Big shifts in the Iowa Electronic Market
Kari Chisholm
It's been mentioned here at BlueOregon before, but I thought I'd draw your attention to the Iowa Electronic Markets. What is it?
The Iowa Electronic Markets are operated by faculty at the University of Iowa Henry B. Tippie College of Business as part of our research and teaching mission. These markets are small-scale, real-money futures markets where contract payoffs depend on economic and political events such as elections.
In short, real people put their own real money behind political candidates. Think Rudy Giuliani is going to be the Republican nominee? Buy shares of Rudy at 28 cents apiece, and if you're right, you'll make $1 per share. Think it'll be Fred Thompson? Just three cents buys you a share that'll pay off at $1.
With lots of people betting their money on various candidates, you get their cumulative sense of the likelihood that various candidates will win their respective nominations. As I wrote at Politics & Technology back on February 19, 2004:
The political world was stunned on January 19th when Howard Dean lost Iowa badly to John Edwards and John Kerry. If you had been paying attention to the IEM Political Market, you'd have seen it coming. ...How were the numbers in the run-up to the caucuses? From mid-November to early January, Howard Dean was trading at a peak of around 76 cents. Suddenly, on January 11 - eight days before the Iowa caucus - his fortunes plummeted, eventually all the way to 51 cents on the night before the caucus. John Kerry had been stuck for weeks at less than a nickel - but from January 12-18 tripled in value.
On the jump, the latest charts on the Democratic and Republican nominations.
First, the Democrats. Naturally, Obama is shooting up, Hillary is collapsing, and John Edwards remains mired.
Here's the latest chart on the Republican nomination. It's a lot more volatile. McCain is shooting up; Huckabee, Giuliani, and Romney are bouncing around; and Thompson is dead-man-walking.
It's not too late to put your money where your mouth is. Visit the IEM.
 |
More Recent Posts | |
Albert Kaufman |
|
Guest Column |
|
Kari Chisholm |
|
Kari Chisholm |
Final pre-census estimate: Oregon's getting a sixth congressional seat |
Albert Kaufman |
Polluted by Money - How corporate cash corrupted one of the greenest states in America |
Guest Column |
|
Albert Kaufman |
Our Democrat Representatives in Action - What's on your wish list? |
Kari Chisholm |
|
Guest Column |
|
Kari Chisholm |
|
connect with blueoregon
Jan 6, '08
Wow! It really IS a horse race! But thanks for sharing this interesting tool.
So, anyone think the Obama Girl's resurrection in time for the primary had an influence on the younger vote? I gotta admit, as a feminist, the video bothers me, especially when I discovered that it was created by a man, and that the model involved is just that -- a model -- not a writer or producer. So, sex for votes, not that it's new. I would feel VERY differently if the model herself had conceived of the idea and decided to use her talents to promote her chosen candidate. It is very clever, of course. Does anyone know the story behind "Barelypolitical.com"? They don't appear to focus only on this election.
This is why I like state and local races better -- fewer opportunities (or reasons) to use mass media in this way. Sigh.
Jan 6, '08
Sorry, that comment above should read "in time for the caucus" not "primary" -- I'm referring to events in Iowa.
Jan 6, '08
Kari,
The graph you show is a old one this graph is the newest IEM graph for the dem race with date from today. Looks a little different from ours but similar in one fashion, Edwards is way done.
Jan 6, '08
Great stuff, Kari. I can't believe how expensive Clinton was prior to Iowa. Inevitability she isn't! Anyone who's interested in exploring the concepts that drive the IEM should read "The Wisdom of Crowds" by James Surowiecki. Great read.
1:39 p.m.
Jan 6, '08
The charts go to October of last year and we are supposed to take them serious?
I would also not that these futures are almost always trailing indicators of polling, not leading.
1:42 p.m.
Jan 6, '08
My bad... the forced column width of BlueOregon's format truncated the right-hand portion of the charts.
Mea culpa.
The point still stands however that these futures markets are usually trailing indicators, not a predictive one.
Jan 6, '08
To me it is a bit distressing that the missions of the candidates and future mandates of the voters are brought so low....a gambling game with Jimmy the Greek style handy-capping. Seems unpatriotic and disrespectful of the long process and the exhausting efforts of those who really give a shit.
Way too early for these figures to mean anything. And after watching Edwards's passion and his message to middle America, I wouldn't discount him. The Iowa caucus should have shown the hacks, pollsters and odds-makers that voter apathy is over, and many many more citizens who've been disengaged in past elections will come out this year. And the long-time political junkies have no idea yet what pushes their buttons.
Polls mean nothing anymore because they are all biased and all are push polls to a certain extent. Surveys who's answers are skewed by the biased and limited range of the questions. (I was willing to take part in one recently....but quit when I saw how leading and tunnel-visioned the questions were.)
And here's a news flash: I'm probably not the only middle American who watched the entire evening of debates and listened to every word, watched for and understood every nuance. Nor am I the only one who engaged at this level for the very first time....in the past I've found the debate format and limitations too dry, dreary & predictable to commit to more than just a passing glance and a recap of the canned material the next day.
