Notes on the Media’s Coverage of the 2018 BC Referendum

From BC Referendum Made Simple (Bee Current): A 3D model of the BC legislature

In 2018, British Columbians were asked whether to continue using the First Past the Post (FPTP) voting system for provincial elections, or switch to a form of proportional representation (PR). During the voting period and preceding months, a handful of journalists published neutral, well-researched articles to help British Columbians vote Yes or No according to their true preferences.¹

Unfortunately, several of the major newspapers appeared to forgo serious investigation, and instead helped the No side sustain a narrative that the NDP government had “stacked the deck” in favour of change. The papers published numerous opinion pieces that undermined trust in the referendum and would have convinced reasonable people to vote No even if they preferred PR.

To illustrate the media’s coverage of the 2018 BC Referendum, I offer notes on five markedly negative and poorly researched newspaper columns and editorials. The first article is well intentioned, but the pieces become increasingly biased and misleading. The final editorial contains factual errors that even the No side advertisers were unwilling to make explicit.

Andrew Coyne: Why not ask a simple question, B.C.?

Andrew Coyne is a National Post columnist who strongly supports reforming Canada’s FPTP voting system.
Here he is in a 20-second clip listing the faults of FPTP. (Click replay to watch the full 5-minute video.)
Coyne’s June 4 (2018) column criticizes the two-question ballot format proposed four days earlier by David Eby, the Attorney General of BC.
This is what the referendum ballot looked like. (source: Elections BC):
Coyne claims the ballot design is biased in favour of the current voting system, though no explanation is given.
Here’s a quote from the article.
Not only does the double-barreled ballot give an unwarranted advantage to the status quo, but it puts voters in the strange position of choosing between a specific system, first past the post, and a broad principle, proportional representation.
Again, no reason is given why he thinks the two-part ballot favours FPTP.² He does say the ballot “puts voters in a strange position” because the options on Question 1 are not both concrete voting systems. However, there’s nothing objectively wrong with asking voters to choose between FPTP and proportional representation. It’s actually just a more neutral way of asking, Do you want change? Yes or No.

Coyne’s alternative referendum ballot – the “simple question” mentioned in the title of the column – would actually have made it more challenging for voters to express a preference for or against change.

Why not simply ask one question: which of the following electoral systems do you prefer? Then list some alternatives, first past the post included — as just one option among several — and ask voters to rank them.

Here’s the format Coyne is talking about. Voters would rank all four options, including FPTP.

	1st	2nd	3rd	4th
Dual Member Proportional (DMP)	( )	( )	( )	( )
First Past the Post (FPTP)	( )	( )	( )	( )
Mixed Member Proportional (MMP)	( )	( )	( )	( )
Rural-Urban Proportional (RUP)	( )	( )	( )	( )

It’s true that Coyne’s format would have allowed voters to rank one or two proportional systems higher than First Past the Post, and any others below. But every format has flaws. The reality is that with Coyne’s ballot, voters who preferred FPTP would have had no influence over which PR system would get adopted if the majority supported change.³ By contrast, David Eby’s two-question ballot ensured that in the event of change, both Yes and No voters would have had an equal ability to determine the PR system. Also, Eby’s ballot is decidedly simpler if all a voter wants to do is support or oppose change.

Coyne also complains about the selection of voting systems on the ballot, though the only specific criticism is a prejudiced remark about one of the system’s inventors.
And there’s something called “dual-member proportional,” which Wikipedia tells me was “invented in 2013 by a University of Alberta mathematics student.”
It’s a common misperception that all significant discoveries and inventions are made by senior professors. When Albert Einstein published his famous four papers, the ones that introduced the theory of relativity and revolutionized physics, he was 26. When Canadian researcher Donna Strickland recently won the Nobel Prize in Physics, it was for work she did while she was a student. The fact that Sean Graham was a student when he invented DMP was not in itself a reason to dismiss the system.⁴ Coyne later wrote that “any of the three would be miles better than the status quo”, so one must wonder if he rushed to judgement in his June 4 piece.
Although the piece was written in good faith with no intent of helping either side, it clearly became a source of ideas for writers and advertisers seeking to attack the referendum process.
For example, the column was cited by Fair Referendum, a disingenuous campaign pretending to advocate for Andrew Coyne’s referendum ballot. The actual intent of the campaign – as well as most other efforts to discredit the referendum – was to convince people to vote No even if they might prefer a proportional system.⁵

