2011 predictions

iPredict's vote share markets ought to be scaring National.

There are eight potential states of the world, sorted by whether ACT, New Zealand First, and United Future make it back into Parliament. iPredict has contracts on each, assuming that the very most likely route into Parliament for them is winning a seat - no chance any crosses the 5% threshold. Here's some first cut analysis, based on the highly dubious assumption of independence across contracts that lets me ignore covariance, and based on simple rounding rather than using the quota system for list allocation.

The most likely state of the world, 26.7% by current prices, sees ACT returned but neither United Future nor New Zealand First. In that state of the world, National and ACT together see 60 seats, based on current prices. But the Maori Party gets 5 seats (overhang of one - expected outcome given prices on contracts in the Maori Seat markets). So even in that favourable state of the world for National, the Maori Party could give the government to a Labour/Green alliance (56 seats plus 5).

Next most likely has none of the minor parties return: 21.2% In this case, National at 58 seats ties Green + Labour; again, Maori makes the government.

There's a 12.4% chance that ACT and UF make it in but not NZ First. National's coalition, assuming UF stays with National, makes 60 seats to Labour/Green's 55. Maori could induce a tie, but that seems pretty unlikely. This is the best state of the world for National.

Next, an 11.4% chance of ACT and NZ First without UF. National plus ACT total 58; Labour plus Green 54, NZ First gets 4 and Maori 5. I'd guess National most likely to form government here as the alternative would require NZ First and Maori to side with Labour/Green.

Next most likely: a 9.8% chance that only UF makes it in. National and UF get 59 seats; Labour/Green 58. Maori Party chooses the government.

Next: a 9% chance that only NZ First makes it in - worst state of the world for National, who get 56 seats to the Labour/Green 55. NZ First could ensure a National government with 5 seats, but I'd expect them to be more likely to go with Labour; they couldn't push Labour over the line on their own. So Maori chooses the government.

Then, a 5% chance that all the minor parties survive. National's coalition gets 58 to 54 for Labour/Green. Either Maori or NZ First could make a National government, but it would take both together to make a Labour government.

Finally, a 4.2% chance that UF and NZ First survive while ACT doesn't. National and UF get 57 seats; Labour/Green 55. Maori or NZ First could give the government to National; both would be needed to make a Labour government.

The market says there's a 78% chance of a National Prime Minister following the next election; this requires that Maori stay onside with National if the vote share markets are to be trusted. I'm always a bit nervous about relying on results from the vote share markets - they have very flat payoff curves and don't pay out for about a year.

But the bottom line seems to be that the Maori Party will likely choose the next government; consequently, whichever party forms government, the Coastal Coalition folks will be upset.