BC voters: what you should know before taking the “HowWeVote” survey
It’s better than the #ERRE fiasco, but still built on false dichotomies
Quick background: In May 2016, there was a provincial election in British Columbia. The then-ruling Liberal party got 40.36% of the votes, and 49.43% of the seats (that is, 43 of 87). Because they didn’t quite reach 50%, the other two parties (NDP and Greens) were able to form a majority coalition against them. A key plank in the coalition platform was to hold a referendum on proportional representation (PropRep for short), so that in the future the 60% anti-Liberal votes wouldn’t get just 51% of the seats (or, as in the election before that, 56% of the votes get just 47% of seats). That referendum is now scheduled for November, 2018; and in preparation, the coalition government has posted a survey for all BC voters (link to take, link to view), asking what they value in a voting method and how the referendum should be structured.
The provincial parties have, as might be expected, lined up for or against proportional representation based on whether or not they were harmed by disproportionality in recent elections: NDP and Greens favor it, while Liberals oppose it. This is ironic because on a national level, Liberal prime minister Trudeau originally campaigned on PropRep; but then, after he’d been elected and it was no longer convenient for his party, deep-sixed it using a public polling process called “mydemocracy.ca” which played up false dichotomies about PropRep. For instance, one question asked voters whether “a ballot should be easy to understand, even if it means voters have fewer options to express their preferences”; there was no way to support both simplicity and choice.
If you don’t know, British Columbia is planning a referendum on switching from horrible FPTP voting to proportional representation (#PropRep) in November 2018. They could lead the way for the rest of Canada, for the US, and for the UK. This is a BFD.
This survey ( feedback.engage.gov.bc.ca/index.php?r=su… ) is laying the groundwork for that referendum. It follows the earlier Canada-wide #ERRE survey, which was deliberately set up to scuttle #PropRep, because Trudeau and the Liberals found it politically convenient to break their promise.
Unlike the ERRE survey, this new BC survey is not deliberate sabotage. But it still falls into the trap of false dichotomies. Modern #PropRep method designs such as PLACE voting (link contains explanation; here are three more links) can allow the best of both worlds. The right answer to the “do you want this or that” questions is “both”.
For instance: “Q16) Which would you prefer, having more small parties represented in the Legislature or a few big parties?” Good #PropRep methods, including not just #PLACEvoting but STV and MMP, can give a happy medium of 3–5 parties. Neither too many nor too few.
“Q17) Do you prefer: MLAs who do what their party promised. MLAs who do what their constituents want.” Um… both? Most voters should have a party choice where those two things are the same!
“Q18a) Ballots should allow voters to support more than one candidate or political party by ranking them in order of preference?” Ranking multiple options is a means to an end: ensuring your vote is not wasted/exhausted if your first choice loses (or gets > enough votes). So ranking methods are better than methods (such as FPTP) where many votes are wasted, but not as good as methods (such as #PLACEvoting) where wasted votes are avoided without fiddling with rankings.
“A ballot should give voters lots of choices, even if it’s less clear how votes get turned into seats.” Gah! Mixing up three separate issues: more choices; simple/complex ballots; simple/complex counting process. A good voting method should definitely have more options for voters. For instance, #PLACEvoting lets voters choose any candidate in any district, or if they want, just choose a party. That’s a lot of options, but with good ballot design, they all fit on a simple ballot.
Note that “More options”≠”More decisions”. For instance, STV has fewer choices than PLACE; instead of all candidates province-wide, you can only pick from candidates in your 5-member riding. But then it has more decisions: you must pick not just 1st choice, but 2nd, 3rd, etc.
If you look at how STV actually works, only one of your various rankings will end up using up most of your voting power. So you end up making several separate decisions, but only one of them really counts. Better to make < decisions but have > options, as in #PLACEvoting.
“Q18c) An election ballot should be easy to understand, even if it means voters have fewer options to express their preferences?” As I said, with #PLACEvoting, you get many options AND a simple ballot. Another false dichotomy.
“Q18d) It is better for several parties to co-operate and govern together rather than one party to govern alone, even if it sometimes takes longer to form government after an election.” This should be up to the voters. One party gets a true majority? Yay! None do? Compromise!
“Q19) Do you prefer: Each electoral district is represented by one MLA Each electoral district is represented by several MLAs Some MLAs represent an electoral district, while others represent larger regions or the province as a whole.” Another false dichotomy (or tri-chotomy?). Under PLACE, each ED would have one primary MLA, but also have one secondary rep per party, so that local minorities would know who to petition. eg, If you’re a Lib in an NDP district, there should be a Lib rep in a nearby district who’s responsible for listening to you.
“Q21) Alongside the option of keeping the First Past the Post voting system, which system or systems of Proportional Representation would you like to see on the ballot? List PR; MMP; STV; MMMajoritarian; Other, describe” I’m not saying they should have included PLACE because I realize that almost nobody has heard of it yet. Same goes for other nice new ideas like RUP fairvote.ca/ruralurbanprop… . But they should have had an option for “Some newer proposal that experts agree is worth exploring, even if it hasn’t been tried elsewhere”.
That’s not at all the same as “other, describe”, because it gives people the chance to be open to new ideas without assuming they are voting method experts.
