IEM Demonstrates The Political Wisdom Of Crowds
What's really going on in the U.S. presidential election?
Is Rick Santorum a threat, and if so, to which party, especially given his weak Wednesday debate? That's the $1 question.
Why one dollar? Because in such situations I push aside my crystal ball, which turned out to be acrylic anyway, back in the Clinton years, and check out the Iowa Electronic Markets (tippie.uiowa.edu/iem). They're a practical effort by the University of Iowa business school to harness the wisdom of crowds on politics by requiring people to put real money where their mouths are.
The problem with public-opinion polls, on politics or anything else, is you can tell pollsters anything, including that your religion is Jedi (or, in Quebec, Catholic, the response of 83% though most couldn't find the local church if it were on fire). Likewise, you can say you support Newt Gingrich, but check Mitt Romney's name or not bother voting at all.
The IEM makes short work of all that nonsense. In it, you buy and sell contracts worth $1 if one of a wide range of political outcomes occurs. For instance, you can "bet" Barack Obama will be re-elected, Republicans will nominate Mitt Romney, the Democrats will win the Senate but lose the House and so on. But (like all futures markets) the IEM is no casino. Rather, once you register and deposit a maximum of $500, you offer to buy or sell contracts for whatever price you think reasonable and actually buy or sell them for real money if someone agrees.
The result is to harness a curious aspect of human intelligence described by James Surowiecki in The Wisdom of Crowds. If you offer a bunch of people a reward for guessing accurately, even though individual guesses tend to be quite poor for reasons from bias to stupidity, the errors tend to cancel out and produce a surprisingly accurate average.
If it's working here, what do we learn? For starters, Romney is still in the catbird seat.
The IEM's colourful price-history charts dramatize how early contenders including Michele Bachmann, Rick Perry, and Herman Cain flamed out and are now going for a tenth of a cent. But look how Ron Paul, now worth just under a cent and a half, never hit a dime (he peaked at 8.7 cents on Christmas Day); he was never really in the race. And Newt Gingrich, hovering around two cents despite bouncing slightly on Santorum's weak debate performance, should read 'em, weep and fold.
As for Rick Santorum (actually listed as RROF_NOM, as in Republican Rest of Field Nomination, because on the IEM's Aug. 30 opening day he was a no-name), Romney went into Wednesday's debate leading him 72 cents to 24, or three-to-one, and emerged around 75 while his boring rival sank to 22.
Does it matter who the GOP nominates? IEM says no; over in the presidential market they're betting the Democrat edges out the Republican in the popular vote but clobbers him in the electoral college—Barack Obama leads 60-40 in the "winner-take-all" market. But he'll probably face a Republican Congress—it's about 60 cents on the GOP having both houses, 16 cents on it having only the Senate and 15 cents on a solid Democratic Congress. (You can get a Republican Senate and Democratic House for six cents, meaning you can't get one.)
All of this makes short work of wishful thinking like my early backing of Herman Cain. But it still seems Republicans will swallow hard and nominate Mitt the Moderate, so I can continue to say Santorum is doomed because, like Senator Henry Jackson in his perennial runs at the Democratic nomination two political lifetimes ago, he's so dull if he gave a fireside chat the fire would go out.
Are such markets infallible? Of course not. Back in 2008, John McCain was leading until the financial crisis hit. But the fact that events can shake up political dynamics doesn't change the fact that real-money markets are way more reliable than public-opinion surveys.
So go ahead. Poll your heart out. Or wait for the Michigan primary results Tuesday, then log onto the IEM to see what the actual money says.