Quebec electoral reform update: Public hearings bearing fruit

Wilfred Day planted the following seed regarding electoral reform in Quebec at my previous thread on Canada’s dysfunctional FPTP system (at the federal level). Quebec electoral reform deserves to a planting of its own, so here it is. I am also copying below a comment that Vasi planted in response to Wilf.

After public hearings that began Nov. 1, 2005, and continued for 25 days across Québec, public reaction has borne its first fruit: the Citizens’ Committee has reported. Nine Members of Québec’s National Assembly, and eight citizens selected by a structured random process from interested volunteers, have consulted the general public and particular experts on the government’s draft bill for an MMP system.

The first shoe has dropped (what kind of fruit is a shoe?), but it has no legal weight. The eight citizens sat at the same table as the nine MNAs and had the same right to ask questions, but have no vote on the Report of the Special Committee (the nine MNAs.) The thud, if any, will come from their report.

Six Citizens have written a masterful 59-page report with no technical assistance. (One more wrote his own, while the eighth signed nothing.) It calls for an improved MMP model with province-wide proportionality, with 40% compensatory seats assigned to regions and filled from regional lists with a 5% threshold using a two-vote model. To the surprise of those who thought citizens would always want an open-list model, they want closed lists in each of Québec’s 17 administrative regions, ranging from a few little three-seat regions to Montréal’s 30 seats (18 local, 12 regional list.)

It also calls for fixed election dates. Uniquely for a citizen panel, it calls for rules to maintain stable government: for example, that a government cannot be defeated in the House except by the formation of a new coalition government holding a majority of seats. (They did not attempt to define exceptional circumstances permitting an early election.)

Will the government pay any attention to the Citizens’ recommendations? The Minister says he too had some reservations about the draft bill, and will take the committee’s recommendations as a starting point for presenting a new version of the reform in coming months. We’ll see.

As a PR wonk, what interested me is the Citizens’ formula for assigning the 50 compensatory seats to the 17 regions. All they said is “start by giving Party A a seat in the region where it is most under-represented” — presumably by highest average, as the draft bill uses — “and continue compensating parties under-represented in regions so long as the region has seats available” until the party’s seats are all assigned. Unlike Germany, the regions have a fixed number of seats, making the assignment process far trickier than in Germany.

Suppose a fourth party won 7 seats. Would parties go turn-about? Then all seven of their seats would be among the first 28 assigned by this process, leaving the large parties to take the left-overs. Or would they go in order of which party has the highest average (is most under-represented) in any region? If so, the fourth party’s 7 seats will mostly be assigned at the end of the process, and some will go to regions which have a seat left over but where the fourth party had little support.

Here the citizens’ talents hit a wall. They ask the Special Committee staff to do simulations of how this would work and attach them to the Citizens’ Report (not attached). Once again, as in BC, we find citizens needing experts to help them write their MMP model, and no experts assigned to that task.

Here are Vasi’s additional remarks on this subect:

In order that list MNAs not become immune to being thrown out, they recommend a limit of two consecutive terms as a list MNA. They do want to permit double-candidacy (list and district), however.

Apparently Quebec law already provides parties with election financing dependent on the number of votes received last election–with a bonus if a certain percentage of candidates are women. The proposal wishes to increase the bonus, increase the required percentage, and make it apply to elected members rather than candidates. There should also be a separate bonus which applies to minorities. Moreover, if this measure does not result in significantly increased representation of women within two elections, then parties should instead be required to alternate between men and women on the lists. (Note that apparently English-speakers and those of minority religions are not considered “minorities”–not that I’m bitter or anything.)

Thanks, Wilf and Vasi, for keeping the orchard planted with the latest hybrid varieties of fruit! (And I am still pondering that question about what kind of fruit is a shoe.)