So, how to answer the HowWeVote survey?
FVBC has an article on how to respond to this survey. Tweet: twitter.com/fairvotingbc/s… @fairvotingbc Article: ymlp.com/z3ZCbM @FairVoteCanada article: fairvote.ca/bc-government-… Both articles are good advice, but deserve deeper discussion, thus this thread.
Both articles take a skeptical view to the survey, as containing traps for the unwary #PropRep supporter. After what happened with #ERRE, that’s understandable. My thread twitter.com/bettercount_us… discusses the false dichotomies this survey has.
But it’s unfortunate that we to some degree have to choose between honest and strategic responses to this survey. Also ironic, because that’s the same situation FPTP puts us in every election, and what we’re supporting #PropRep to break away from.
(Geeky aside: I know of course that the Gibbard-Satterthwaite theorem shows no voting method, not even #PropRep, is entirely free from strategic voting possibilities. But using that to argue for keeping FPTP over PropRep would be like saying “vote Hitler because nobody’s perfect.”)
As structural reformers, we should recognize that the most important question here is process. A badly-structured referendum could defeat #PropRep. I have faith that a well-structured one will choose an excellent PropRep method.
So that means the most important two questions on the survey are Q8 (referendum ballot type) and Q9 (other comments). Both FVBC and FVCan recommend pushing for a 2-part ballot question: first, FPTP vs. PropRep; then, ranked ballot on which kind of PropRep.
I think this (2-part question) is essentially a good idea, and certainly I don’t want to undermine FVBC and FVC who have better knowledge of what’s possible. So yes, strategically: what they said. But honestly? We could do even better.
What’s the utopian ideal structure for this referendum process?
- Step 1: Citizens’ assembly to analyze options. BC had one of these back in 2004 but that’s a long time ago; new, better options have emerged, and public awareness has improved.
What kind of thing should this ideal Citizens’ assembly be looking at? Here’s a few: PLACE voting: wiki.electorama.com/wiki/PLACE_FAQ Dual-member proportional: en.wikipedia.org/wiki/Dual-memb… Rural-Urban proportional: fairvote.ca/ruralurbanprop… Heart and Head: heartandhead.ca/home etc.
That may look like a bewildering array of choices. Certainly it would be far too many to put on a referendum. But that’s what a citizens’ assembly is for: to dive into details. Also, these proposals have a lot in common. All are essentially hybrid forms of STV and/or MMP.
Is a citizens’ assembly even possible at this point before a November 2018 referendum? Not unless somebody gets the ball rolling *right now*. And from the looks of things, that’s not part of the provincial government plan.
So I’m willing to follow the lead of FVBC and FVCan on this one. If they drop the citizens’ assembly idea, np. Not a lot I can do about it from here at Harvard. But if anybody wants to organize a CA, then I want to lend any support I can. dm or email: fname.lname at gmail
- Step 2: Citizens’ assembly chooses 3 top methods. This should include: STV w/5member districts/ridings; MMP with *open list*; One newer proposal like the ones I listed earlier. Not worth including: MMM (multimember majoritarian, aka “parallel voting”); any closed-list idea.
- Step 3: two-part referendum. Question 1, FPTP vs #PropRep; Question 2, best form of PropRep. For q2, both FVBC and FVCan are suggesting using IRV / ranked voting to pick. Sigh. IRV is better than plurality but other methods are yet far better.
Don’t wanna get too distracted by this old IRV/approval debate, so up front, I’m gonna say that strategically I embrace IRV recommendation here. Bickering on this minor point would be counterproductive. But honestly? Approval voting is far better, and 3–2–1 voting better yet.
OK, that’s it for process questions. Now, on to the false dichotomies. Several of these I already discussed in my earlier thread, not gonna repeat. Here’s the link to the earlier thread again: twitter.com/bettercount_us…
Q5: values. FairVote recommends avoiding “easy to understand” as a value. While I understand their logic, I disagree. Showing that you can value both proportional results and simplicity is useful. And proposals like MMP, PLACE, or DMP offer ballots that are very easy to understand.
FairVote BC also says to avoid both “focus on local community” and “focus on province as a whole”. Again, I disagree. The very illogic that two contradictory positions would both be seen as anti-PR shows that this is another false dichotomy. Choose either of these if you want to.
As for “single-party majority governments”, yes, avoid that BS. A good proportional method like PLACE will actually not lead to excessive party splintering, but this is not the place to have that debate.
Q7c: “It should always be clear which party is accountable for decisions made by the government, EVEN IF this means that decisions are made by only one party.” More false dichotomy bs. Best to answer “strongly disagree” or “prefer not to answer”.
Q21, which systems? (pedant alert: “method” is a better term here than “system”. System is a superset; also includes voter and candidate eligibility, counting authority, calendar, etc.) I’d suggest “PLACE voting”. Or at least, some new proposal of those I listed above.
Q24, extra comments: I agree completely with @FairVoteCanada. You should suggest a two-part question, and point out that hybrid/customized methods should be an option.
Sorry, much of this article is copied from two twitter threads and lightly edited, so the paragraph separations may come off as a bit choppy.