Wednesday, November 20, 2019

What lessons can be learned from the results of the 2018 BC referendum on proportional representation?

As I expected, a post-referendum poll shows that BC residents still like PR.

“A simple pro-rep question could have won” said the headline:


From this poll, it's clear that some potential supporters voted against PR because they did not feel informed enough, or perceived the whole process unacceptable, or found the options unclear or unconvincing, or found the questions confusing, or feared MLAs being appointed from party lists, or they were just afraid to vote for "the mystery box." (See below).

A poor process

One can argue that this referendum should never have been held, but it was in the NDP Platform. However, after Attorney-General David Eby was appointed to be the independent Minister in Charge, he conducted consultations and made recommendations concerning how the referendum process should be conducted. It ended up with a two-part question with three systems on the ballot.

The referendum timing was intended to be fast, to leave time for Elections BC to hold boundaries hearings for new ridings for the 2021 election, with a simple referendum. It did not allow enough time for public education on three systems. The consequences of this are discussed below.

Thousands of electoral reformers spent six months defending the process and how it was arrived at. It’s hard to admit that the process was the problem. But the polling evidence is clear. With a referendum on first-past-the-post versus one fully fleshed-out alternative, designed and explained via deep public consultation, PR would have won. Reformers should not be afraid to say so.

Research Co. poll:

An online study conducted from December 18 to December 20, 2018, among 803 adults in British Columbia.

(There was also an Angus Reid poll, but it asked less useful questions, and gave no results from non-voters. And its sample did not match the referendum outcome.)

The Research Co. poll mirrored the results of the referendum, with 49% of respondents reporting they voted for the first-past-the-post system, 31% voting to move to a proportional representation system, and 20% not casting a ballot. Once the non-voters are removed, the result of the “exit poll” is 61% choosing to retain the existing system and 39% opting to change it.

The survey shows that large majorities of British Columbians support basic PR concepts such as attempting to eliminate “strategic voting” (75%), that a party should only win majority power if its candidates won a majority of the votes (70%), a party not governing with a majority of seats but with fewer than 40% of all votes cast (63%), and MLAs being elected from different parties in close proportion to how voters voted in each region voted (57%). (Also, the notion of smaller parties holding the balance of power influenced the way 55% of voters voted, but the poll does not say what percent found that good and what percent bad.)

Majorities of British Columbians also endorse other concepts, such as that the voting system should not disadvantage independent candidates (70%), almost all votes helping elect an MLA (64%), voters being able to choose among different candidates from their preferred party (58%), and voters being able to make their vote count for a more popular candidate rather than having it ignored (52%).

Why did PR lose?

The No. 1 issue for those who chose not to vote is “not feeling informed enough” (48%). While some non-voters also claimed they never received a ballot (18%) or simply forgot to mail it (17%), nothing came close to the uneasy feeling of being in the dark.

When asked about issues that influenced the way they voted, majorities of British Columbians mention

1. the details of the three options on the second question not being fully fleshed out (59%),
2. the three options listed on the second question being confusing and not clearly explained (55%), and
3. MLAs being appointed from party lists (53%).

Many refused to vote yes on the first question when the result of the second question was a "mystery box," which might be an unknown system.

The no side was highly effective at convincing people that many details would be decided later. In fact, other than closed lists, not many details that made much of a practical difference were left to be decided later, but since voters were confused and ill-informed, explaining this was too hard. The conservative news media were no help.

This is no surprise. The No campaign’s best messages were bound to be “we don’t understand this enough” and “we don’t know enough.” Research has shown that low info voters will take the status quo when they don't understand the alternative. The opposition preyed on those fears.

Unfortunately, the Yes campaign played into its weakness, merely playing defence, reassuring voters of the benefits. This didn’t work. Voters couldn’t focus on what’s wrong with winner-take-all voting, when they were confused about PR.

Confusion on PR caused other problems: 50% of respondents were concerned over coalition deals being worked out “in the back room,” 49% over fringe or extremist parties winning seats, urban centres having disproportionate influence over future governments, and the details of the chosen proportional representation option being left to an all-party committee. Slightly smaller proportions of voters were influenced by the notion that voters from rural areas might lose local representation (45%) or the government possibly rigging the process for partisan gain (41%).