And I've been thinking lately that blueoregon would not be outside it's mission if it sponsored (or co-sponsored) a visible and vocal voter registration drive....to help ensure that those who've dropped out of the process in past years are registered to vote and be counted. They're out there. Blueoregon has a presence and a somewhat evolved readership, with plenty of energy for commenting here. Why not channel that energy...put some wheels on it? Do something good and worthy and more than just talking about it.
3:13 p.m.
Jan 6, '08
MCT wrote... Surveys who's answers are skewed by the biased and limited range of the questions. (I was willing to take part in one recently....but quit when I saw how leading and tunnel-visioned the questions were.)
MCT, sounds like you actually got a real candidate-driven poll, rather than a media poll.
This is one of those misconceptions that the public has about polling.
The media will ask you a horserace question - "If the election were held today, would you vote for X or Y?" - or perhaps a favorability question.
But campaigns don't care as much about that. Or, more to the point, that's only the beginning.
Campaigns care about actionable data. They ask questions to help determine the efficiency of various messages.
The "biased" and "tunnel-vision" questions you were getting were likely a campaign testing messages.
For example... "Joe Blow argues that the Congress should immediately implement a universal health care program for all Americans. Does that make you more or less likely to vote for him?"
Do that for a dozen different messages, and you'll see what moves voters - especially key segments of voters - in the furthest direction.
Jan 6, '08
I really do prefer intrade.com for this type of thing.
3:21 p.m.
Jan 6, '08
I've been following the IEM pretty closely, and was interested to see what Iowa did to it--then promptly forgot to look. Thanks for the update.
Interesting that Huckabee wasn't even in the market until December, isn't it. Still no love for him--and perhaps too much love for McCain. That market seems like a crapshoot.
Jan 6, '08
Guess I can save that stamp on my ballot this time around since its already been decided.
9:36 p.m.
Jan 6, '08
I can post some nice links to articles analyzing the IEM if people are interested. There is a recent critique of the market, comparing its performance to polling, written by Erikson and Wlezien.
The essential thing to keep in mind when looking at the market shares compared to polls is this: polls typically ask precisely what Kari posts above: "If the election were held today " (emphasis added), while the markets ask people to make projections to November.
The polls and the IEM are measuring fundamentally different things, and the polls have been and should be expected to be more volatile and responsive to events.
That being said, here's a nice money maker for you: invest in Dems before the convention and sell after, and same for Reps. There is always a convention boost in the market.
Jan 7, '08
Paul G,
As far as the Winner Takes All Presidential markets are concerned, it's not clear that one can exploit a convention-timed investment opportunity. Price graphs and daily quotes for the 2000 and 2004 WTA IEMs appear not, at least in general, to support the hypothesis of a "convention boost" in futures markets. If there is a boost in price come convention time (as occurred in 2004 BEFORE the Republican convention with a boost of something on the order of $0.04-5/share), it remains to be seen just how a convention and any price change correlate ... and/or whether any short-term speculation can overcome the bid-ask spread in this particular case.
9:35 a.m.
Jan 7, '08
The markets have an advantage and a disadvantage, Paul. The advantage is that they're agnostic: when you put your money down, you jettison sentimentality. It provokes more clear-eyed judgment.
The disadvantage is that it necessarily follows conventional wisdom. In the week before Iowa, as Obama got closer, the trend line followed, but still had Hillary winning. Now that the Iowa results and subsequent polls have come out, he's soaring. Again, that reflects what has happened, not what will happen. Should he stumble tommorrow, the trends will adjust to circumstances. That's how all markets function.
2:32 p.m.
Jan 7, '08
Exactly. It is a trailing indicator of what the "CW" currently is, not an accurate predictor of what it will be as events unfold.
3:01 p.m.
Jan 7, '08
Paul, please do post links if you've got time.
Is the unmarked blue line on the R graph Giuliani?
Which is what Mitch said, more or less, isn't it?
I can't find a "market" specifically on the Iowa caucuses at IEM, so their efficacy in capturing what was going on there is hard to see.
But in terms of this prediction of the nomination likelihood, Obama actually suffered a significant decline in judged likelihood (ca. 10%) in the two weeks before the caucuses. I think that probably means that the outcome was expected to have Clinton, Edwards & Obama all bunched up which would be favorable to Clinton on the whole. So this market got caught out by the caucus results.
Now it seems either to be giving Obama a huge bounce for New Hampshire, or saying that a relatively even outcome in New Hampshire is all he needs to be front-runner, and apparently that Clinton needs a large NH win to recover.
Was Iowa really as crushing for Clinton as the reversal of Jan 6 here suggests (Jan 2 Clinton 65¢, Obama 26¢; Jan 6 Clinton 32¢, Obama 67¢)? Or do "investors" in these markets operate in them to make money in the interim, i.e. betting not just on the final outcome but on likely price fluctuations between now and then, as would happen in real commodity markets?
In other words, what game are the investors playing (are they all playing the same game?). Do you win if you pick the right candidate, or if you make the most money along the way?
5:09 p.m.
Jan 7, '08