In recent Fair Vote Canada panel discussion*, a speaker mentioned that in London's election, a Black woman was elected in the London Ontario city election under ranked voting. This was a reference to Arielle Kayabaga, who according to the speaker would not have been elected under First Past The Post. This was wrong as explained below. But is just one example of the unknowns of this election, the first election held not under FPTP (or Block Voting) since 1971.
The speaker stated that Dave Meslin had done great work with ranked ballots and published "London Leads" (available online and published in part below).
The speaker stated that ranked ballots or party-list or any kind of fair vote should be encouraged, as we all agree that any kind of fairer vote is better than what we have now.
Anita Nickerson (the moderator of the panel discussion) agreed that uniting behind any move toward PR was the strategy of the Fair Vote Canada group. Any proportional improvement is a win for voters. She did say though that Fair Vote Canada was not supportive of majoritarian systems (exactly the kind of system used in London).
What was wrong with the Alternative Voting used in London's 2018 election?
Here's my thoughts on this.
===================================
London Ontario AV election in 2018
In every ward and in the mayor's contest, the candidate who lead in the first count was elected every time. So the winner would have been the same if the election had been conducted using FPTP (assuming that voters would have voted the same).
Perhaps the speaker was assuming that many voters might have assumed that Kayabaga would had little chance and she would not have received the votes that she in fact did under Alternative Voting, where voters were prepared to take a chance knowing that their votes might be transferred to another candidate later instead of being wasted, as is always the case for votes cast for less-popular candidates under FPTP. (Later in the panel discussion the speaker does say this same thing herself.]
================
From "London Leads":
"London City Council is the first government in Canada to reject the discredited First-Past-the-Post model and adopt a fair electoral system that respects voters
On October 22 2018, London used ranked ballots in their local elections - a first for Canada! This decision places London as the #1 innovator and pioneer of democratic reform in the country.
Ranked Ballots are a small and simple change to make local elections more fair, inclusive and friendly. Explore our website to learn more ..." =============
It should be noted that London is NOT the "first government in Canada to reject the discredited First-Past-the-Post model and adopt a fair electoral system that respects voters"
Two provincial government used ranked ballots in their elections - Alberta in eight provincial elections and Manitoba for nine elections. That happened in the 1920-1955 period.
Previous to 1930, 20 Canadian cities and other places used ranked ballots, the voting system that London adopted in 2018. They used Alternative Voting for election of mayors and other single-office elections and used STV (a proportional representation system) elsewhere.
London by the way was not one of these PR cities. They were all in western Canada.
=============
Montopedia blog on London's election:
The 2018 London, Ontario city election was the most recent use of AV in Canada.
This was the first government election in Canada to use preferential balloting since Calgary's last STV election in 1971. In that election Calgary's mayor was elected using Alternative Voting.
Calgary had also used Alternative Voting for election of its mayor and city councillors in the 1960s. From 1917 to 1961 Calgary had used STV to elect its city councillors, and AV to elect its mayor.
AV had been used in various elections in Alberta and Manitoba, from 1924 to 1955, (these included all provincial elections held outside the major cities and all provincial by-elections) and in BC elections in 1952 and 1953, as described in my blog "Variety of electoral systems used in Canada".
Alternative Voting has been used for election of the leaders of our major political parties for many years. Long used for fairness at this level, it has not been adopted for elections of representatives of the general public until now - at least not since 1971.
Only up to three preferences allowed on London ballots
The London election system allowed each voter to mark no more than three preferences.
In London's 2018 election, with 15 candidates running for mayor, many votes - more than 20 percent of the total - were exhausted by the time one candidate accumulated a majority of the votes still in play to secure the mayor's seat. This was likely caused by the limit of three choices.
Thirteen eliminations occurred, to thin the field of candidates to only two. In each one, about a third or quarter of the votes were exhausted, rising to the one-half level in the 13th round after many votes had used up their three preferences.
To make things worse, when the field thinned down to two, 21,000 votes were exhausted and only 13,000 separated the two remaining candidates. (from "City of London, Ontario, Municipal Election - certified results" website)
It must be difficult to know, with that many exhausted votes, whether the correct person won the election.
==============================
Mysterious post-election vote transfers
Unusually, in the vote count for the 2018 London election, transfers of votes between candidates continued even after a candidate had accumulated a majority of votes still in play and had been declared the winner:
The calculation of the election of mayor should have ended on the 13th count, when one candidate (Holder) got a majority of the votes still in use.
The calculation of the Ward 5 election should have ended on the 5th count, when one candidate got a majority of the votes still in use.