The PEI plebiscite had shown that a multi-option referendum could work well. Some reformers told Attorney-General Eby that a multi-option referendum would be better in BC. However, when New Zealand held their first multi-option referendum in 1992, every household in the country received a booklet in the mail, with detailed explanations of all four models, much more detail than Elections BC could give. New Zealand’s public education program ran for many months. It included TV debates with proponents of each of the four models. Of the 55% of registered electors who took part, an overwhelming 85% voted to change their electoral system. In the second part of the same vote, 70% favoured MMP.

So a good multi-option referendum is indeed possible. There was no time for that in BC, unless the referendum would have been the next spring, two years into the government’s mandate, when it might well have been suffering from a mid-term slump that would have led to PR’s defeat. Besides, Elections BC needed more time for new boundaries. No one wanted to wait that long. But a fast process was incompatible, as we have just seen, with a multi-option referendum.

Reformers expected voters to accept that the first question was the most important (although Elections BC materials did not say that). But many felt overwhelmed deciding on systems when they were not familiar with the systems on the ballot.

Can reformers learn how to scale up their greatest strength – relationships -- to withstand the forces of No? Not in a fast process.

As a Globe and Mail editorial said “Even the details of the proposed mixed-member proportional system, the system that was supposed to win the referendum, were to be left to the discretion of the government. It was like asking someone if they wanted the usual for lunch, or a sandwich. What kind of sandwich? Can’t say. Vote “sandwich” and leave it to the politicians to figure what goes between the bread."

Political leadership

Many have complained of the lack of political leadership from the NDP. However, in March 2017 when reformers interviewed Horgan, he said “Mixed Member Proportional (MMP) is the preferred electoral system if we’re going to change from what we have now. It’s party policy and if people don’t like it, there’s always the party convention." (Meaning: don’t try to change party policy undemocratically, it was set by Convention.)  "I’m confident that the group I’m leading now is behind me on this.” Regardless whether the platform mentioned MMP, every MLA knew it was party policy.

Right after the 2017 election, the BC NDP and Greens signed a Confidence and Supply Agreement committing both the New Democrat Government and the Green Caucus to proportional representation. "Legislation will be introduced in the 1st session that: (1) A referendum on proportional representation will take place in the fall of 2018, concurrent with the next municipal election; and (2) The form of proportional representation approved in the referendum will be enacted for the next provincial election. The parties agree to both campaign actively in support of the agreed-upon form of proportional representation." A fast
 referendum on a single form of PR. Not enough time for public education on three systems. 

Eby’s non-partisan experts took FVC’s input and came up with a new referendum model. However, no NDP MLA was elected on a platform of holding a referendum on a "mystery box" that might have chosen an unknown system. 

Although Fair Vote Canada has not supported any closed-list model since 2013, FVC somehow failed to convince Eby to keep it off the table. About two months before the campaign started, the BC NDP's own pro-PR website said the MMP model would be open-list, with a sample open-list ballot; but this contradicted the Elections BC materials. So by the time Horgan belatedly took closed-list MMP off the table, at least 53% of referendum voters either didn’t hear him or didn’t believe him.

Reformers could have had strong NDP leadership. Instead, the process reduced Horgan to asking voters, feebly, to “take a leap of faith.” Many NDP voters liked the result of the 2017 election, and had lost some appetite for PR.

By contrast, the MDN in Quebec worked to build up five-party support for a specific model of regional MMP which MDN had endorsed first.

Back to the Research Co. poll: “While BC Liberal voters from 2017 were decidedly more likely to support first past the post (82%), only 53% (really?) of those who say they voted for the NDP and the Greens favoured proportional representation.” (The Angus Reid poll found the NDP voters who voted in the referendum split 70/30, but 70% of 42% of 716,421 is 210,628, or 39% of the yes votes, so it’s not that simple.)

“Almost four-in-five British Columbians (78%) agree that politicians are in a conflict of interest when it comes to making decisions about how we vote, and would like any future proposals to involve an independent, non-partisan citizens’ body. This includes 77% of those who voted for Proportional Representation and 82% of those who voted to keep the First Past the Post system.” A logical conclusion when BC voters look at how the politicians screwed this process up.

