Tuiliao. CC-BY-2.0 via Wikimedia Commons.

Recent Comments

  1. […] that the US mid-terms are over, read my latest Oxford University Press blog post on political prediction […]

  2. Adam Siegel

    Good post. Intrade indeed provided a nice summation of all the polls and opinions out there about the election in a helpful, probabilistic driven model. I realize conventional wisdom says real money markets are the most reliable indicators, but you may still want to look at some alternative prediction markets in the future.

    I founded Inkling @ http://home.inklingmarkets.com so of course I’ll mention that one :), but there are others if you google for them.

    A research project I suspect you’d be interested in as well if you haven’t heard about it already is the ACE project run out of IARPA in the intelligence community: http://www.iarpa.gov/index.php/research-programs/ace This project has shown that through other primary incentive mechanisms you can drive highly accurate geopolitical forecasts.

  3. […] This article originally appeared at the OUPblog.  […]

  4. Jose Olivas

    There is a new site that is almost in every way similar to Intrade that is pretty new. It’s a prediction market and they mostly cover elections (folks can buy shares and sell them, just like they could in Intrade.) The site is called PredictIt (http://www.predictit.org)

    It’s run out of Victoria University in New Zealand and apparently it’s all above board (even moreso than Intrade was). Could be what you’re looking for, Professor Grossman.

  5. Luana Lopes Lara

    Dear Professor Grossman,

    Great post! My name is Luana and I’m an MIT graduate and the co-founder of Kalshi, a Y Combinator funded company. Kalshi is a prediction market game, where users play for free but get prizes according to their PnLs. I’d love to talk to you about Kalshi and InTrade.

Comments are closed.