Globe Editorial: B.C. stacks the deck in its electoral-reform referendum

Though the editors claim to be “torn”, this June 26 (2018) piece was published months before the start of voting and weeks before the official Yes and No sides were chosen.
The official Yes and No groups were chosen during the second week of July, and the voting period began on October 22. Unless their intent was to help the No side’s narrative take root, the Globe and Mail could have waited until the voting period to take a position.
While Andrew Coyne’s column (above) claims the referendum ballot gives an advantage to the status quo, the Globe and Mail asserts without explanation that the ballot is biased in favour of change.
Here’s the closest they come to supporting their claim that the two-question ballot was biased.
It’s not hard to see what’s wrong with this, if you imagine running a federal or provincial election on this template. How fair would that be? In the first stage, voters would be asked if they preferred an incumbent politician to some hypothetical, unnamed Anybody Else. Only if they voted for Anybody Else would they be presented with a list of actual candidates, not including the incumbent. Such a system is pretty clearly tilted against incumbents.
Aside from an error in the analogy,⁶ nothing in this quote explains why a two-question ballot favours change. In fact, Global News reporter Keith Baldrey – who himself appeared quite opposed to reform – put forward the opposite theory: that the referendum might have been “designed to fail”.⁷
By choosing a title with the phrase “stacks the deck”, the Globe and Mail endorses and transmits a political party’s campaign message.
A month before the editorial, the Globe and Mail quoted Andrew Wilkinson, leader of the BC Liberals, referring to the referendum as a “stacked deck”. The party had been using this language for many months prior, as the Globe would have been well aware. The Globe’s opinion piece, and others like it, allowed this messaging to remain effective throughout the campaign.

Vaughn Palmer: Eager watchdog walks into NDP ‘trap’ on wording for electoral reform

This July 13 (2018) Vancouver Sun column appeared shortly after BC Chief Electoral Officer Anton Boegman suggested changes to the wording of the referendum questions to make them more clear to voters.

These are the changes Boegman recommended.

Original Wording	Recommended Wording
Which should British Columbia use for elections to the Legislative Assembly?	Which system should British Columbia use for provincial elections?
Vote for the voting systems you wish to support by ranking them in order of preference.	Rank in order of preference.
Rural-Urban PR	Rural-Urban Proportional (RUP)

Accordingly to Vaughn Palmer, the referendum ballot was so obviously flawed that Boegman should have refused to improve it.

Undaunted by the lack of clear direction from the public, Eby concocted a two-ballot structure that fudged Horgan’s promise of a single question.

The first offered what Eby’s own report characterized as an “apples to oranges” choice between a specific voting system (first past the post) versus a concept (proportional representation).

The second question invited people to rank three versions of PR chosen by Eby himself. Two were “not in currently in use” anywhere. For the third, Eby deliberately left a dozen key aspects to be decided after the referendum by a committee where New Democrats and their partners, the Greens, would have a majority of the votes.
Palmer explains what he means by “promise” in another article:
“You are going to give them one system to vote on?” the NDP leader was asked on the eve of the election. “Yeah, yeah, exactly,” he replied.⁸
The “apples to oranges” comment misrepresents Eby’s report, which was referring to a different ballot design (see page 52). Palmer may have been inspired by Andrew Coyne’s June 4 column. Again, the first question on the referendum ballot was simply asking if British Columbians wanted change. Listing “Proportional Representation” and “First Past the Post” and as the two options was less biased than asking people to choose “Yes” or “No”.

Palmer questions the PR voting systems Eby selected for the second question. All of the systems had either appeared on previous public votes in Canada (DMP⁹ and MMP¹⁰), or were a combination of such systems (RUP¹¹). The options were normal and the government was well within its rights to ask for a mandate to proceed as outlined in the report.

Contrary to Palmer’s claims, there was no sensible reason for the Chief Electoral Officer to refuse to offer superior wording for the referendum questions. Boegman did not fall for any “trap”.
The main purpose of the piece is almost certainly not to expose any wrongdoing on the part of the Chief Electoral Officer, but rather to undermine trust in the referendum process and thereby encourage people to vote against change.
After the referendum, official No side president Bill Tieleman claimed his anti-PR ads were responsible for the 61%-39% defeat of proportional representation. Mr. Tieleman takes too much credit. For many British Columbians, the issue became a matter of trust rather than a question about what type of voting system they preferred. This was largely due to the way the mainstream media chose to cover the topic.

