The 2004 BC Citizens’ Assembly on Electoral Reform recommended using a multi-member approach in which adjacent ridings are merged and voters use a ranked ballot to express their preferences between individual candidates.

How Multi-Member Voting Works

The core idea behind multi-member voting systems is that, if neighbours with different political perspectives are to be represented in the Legislature by an MLA each of them supports, more than one MLA will have to be elected to represent each locale. The most straightforward way to do this would be to merge a handful of existing ridings (the BC Citizens’ Assembly recommended groups of between two and seven current ridings) and elect a set of MLAs who would represent the local diversity of political opinion in the locale.

This approach of merging adjacent ridings is used in two main families of voting systems:  the Single Transferable Vote recommended by the BC Citizens’ Assembly, and List systems commonly used in Scandinavia.  One critique of the idea of using this approach in BC is that we have large distances between population centres in parts of the province, so it would be good to guarantee that, if we use a multi-member approach, we have MLAs based in different places across the region.  Fortunately, it’s possible to do this using a variant called Local PR which can guarantee that we elect one MLA from each existing riding, so there would be no change in where MLAs come from.  For more details about this approach, check out the Local PR website developed by our friends in Guelph, Ontario.

Multi-Member Example: Burnaby/New Westminster: For example, the cities of Burnaby and New Westminster currently elect five MLAs. In the 2017 election, all five seats went to the NDP, even though NDP candidates won less than 50% of the vote. Although Liberal supporters accounted for 34% of the voters there, no Liberal candidate won a seat. A fairer outcome would have been for the Liberal voters to have been able to elect at least one of these seats and perhaps also the Green voters.

Burnaby MM
Example of combining five existing seats into a multi-member region (cities of Burnaby and New Westminster). The set of MLAs elected will more closely reflect how voters voted.

Multi-Member Example: Kelowna/Penticton: Similarly, there are four seats in the Kelowna-Penticton area. The Liberal candidates won 58% of the vote and all four seats. The 24% of voters who voted NDP and the 16% who voted Green won no representation at all. Again, a fairer outcome would be for at least one of these seats to go to an NDP candidate, and perhaps also one to a Green Party candidate.

Kelowna MM
Example of combining five existing seats into a multi-member region (cities of Kelowna and Penticton). The set of MLAs elected will more closely reflect how voters voted.

The Ballot

If more than one MLA is elected, how do we choose who wins? The most frequent recommendation is that we use a ballot which lists more than one candidate from each party. A typical ballot might look like this (for the example four-seat Kelowna-Penticton riding shown above):

Kelowna MM Ballot

Types of Ballots: Parties would be free to nominate as many candidates as they wished. Voters could be asked to do something as simple as choose their favourite candidate by marking an ‘X’, or, much the way most parties handle their own leadership races, voters could be invited to rank candidates both within and across parties.

Open List Counting: Counting the ballots could be done in a number of ways. The easiest would simply be to count up the total number of ballots cast for candidates of each party, figure out how many seats each party deserves based on their candidates’ share of the popular vote, and give the seats to those candidates with the most votes. This is known as the open-list method.

Ranked Ballot Counting: The BC Citizens’ Assembly felt that voters should have even more choice, so they recommended using a ranked ballot instead. With a ranked ballot, voters indicate their top choice with a ‘1’, their second choice with a ‘2’ and so on. To determine the winners, a simple way to count ranked ballots is to eliminate the candidate with the fewest votes and distribute those votes to the next-named candidate on each ballot, the same way most parties do with their own leadership contests. This is repeated until there are only as many candidates left as seats.  With the Local PR approach, we ‘protect’ each riding – a candidate can’t be eliminated if they’re the last candidate left from that riding; instead, they’re declared elected and the count proceeds.  This protection rule ensures that there is one candidate elected from each current riding.

With either the open list or ranked ballot approaches, the final group of candidates elected would broadly represent the diversity within each electoral district and would be the individual candidates most supported by the voters.