0 thoughts on “Quebec electoral reform update: Public hearings bearing fruit

  1. Points of clarification:

    1. Parties are to be given incentives under the government’s draft bill (not under existing law) to run more women and more ethno-cultural minorities. They will get even higher incentives for electing them if the Citizens get their way. Yes, that includes English-speaking minorities. The former “Westmount Rhodesians” are now on a par with Haitian taxi-drivers, in need of affirmative action.

    2. Even the government’s draft bill proposes that incentives continue until women hold 50% of the Assembly seats. The Citizens say that, by giving incentives for the number of women elected rather than candidates, this will encourage the use of alternance (“zippered” lists) which they encourage parties to use. They say this should be reviewed after two elections, but they don’t quite say “then parties should instead be required to alternate.” They say if the number does not rise after two elections — at 32% Quebec already has the highest number of women in a Canadian legislature — then “it would be necessary to consider obligatory alternation.”

    3. If I had read the whole report before posting, I would have seen that these diligent citizens prefer, as they should, Ste-Lague rather than highest average.

  2. I appreciate the update. I wonder what the motivation for the support of closed lists was.

    As for the question of what kind of fruit is a shoe, I’d have to go with fruit cobbler.

    {GROAN!–MSS. Actually, that’s a really good one!}

  3. Wilf,

    Thanks for correcting my mix-up between existing legislation and the govt’s draft legislation.

    Re: English and minorities, I didn’t mean to say that anglophone individuals are excluded from being considered members of minority groups. Rather, I meant that anglophones as a group are not considered a “minority” for the purposes of electoral equality–even though later on in the report, it actually refers to anglophones as a “minority”. I think it’s unfortunate that Canada is moving from a nation with two languages, towards a nation with two areas which speak separate languages. Slowly but surely, both Quebec’s anglo community, and the francophone communities in Ontario (and Manitoba, Nova Scotia, maybe even New Brunswick) seem on the decline.

    You’re right that the paper only calls on the director-general to “consider obligator alternation”. However, it says this should be considered not just if there’s no rise, but if there’s no “significant rise”. I suppose it’s open to interpretation what “significant” means. Judging by the details of the plan, it’s probably somewhere between 35% and 45%, but where exactly is hard to say.

    Some other random notes:

    The report comes out against almost any use of electronics for either voting or counting votes. Take that, Diebold!

    A new option for “intentionally spoiling one’s vote” is suggested, so that ballots which are improperly marked can be separated from protest ballots. I really like this, even if this protest option can’t win (a la “None of the above” in Russia). Interestingly, you can vote for “spoil” in your district and still have your province-wide vote counted (or vice versa).

    Here’s a real kicker: Wilf was talking about the possibility that a party would be assigned a seat in a region where it has little support. The report actually discusses the case where a party wins a list seat, but the only available seats are in regions where the party did not run. In this case, the report recommends simply “canceling” the seat that was one, though it seems hesitant about this. It remains unclear what would happen to this seat. Would it remain unfilled? If another party picks it up, would we now have to go back and reallocate all the seats?

  4. Declan,

    Here’s a rough translation of the relevant section:

    “We recommend closed lists rather than open ones. First, because closed lists offer parties a way to promote the election of women and minorities, by way of their list placements. Second, because this frees voters from having to learn about too many candidates (as many as dozens in some cases), which would probably cause them to vote for the only candidates which they have heard of.”

  5. “closed lists offer parties a way to promote the election of women and minorities, by way of their list placements.”
    Polls in Canada have shown that 90% of voters want more women in politics. Open lists might help women slightly, as they did in Sweden, except for the risk that incumbents, big names, and rich men with big private ad budgets would have an advantage. The bigger concern is minorities. Parties in Quebec have a recent history of trying to promote minorities, only to find a risk of the minority candidate failing to win a contested nomination. The voters at large might rank a familiar name higher. This has happened, even in left parties, in some European open-list models.

    “this frees voters from having to learn about too many candidates (as many as dozens in some cases), which would probably cause them to vote for the only candidates which they have heard of”

    Think especially of Montreal, with 18 single-seat districts and 12 regional list MNAs, the very place where minorities should be elected.

