There’s a theory out there that if you want the most accurate possible forecast of election results, ask the people who have money riding on it.
The University of British Columbia has been running an election prediction market since 1993, and today investors can place between $25 and $1,000 on the outcome of the Oct. 19 federal election.
Sauder School of Business associate professor of business Werner Antweiler, who runs the market, says the results are “often more accurate than public opinion polls.”
That’s because “efficient markets are very good at reflecting all available information and can often reflect information faster than opinion polls that take several days to complete and process,” Antweiler writes.
So who does the market see as the winner of Canada’s federal election on Oct. 19? Well, it sees a controversial outcome — a Conservative minority government, despite the Liberals winning the largest share of the popular vote.
Based on trades made as of Monday, the market sees a roughly 51.5 per cent chance of the Conservatives winning the most seats in Parliament, with a 28 per cent chance of the Liberals winning the most seats, the second-likeliest outcome. The Tories took the lead from the NDP about two weeks ago.
This chart shows bets placed on which party will win a parliamentary plurality. The Conservatives lead with 51.5 per cent. Source: Sauder School of Business
But in terms of the popular vote, traders are betting that Justin Trudeau’s Liberals will beat Stephen Harper’s Conservatives, giving the Liberals 35.5 per cent, to the Conservatives’ 31.5 per cent. The NDP trail around 27 per cent.
That contradicton may be intentional — investors may see Canada's first-part-the-post system electoral system skewing results. But more likely, the contradiction is a reflection of just how tight this race has been.
The popular vote market.
Whoever wins, the market is largely convinced it will be a minority government — traders see a 75.59 per cent chance of that happening. The Tories are most likely to form a majority, but there’s only a 17 per cent chance of that happening, according to the market.
Odds of a majority government.
But markets, like voters, are fickle. The prediction market had forecast an NDP victory as recently as mid-August, but as of this week the NDP had fallen to third place, with a 19 per cent chance of winning the most seats.
For UBC’s Antweiler, this is more than just gambling on an election. He says part of the point of the market is to help investors predict election outcomes.
“To this extent the markets have a public service function,” he writes.
And he argues the market can forecast better than polls because “things are a bit different when people have money at stake.”
Investors “tend to put their money where their mouth is, so they might be more inclined to vote for who they think will win rather than who they want to win, and in a way that can be a lot more truthful in some respects,” Antweiler told Business in Vancouver earlier this year.
But in the end the prediction market is composed of people, who, as we know, make flawed investments.
In the 2013 British Columbia provincial election — which the Liberals won in a stunning upset the pollsters hadn’t predicted — the market got it wrong too, calling for an NDP majority.
“I think the traders got duped by a lot of the polls, much like everyone did,” Antweiler said.