British voters are heading to the polls Thursday, and it looks as if it’s going to be a nail-biter. Political prediction markets suggest that it’s basically a 50-50 bet as to whether the Conservative prime minister, David Cameron, will serve another term or be ousted by Labour’s leader, Ed Miliband. I can recall very few elections where the bookies couldn’t pick a favorite by election eve.
The traditional two-party system is breaking down in Britain, which makes the outcome more uncertain because it renders traditional approaches to predicting the outcome less useful. Multiparty elections bring a type of complexity that traditional polling — and computer models based on polling — are ill-equipped to deal with.
Voters become less predictable when they face more than two plausible candidates, because they might vote strategically in an effort not to waste their vote. The logic is that if you expect your candidate to come in third, you can make a difference by opting instead to cast your ballot for the least bad choice among the two more likely candidates. Consequently, you may prefer one candidate, but end up voting for another. As much as polls try to pick up whom voters intend to vote for, people are still likely to tell pollsters whom they prefer.
And the dynamics can get complicated, as strategic voters will choose based not just on whom they want to see elected, but also on whether they think that candidate has a chance. In turn, their assessment of whether a candidate has a chance depends on whom they believe other voters will vote for, which depends on whom those voters believe other voters believe will win. It’s not just beliefs about beliefs that matter, but beliefs about beliefs about beliefs, and so on. It’s turtles all the way down. Abrupt swings in voter sentiment can occur when any hint that a candidate’s support is fading leads others to jump ship, too.
Then there’s the problem of figuring out who will be the next prime minister, which is not decided directly by the voters, but rather by the members of Parliament they elect. The problem is that there’s a less than one-in-ten chance that any party will win a clear majority. And so the next prime minister won’t be decided on Election Day, but rather in the subsequent rounds of horse-trading, as the major parties seek to find partners willing to form a governing coalition. Pollsters are obviously silent on how this will play out, even as traders in political prediction markets are gaming different possibilities.
Despite all this complexity, it is likely that more than $150 million has been wagered on the election.
So here’s what the prediction markets are telling us. The Conservatives have more than a four-in-five chance to win the most votes and the most seats. Even so, they are seen as having less than a one-in-ten chance of being able to form a majority without coalition partners. This is partly because of the Scottish National Party, which is now likely to become the third-largest party, after picking up nearly all of the traditionally Labour-voting seats in Scotland. It is expected to throw its support behind the Labour Party.
Nick Clegg’s Liberal Democrats — the Conservatives’ current coalition partners — are expected to lose a lot of seats to both the Labour Party and the Conservatives, as well as in Scotland. The Liberal Democrats could potentially make a coalition partner with either major party.
Add in the minor players — the Welsh independence party Plaid Cymru; Northern Ireland’s Democratic Unionist Party and Social Democratic and Labour Party; the right-wing populist UK Independence Party; and possibly the Greens or other independents — and a number of possible coalitions emerge. Many of these potential coalitions go Labour’s way, meaning that it’s as likely as the Conservatives to form the next government.
These negotiations are expected to take time, and there’s roughly a 50-50 chance that we still won’t know the outcome a week from now. NYtimes
No comments:
Post a Comment