The Liberals made it a partisan question

The party opposed campaigned for NO like their life depended on it, the most Trumpian campaign Canada has ever seen. Many say that partisan interests lie at the core of the problem, including the partisan interests of political parties elected to form government thanks to FPTP, the partisan interests of incumbents who fear losing their seats were the electoral system to be changed or the partisan preferences of voters themselves. One of their messages was that the referendum was RIGGED (“stacked deck” “rigged game”) in favour of the NDP and Greens, and the idea was that the NDP/Greens would be in power forever with PR - they made sure their Liberal voters heard this, despite there being no factual basis. When the debate was held between Horgan and Wilkinson, that made the referendum even more of a partisan fight than it already was.

Referendums rarely achieve the social license for changes. As the recent Brexit and BC referendums have shown, referendums are divisive and not a way to develop social consensus, especially with “opponent” and “proponent” groups expected to duke it out as a dystopian form of “public education.”


Does low turnout correlate with low support for PR? Yes, and the Yes campaign failed on ethnic/minority out-reach. But that’s only part of the story. In last year's provincial election in BC, 1,973,914 voters cast ballots. In this referendum, 1,403,358 completed voting packages were returned to Elections BC. That's 570,556 previous voters who did not vote. Still, if they had all voted, it would have taken 441,137 of them (78% of those non-voters) to vote for PR to change the result. Plus everyone who turned 18 in the past 16 months, which should have given the PR side a boost. But the mail-in ballot made the turnout worse because young people don't use mail. Being more likely not to have up to date voter registration, many didn't receive their ballot.

The silver living: MMP won

The referendum was not a complete loss. Some PR supporters in BC were still trying to fight the 2009 referendum over again, hoping to resurrect BC-STV. MMP won the first round on Question 2, and then won 63.05% on the final count. Since it also won the 2016 PEI plebiscite, and is the only model being considered in Quebec, this simplifies the debate on both sides of the Rockies.

As previously noted, the BC-STV model designed by the Electoral Boundaries Commission was not very proportional. With each district having, on average, only 4.25 MLAs, Green Party voters, who deserved to elect seven MLAs on the votes cast in 2009, would have been lucky to elect even three MLAs. Yet many BC voters complained that the proposed 20 districts, with an average population of about 222,000, were already unmanageably-large. This is not a trivial point. In a country with as much geography as Canada, fitting our geography into the voting system is the major design issue.

Anecdotal evidence

Every one has their favourite anecdotal comments. Here are mine, which match the polling results. Two of them are from friends I trust, retired Ontario teachers who moved to BC.

“I voted to keep FPTP not because I reject PR, but I was against all the devilish details of implementation being left to be decided later – no doubt in a way that favours the ruling party. We had no option except FPTP. The PR choices were vague and scant with details and explanations. It appeared that if FPTP was defeated, politicians would decide how to implement PR.  Unacceptable to 61 percent of us. The yes PR side campaign was a cluster f of “the right thing and the high road” but no details. That’s why 61 percent of us said NO!!”

“This campaign DID NOT properly explain what PR meant and until people get an understanding of what it is, they will be afraid to vote for it. There must be a way to make it simple and understandable. So disappointed. Let's keep advocating for it in the sense of educating people about it.”

People don't trust politicians to decide anything about electoral reform for totally understandable reasons. Unfortunately, politicians don't want to give up control to processes like citizens assemblies that people do trust more to decide details. “If I’d lived in BC, I wouldn’t have been happy with a blank cheque checkbox.”

“The politically informed and engaged typically understand the big picture (and often the finer details) of PR -- not so for the average voter or citizen. Anyone who doesn't understand (or take the time to understand) an alternative will favour the status quo or not vote.”

Last word to former BC Citizens Assembly member Shoni Field: "Did women give up fighting to get the vote after their first few failures? No, this is a long haul fight. We'd hoped that BC would be the first in Canada, and maybe now they'll be the last. But change is inevitable and one day we'll be using a proportional system and we'll keep working to make that happen.”

(Updated August 9, 2020.)