As previously noted, the media first undermined trust by declaring the referendum had been designed in a manipulative fashion to produce a particular outcome. The experts did not support this theory, and even the journalists couldn’t agree on which outcome had been unfairly advantaged. But they printed it anyway. Palmer’s July 13 piece is an attempt to expand the scope of the public’s distrust by creating suspicions about Elections BC. The next article also strives to undermine trust, but targets the advocates of proportional representation.

Rob Shaw: Proportional representation debate off to a sad, misleading start

In this Vancouver Sun column from October 28 (2018),¹² Rob Shaw reflects on the first week of voting by first criticizing an official No side campaign ad.
The official anti-PR group found itself on the defensive last week for a TV ad that depicts marching soldiers and violent protests, while warning of neo-Nazis and extremism under European governments that have adopted PR. It’s a shameful bit of fearmongering that does more to undermine the credibility of the No side than backstop any potential benefits of FPTP.
The ad Shaw refers to is more than deserving of this assessment. In fact, one could say it epitomizes fear-based propaganda. The ad is so over-the-top, it’s hard to believe it’s a real campaign video from 2018 and not some sort of parody. Take a look.
However, the purpose of the column is to convince the reader that the Yes side is equally culpable for the low level of discourse.
The Yes side is just as misleading, in its own way.

It has been promoting several videos its ally groups have created in which energetic young hipsters and animated characters pretend to break down the referendum using “facts” and “truth” — while glossing over important unanswered questions about the three PR models, only focusing on the negatives of FPTP and over-exaggerating or distorting things just enough to put the very concept of “truth” into question.
Shaw argues the Yes side is “just as misleading” as the No side, which is a remarkable claim given what we’ve seen above. His only evidence is a pair of pro-PR videos, one with “hipsters” and one with “animated characters”. The article fails to name or reference the videos, so readers had to take him at his word. But let’s actually examine the videos and compare them with the No side’s ad.

I happen to be quite familiar with the video featuring the animated characters. It was made pro bono by BC artist Brendan Gallagher, and I helped write the script. Here’s a 45-second clip. It’s true that the video has an anti-FPTP bias. It is, after all, a campaign video. What’s important is that we took care to ensure all the information presented was accurate. The examples in the clip were reflective of past elections in BC. We did not exaggerate the distortions caused by FPTP. Shaw describes the video with the words “facts” and “truth” in quotes, yet fails to point out even a single example of misleading information. The video was made independently of any of the registered Yes side advertisers. When a few Fair Vote Canada campaign teams shared it on social media, they were not being “as misleading” as the No side was when they spent tens if not hundreds of thousands of taxpayer dollars producing and airing the ad with the “goose-stepping neo-Nazi soldiers”.¹³

Here’s a 45-second clip from the other video, the one with the “energetic young hipsters”. This video is shorter than Brendan Gallagher’s, yet broader in scope because it covers both the voting systems on offer and the effect PR can have on law-making. The simplifications they made to fit everything in are reasonable. It is not “sad and misleading” that a group of relatively young people volunteered their time to produce a high-quality video engaging British Columbians in the referendum.
After attacking the credibility of both sides, Rob Shaw represents himself as a more trustworthy source of information by providing a seemly balanced yet severely misinformed comparison of First Past the Post and proportional representation.
PR’s main strength is its promise of a system that decentralizes the power structure and redistributes it back to MLAs.
Proportional representation has many qualities that could be characterized as its “main strength”. However, no genuine expert would describe a transfer of power from party leaders to MLAs as the main reason to switch to PR.¹⁴ Rob Shaw could have said PR’s main strength is to reduce distortions between the popular vote and the seat distribution, as shown in the animated video. He could have written that PR increases the level of cooperation that is usually necessary to pass laws, as explained in the “hipster” video. He could have written that PR makes election campaigns competitive in ridings that are currently safe, or that it increases the diversity of views in the legislature, or that it gives every voter a more comparable degree of influence over the number of seats won by each party. Any one of those qualities is defensible as a primary benefit of PR. What Shaw actually wrote is more misleading than any statement in the two pro-PR videos he equates to the No side’s neo-Nazi ad.

Vancouver Sun Editorial: Vote ‘No’ in NDP’s badly flawed electoral reform referendum

On October 29 (2018), a day after they published the Rob Shaw column, the Vancouver Sun produced one of the most error-ridden of all pieces of writing on the 2018 BC Referendum.

Although the Vancouver Sun waited until the voting period to make their position official, they had already been active for many months promoting the No side’s narrative about the referendum being “flawed”. The following articles are from Vaughn Palmer alone.

