On Monday, the Canadian province of Ontario held its local elections. For the first time, municipalities were given the option of using a ranked ballot system, and London became the first city in Canada to hold elections for Mayor and Council in this way. Two other cities, Cambridge and Kingston, both had referenda on the same question, and both cities will be adopting a similar system in 2022.
How does it work?
When voters receive a ballot paper, they are instructed to rank the candidates in order of preference. In London, voters were able to rank up to three candidates as their first, second and third preferences.
On the first round of counting, only the first preferences are counted. If any candidate reaches the election threshold (defined as 50% of the valid votes plus one, rounded down if necessary), that candidate is elected. If not, the candidate in last place is eliminated, and their votes count in the second round for the voter’s second preference (or count as exhausted, if none of the voter’s preferences are still in the running). This is repeated until a candidate reaches the threshold, or until a single candidate remains, whichever comes first.
A total of 14 candidates ran for Mayor of London, and 14 rounds of counting were required to determine a winner. In the end, Eric Holder won in the final round, taking 34.2% of the first-preference vote, and keeping a relatively consistent majority (between 11.5k and 13.3k votes) over his nearest rival.
However, there were a significant number of exhausted ballots, resulting in only 60% of voters ranking the winning candidate at all. Right from the first distribution of preferences, more than a fifth of votes for the eliminated candidate were exhausted each round. Given that this started right from the beginning, it’s highly likely that most of these ballots (particularly in the early stages) did not mark all three preferences, though we don’t know the exact figures. To put this rate into context, the 2011 election for President of Ireland, which uses a similar system, saw only 7.7% of ballots exhausted by the final round (London’s rate at the same stage was nearly three times as high).
London is divided into 14 wards, each of which elects one City Councillor. Of those 14 wards, three wards (3, 6, 7) only had two candidates (meaning a winner on count one is assured), and another five wards (1, 2, 4, 10, 11) were also won in the first round. Six wards (5, 8, 9, 12, 13, 14) required multiple rounds of counting to determine a winner.
In all six of these wards, the candidate who got the most first-preference votes ended up winning the seat. In most wards, the leader had a relatively consistent majority in each round. However, in Ward 5 (north-centre London), we saw preferences make a significant shift when the third-place candidate was eliminated.
As with the mayoral election, there was a high rate of ballot exhaustion in the council wards as well. In two wards (8, 13), there were so many exhausted ballots that the winning candidate didn’t even reach the threshold when they were the only ones remaining!
What about next time?
The high rates of ballot exhaustion suggest that there is plenty of education that needs to be done with voters on this voting system. A common misconception of this system is that indicating a second or third preference harms your first preference candidate (it doesn’t). There may also have been additional confusion due to the fact that School Board Trustees were still using first past the post. There might be other misunderstandings about the system, and having completed the process for the first time, the City would benefit from conducting some focus groups to understand why ballots are being exhausted at such a high rate.
The candidates in this election treated it like a traditional first-past-the-post election. Candidates who have a negligible chance of winning themselves may be able to influence the later rounds by encouraging their supporters to vote a certain way for their second and third preferences.
With Cambridge and Kingston joining London in the ranked ballot system in 2022, it’s clear that this is the general direction elections in Ontario is going. While we can’t call this experiment a failure, it’s clear that there are improvements that will be needed, for both the system and the candidates who participate in it, in order for it to become second nature.
The full results of the London municipal elections can be viewed here.