The calculation of the Ward 8 election should have ended on the 8th count, when one candidate got a majority of the votes still in use.
The calculation of the Ward 9 election should have ended on the 2nd count, when one candidate got a majority of the votes still in use.
The calculation of the Ward 12 election should have ended on the 4th count, when one candidate got a majority of the votes still in use.
The calculation of the Ward 13 election should have ended on the 7th count, when one candidate got a majority of the votes still in use.
In part due to the high number of exhausted votes, which in turn was in part due to the ceiling put on the maximum number of back-up preferences that voters could mark on ballots -- only two-- some of the elected reps did not receive a majority of the votes cast.
Alternative Voting is majoritarian but it is possible to be elected without majority of votes cast due to the production of exhausted ballots. Successful candidate are required to have majority of votes still in play but not necessarily majority of votes cast.
And in the London election such was the case in six of the fifteen contests:
Holder, the winning mayoral candidate, was elected in the end with 44,373 votes when 48,320 was a majority of valid votes that were cast.
In Ward 5, Cassidy won with 3922 votes, 44 percent of votes cast.
In Ward 8,Lehman won with 3058 votes, 39 percent of votes cast.
In Ward 12, Peloza won with 3139 votes, 48 percent of votes cast.
In Ward 13, Kayabaga won with 2325 votes, 41 percent of votes cast.
In Ward 14, Hillier won with 2522 votes, 48 percent of votes cast.
District by district results
At-large mayoral contest
Holder, the winning mayoral candidate, was elected in the end with 44,373 when 48,320 was a majority of valid votes that were cast. Holder had led in every round but whether he had a majority of of the votes behind him was not determined in this election, due to the large number of exhausted votes and the narrowness of the win in the end.
(He is accorded 57,609 votes but that is only after his closest competitor (Paolatto) is eliminated and the votes marked for Holder as back-up preference were transferred. But that is faulty result - it says that where all other candidates are eliminated, Holder was marked as back-up preference, but not that if Holder and Paolatto are the last two competitors, which of the two would have the most support - if all the voters had input. By the point in the vote count as held in 2018 where the field of candidates had been reduced to just those two, 21,000 votes cast had been ruled exhausted and the two contenders were only 13,000 votes apart. (see my July 2021 blog on the mayoral election for more info on this.)
Ward elections
Ward 1 First Count victory - no transfers
Van Holst won with 2581 51 percent of votes cast (First Count victory - no transfers)
Ward 2 First Count victory - no transfers
Lewis won with 3481 64 percent of votes cast
Ward 3 First Count victory - no transfers --- two-candidate contest
Salih won with 3421 73 percent of votes cast
Ward 4 First Count victory - no transfers
Helmer won with 2559 51 percent of votes cast
Ward 5 Cassidy won with 3922 44 percent of votes cast
(official results have this result --
Cassidy winning with 4741 votes 53 percent of votes cast)
Ward 6 First Count victory - no transfers --- two-candidate contest
Squire won with 3959 70 percent of votes cast (two-candidate contest)
Ward 7 First Count victory - no transfers --- two-candidate contest
Morgan won with 6117 75 percent of votes cast (two-candidate contest)
Ward 8 Lehman won with 3058 39 percent of votes cast
(official results have this result --
Lehman won with 3823 49 percent of votes cast)
Ward 9 Hopkins won with 4948 53 percent of votes cast
Ward 10 First Count victory - no transfers
Meerbergen won with 4402 53 percent of votes cast
Ward 11 First Count victory - no transfers
Turner won with 4255 54 percent of votes cast
Ward 12 Peloza won with 3139 48 percent of votes cast
(official results have this result --
Peloza won with 3403 52 percent of votes cast
Ward 13 Kayabaga won with 2325 41 percent of votes cast
(official results have this result --
Kayabaga won with 2894 49 percent of votes cast)
Arielle Kayabaga is the first black woman on London council.
She was leading in the first count and every count thereafter until finally on the 7th Count when the field of candidates thinned to just two remaining candidates she took a majority of votes still in play to win.
Ward 14 Hillier won with 2522 48 percent of votes cast
(official results have this result --
Hillier won with 3370 64 percent of votes cast
The six marked in bold won the seat with less than the majority of vote cast. (However, that 9 out of 15 were elected with a majority of votes cast in their wards is legendary. In Edmonton's last election conducted through FPTP Single-Member Plurality, five of 12 councillors - almost half of the council - were elected without majority of votes cast in their districts.)