NDP stacking the deck in favour of proportional representation (April 24)

Electoral reform suspiciously smells like a blank-cheque vote (May 15)

Eby to keep electoral reform vote ‘simple,’ as in simply confusing (May 31)

Eager watchdog walks into NDP ‘trap’ on wording for electoral reform (July 13)

Horgan selling positives of electoral change, but there are negatives (Sept 7)

Leap of faith with Horgan on electoral reform has dark side for voters (Oct 4)

The editorial falsely reports that if proportional representation were to be adopted, riding boundaries would be drawn by politicians.
Then there is the fact that voters are not being given a clear choice as in past referendums about what new voting system would replace FPTP. The ruling politicians, those with the greatest vested interest in the outcome, are telling British Columbians to trust them to work out dozens of details, including what kinds of MLAs will sit in the legislature and the size and shape of the ridings. [emphasis mine]
It is well known that Canadian elections avoid the problem of gerrymandering by requiring the “size and shape” of ridings to be determined by independent electoral boundaries commissions. There was never any credible indication that this policy would change depending on the result of the referendum. In fact, the Attorney General’s report states that, “if any of the proposed PR systems are adopted, an independent Electoral Boundaries Commission would be required to recommend new electoral boundaries for the province” (page 60).

The other claim made in the highlighted sentence, that politicians would decide “what kinds of MLAs will sit in the legislature”, is also incorrect. There would be either local MLAs or a combination of local and regional MLAs, depending only the decision made by voters on Question 2 of the referendum.
The editorial then takes one of the No side’s most deceptive insinuations – that MLAs might be chosen by parties after elections – and makes it explicit.
For example, under mixed-member proportional representation, one of the three systems being considered, 40 per cent of MLAs would not be directly elected. They would come from party lists and we don’t know whether voters would be able to choose them or whether parties would decide after elections who would fill those seats, possibly insiders and patronage appointees. [emphasis mine]
The notion that MLAs would be appointed after an election is completely false. Regardless of which PR voting system prevailed or how it was implemented, Elections BC would continue to declare which MLAs got elected by applying the electoral formula to the votes cast. Parties would continue to nominate candidates, but that process would still occur before elections and not after.

At the time the editorial was written, there was a remote possibility that some candidates would be elected from closed party lists. The way these lists work is common knowledge among those familiar with voting systems. As explained in the report, “voters endorse a party and its candidates as listed in the order put forward by that party” (page 59). It is not at all the case that “parties would decide after elections who would fill those seats”, as stated by the Vancouver Sun. While there are legitimate arguments to be made for and against party lists of both the closed and open variety,¹⁵ there was no justification for misinforming British Columbians about how they work.

If the Vancouver Sun wanted to give an example of a decision the legislature would have had to make about Mixed Member Proportional after the referendum, they had nine options to choose from on page 71 of Eby’s report. Why did they avoid mentioning the nine decisions that the Attorney General had explicitly listed? Why did they invent a new possibility that was never on the table?

One theory is that the Vancouver Sun was influenced by a successful yet deceitful campaign video from San Sullivan, a former mayor of Vancouver. Here’s a 15-second clip from the video.At the end of the clip, we hear Sullivan say, “You give your vote to a party, the party chooses these MLAs for you.” The cleverness of this quote lies in the order of the two statements. First he says the voter chooses a party. Then he says the party chooses the politicians. The video may not explicitly say the party’s decision comes after the voter’s decision – as that would be an outright lie – but the intent is clearly to make the viewer believe it.

The video insinuates that parties exercise their influence over who gets elected after the voters, and hence after elections. As with many of the elite members of the No side, Sullivan stopped short of making this particular lie explicit. He does not speak the words “after elections”. But the editors of the Vancouver Sun went ahead and printed it anyway.

To give credit where it’s due, there were some news organizations – mostly small, independent, or local outlets – that interviewed genuine experts and admirably covered the 2018 BC Referendum. But the mainstream media showed little interest in researching the topic, and ended up playing far more of an influential role than an informative one.

Newspaper columnists and editors repeatedly claimed that the referendum was fundamentally flawed, and that those who facilitated or promoted the reform process could not be trusted. Hardly any information was given to back up these claims, except in the case of the Vancouver Sun editorial where the evidence provided was false. Despite the factual errors and unsupported arguments, reasonable people would have believed the media’s coverage to be reliable due to the professional standards journalists are expected to uphold.

