Prediction markets and Obama Vs Romney

Four years, I posted a blog on the Iowa Electronic MarketsUS Presidential Election Market.  (For more explanation of the market go to this page.)  For much of time from his nomination in August 2008, it looked very clear from the winner-take-all market that President Obama was going to beat Senator McCain.  Recall that from early September 2008, when the winner take-all market was trading at nearly 50:50, President Obama’s star started an ascendency of such strength that by the date of the 2008 election on 4^th November, the market was trading 90:10 in favour of Obama.   This means that it cost 90 cents to place a bet that the Democrats would poll the majority of votes cast, which would return $1 if this transpired and $0 if not.  To coin a phrase: this time is different (to some extent!). 

In 2012, a similar divergence in the winner-take-all (WTA, the upper chart) market started this summer and continued until late-September when by close of trade on 27 September, WTA had 82:18 for President Obama.  But rather than continuing to cascade towards Obama the prediction market went back to 58:42 by 24^th October: the US public seemed to waver and Mitt Romney built up considerable momentum.  In the last week of October, much was therefore made of an election that was going to be too close to call - which seems to be even more so the case if you look at the Vote Share market (VS, the lower chart).  

But since Hurricane (Cyclone) Sandy the prices in the WTA market have gone back to favour Obama and on Sunday close 4^th November and we stood at 72:28 in favour of the incumbent.  It seems that Iowa prediction markets are calling it for the President Obama, or at least that he will get the majority of the popular vote.  Unfortunately prediction markets have not taken off in the way that I had hoped a few years ago and the Iowa markets are still far from deep and liquid with a maximum bet of $500.  But the ability for market participants to trade assets in a manner to capture the probability of different states of the future remains of great use.  At the very least to someone over here in the UK, who cannot follow every turn in the US news - and yet from this market, I get a clean probability on a daily basis.  At the least we should probably ignore opinion polls and place a bit more weight on the implied probabilities from markets in which people place real money on their views.  A surprise may, of course, await us but at least we do not have very long to wait.