The difference between my winning totals and the official results, where indicated above, is that according to the results (as published online in https://london.ca/sites/default/files/2020-09/2018%20Election%20Results.pdf )
vote transfers continued in those wards even after a candidate had received a majority of votes still in play.
I have not seen this phenomena of after-victory transfers elsewhere. In the London vote, the votes of the last remaining defeated candidate (or one of the last remaining defeated candidates, if there was more than one at the end) were transferred to the winner if the ballot was marked that way.
This is correct as it shows that the winning candidate was the marked choice of the voter. But it is not normal.
It appears to be done for propaganda purposes to present the winning candidate as the choice of as many voters as possible but I have not seen these "after-victory transfers" take place in any other ranked ballot election. Generally vote transfers end when someone has been elected.
And in historic STV elections some votes were declared exhausted (and considered wasted) even though marked for candidates who had already been elected. This made the waste appear worse than in fact it was, while the 2018 London election conducted an after-victory transfer in many wards to present the winning candidate with the most support possible - or at least as much as could be legitimately added in one after-victory transfer.
(In Ward 12, even with one after-victory transfer (the transfer of Evans' votes) , the back-up preferences on Weniger's original 1800 votes were never considered. Some of them might have been marked for Peloza as well!)
London's 2018 city election was fair
The result of London's 2018 city election is fair as compared to previous London elections. In 2018, 9 out of 15 were elected with a majority of votes cast in their wards, a much larger number than in previous London elections that used FPTP.
2006 London election
In the 2006 London election, Josh Morgan, a former aide to Liberal MP Sue Barnes, had sought a council seat, and had lost by just 23 votes to Paul Hubert, who had support from barely more than 20 percent of the voters. Morgan, Hubert and two others all received about 20 percent of the vote and Morgan noted that Hubert then had the curse of knowing that three-quarters of the voters had voted against him. (info from interview online: https://www.unlockdemocracy.ca/londonleads "Why London changed to ranked ballots")
2010 London election
In the 2010 election Fontana was elected mayor with only 47 percent of the vote. The incumbent mayor DeCicco-Best received only three percent fewer votes than Fontana while another woman came in third with enough votes to give DeCicco-Best the lead. Perhaps vote-splitting among women voters (and among women-sympathetic male voters) caused Fontana's victory.
That sort of thing could not happen under Alternative Voting.
Or I should say should not happen under Alternative Voting. Unfortunately the large number of mayoral candidates that ran in 2018 and the arbitrary three-preference ceiling meant that the successful candidate won with only 46 percent of the votes cast, there were 21,000 votes exhausted by the time one candidate received a majority of votes still in play (the 13th Count), and the difference between the last two surviving candidates was much less than the number of exhausted votes. This casts some doubt that the successful candidate Ed Holder was actually the choice of more voters than the second-ranking candidate Paolatto.
These doubts were supposedly laid to rest when an after-victory examination of the back-up preferences on Paollato's votes found that Holder had assembled a majority of votes. This proved that a majority of voters had given Holder a place among their top three choices for mayor. (Although theoretically three could hold this honour.)
Other grounds for thinking the result fair is that Holder had the most votes in every count; that at no time was the number of exhausted votes sufficient enough to allow a third or lower candidate a lead in votes over Holder; at no time did any two candidates have enough together to surpass the quota. (Under normal AV you could rely on vote transfers to bring this out, but with London's arbitrary three-preference ceiling combined with a vote count process that included 13 Counts, the ability of two camps to join together to achieve election of one of their candidates was not a given.)
Paollato's ballots showed that his voters favoured Holder as well - enough to give him a post-victory majority anyway.
Wards cause unfairness
London having small one-seat districts means there can be variance that way.
Vote totals district to district varied from high of 9418 (Ward 9) to low of 4702 (Ward 3).
majorities "required" thus varied from 4710 to 2351.
with total vote city wide being 100,000 (the total for the mayor's vote), and 14 seats to be filled,
Quota (Droop) would have been about 6700 votes, a total that none of the successful candidates received under AV in 2018. (But STV vote transfers would have ensured that several would have gotten that many - as long as voters were allowed to rank all the candidates or at least many more than three.
In the end, perhaps some of the candidates would have been elected without quota by merely surviving until the field of candidates was thinned down to the number of remaining open seats.)
===================================
Thanks for reading.
====================================
* Fair Vote Canada 20th Anniversary Panel, June 4, 2021
https://www.youtube.com/watch?v=G7qh5JqQtGw
partial transcription in Montopedia blog:
https://montopedia.wixsite.com/montopedia/post/2021-fair-vote-canada-discussion-to-55-40
Comments