End Notes

For an example of a well-researched article on the BC Referendum, see Electoral dysfunction? Experts abroad weigh B.C. voting systems in referendum by StarMetro Vancouver reporter David P. Ball. The provocative title and main image do no favours to the Yes side. But the content is balanced, and the piece features interviews with two local experts, two international experts, and the inventor of one of the proportional voting systems.
Perhaps Andrew Coyne is assuming all options have an equal chance of being selected. In that case, FPTP would have a 50% chance of winning Question 1 on the actual referendum ballot, whereas it would have a 25% chance if it were listed side-by-side with three PR options. But that logic doesn’t hold water, as voters don’t make decisions at random.
Andrew Coyne’s preferred referendum ballot format would have let No voters rank the three PR systems. The problem is, their rankings would have had no effect. In all likelihood, all but one PR system would have been eliminated on the first several rounds of counting, and votes for these systems would have mostly transferred to the remaining PR option. If the last PR option prevailed over the status quo, it would have been the system chosen only by Yes voters. Anyone who marked FPTP as their first preference would never have seen their vote transfer to any of the PR options.
DMP ended up garnering a respectable level of support in the referendum. For every 10 voters who ranked MMP as their 1st choice, there were 7 who ranked DMP on top. RUP also proved quite popular, receiving almost as many 1st choice votes as DMP. These results suggest David Eby did in fact choose PR options likely to appeal to significant numbers of voters.
The referendum ballot format had already been announced by the time Fair Referendum started their advertising campaign. They knew full well their changes could not be adopted at that point. In fact, they were counting on it. One of their “demands” was that all four systems receive equal funding. That would have given the Yes side three times as much money as the No side.
In the Globe and Mail’s hypothetical election, you have to first vote for “Anybody Else” in order to choose among the non-incumbent candidates. In the referendum, all voters were allowed to choose among the alternative voting systems, even those who supported the status quo. But even if the Globe hadn’t made this error, there’s no reason why the format of a referendum ballot should match the format of an election ballot. Voting on an issue is different from voting for a person.
Keith Baldrey’s suggestion that the referendum was designed to fail is slightly more logical than the Globe’s theory that it was “stacked” in favour of change. But there’s a much simpler explanation for the referendum’s design. In all likelihood, David Eby’s aim was to delegate the decision-making and limit his own influence over the new system. He gave voters the choice of which PR system would be adopted. And he handed a future all-party committee and electoral boundaries commission the responsibility of deciding a number of details for the winning system. There are, of course, pros and cons to this approach. But there was never any evidence that Eby was trying to give either side an unfair advantage.
“Yeah, yeah, exactly” is not quite an election promise as Palmer implies in his piece. It certainly doesn’t suggest the same degree of commitment as Trudeau’s, “the 2015 election will be the last federal election using First Past the Post”, for example. It is fair, of course, to criticize John Horgan for answering a question in careless fashion. Nevertheless, it’s beyond absurd that the words “yeah, yeah, exactly”, spoken off the cuff, should have been interpreted by Elections BC as the sole legitimate basis for the design of a multi-million dollar referendum.
Dual Member Proportional was an option on Prince Edward Island’s 2016 Plebiscite on Electoral Reform.
Mixed Member Proportional appeared on the public votes in PEI 2005, Ontario 2007, and PEI 2016 (which it won).
Rural Urban Proportional is a combination of MMP and STV, the latter of which was the alternative in BC’s 2005 and 2009 referenda.
The article is dated October 30 due to an update, but I assume the original publication date coincided with the newspaper’s tweet.
At the time Rob Shaw’s piece was published, the official No side had also been using some of their government-granted $500,000 on an ad implying that proportional representation would leave some voters without a local MLA. This ad was as misleading as the neo-Nazi video, as the creators were well aware that every voter would still have had at least one local MLA under any of the PR systems. Some of the systems would have provided list MLAs as well, and these list MLAs would also have had regional offices for British Columbians to visit or phone.
The issue of party loyalty became a red herring throughout the referendum campaign. In theory, certain implementations of PR could reduce party discipline, as explained by the University of Victoria’s Daniel Westlake. But in the end, MLAs will only become more independent if voters choose to reward independence. Westlake cautions, “It is important not to overstate the influence that electoral systems have over party discipline.”
Open lists allow the voter to mark an ‘X’ beside their preferred candidate on the party list. The order in which candidates are then elected is based on the number of votes they receive from citizens.

Rhys Goldstein's Blog