    These points would have been well-understood by the eight Citizens, such as Mustapha Acharid (cross-selected by a computer as the male from the 32-45 age cohort and the region of “Laval & Laurentides”).

    But they were a marvellous group: Yohanna Loucheur, with a Masters in international relations and a second Masters from London School of Economic and Political Science in methodology and research, who spent two years in the Republic of Guinea, as well as time in France and England. (She was selected as the woman from the 32-45 age cohort and the region of “Outaouais and Abitibi-Témiscamingue.”) Guillaume Boivin is working on his Masters in History (the male from the 18-31 age cohort and the region of “Capitale-Nationale & Chaudière-Appalaches.”)
    Charles Gaboury has a Masters in Regional Planning and Regional Development (the male from the 60+ age cohort and Montréal.) Random selection works nicely when the pool are the 2,300 volunteers most interested out of a population of 7.5 million.

  6. Thanks, Vasi, Wilfred. I don’t reallly agree with the reasoning, but then I’m not really a fan of ‘activist’ electoral systems.

  7. Declan, I don’t much like anything that smells of quotas, either. But Quebec isn’t exactly known for hesitating to use laws to push social goals. I’m actually surprised that nobody (to my knowledge) has proposed incentives towards the election of citizens of lower income, or a wider range of professions.

  8. Oops, that last comment of mine was a little confused. Let me try again.

    Closed lists could indeed increase the representation of minorities. Moreover, they could do this even with color-blind government policies. For example, say the PQ has a list that’s 50% women and the ADQ list is mostly men. The PQ could make an election issue out of this difference, causing the election of more women by convincing voters that it’s important. I don’t see anything wrong with this, and I can understand why people consider this a benefit of closed lists.

    What I find objectionable are the government incentives towards the election of minorities. With a proportional electoral system, it’s likely that smaller parties will form, and some of them may have a homogenous pool of potential candidates. If enough voters wish to elect a “Men’s Rights” or “Pure Laine” party, well I think that’s silly, but it shouldn’t be punished by the government.

  9. I’m surprised that anyone would find government incentives towards the election of minorities objectionable. Such incentives for the election of women are a core feature of France’s Parity Law. Similar incentives for the election of women will be a core feature of the Quebec election reform. Why should similar incentives not apply to encourage parties to elect more minorities?

  10. Er, I guess was I unclear again, sorry Wilf. I don’t mean that I want incentives for the election of women, but not minorities. Rather, I oppose incentives in both cases. Simply replace “minorities” in my last post with “minorities and women”.

    I consider it morally troubling for an electoral system to care about a candidate’s personal status, whether in language, religion, ethnicity, gender, class, or anything else. For those who want more equal representation–including me–the solution is to vote accordingly. Sometimes one may have to choose a member of under-represented groups whose politics one opposes. But this unappetizing choice isn’t a unique situation in elections, it’s quite common to have to choose between priorities. (One is unlikely to be able to vote for a NDP member who supports armed intervention in Darfur, for example.)

    I heartily support electoral systems that make it easier for voters (rather than the government) to make their preferences on equal representation known, without sacrificing other choices. For example, a flexible-list system could allow one party to present multiple lists, and then a voter could select the “Women’s List of the Apple Party” to favor representation of women.

  11. If one supports electoral systems that make it easier for voters to make their preferences on equal representation known, then clearly something must be done to make parties give voters this choice. Polls have shown that 90% of Canadians want to see more women elected. Yet parties persist in nominating only 26% women (Liberals) or 12% women (Conservatives). Partly this is because the “gatekeepers” (local ridings associations) are nominating only one candidate at a time, so that any system with some lists could lead to some improvement. But some parties will still not give voters the choice without some incentives. This, of course, is why France enacted their Parity Law.

  12. The final report of the Select Committee on Quebec’s planned MMP system has been published in French. (English version not yet posted.)

    I’ll withhold detailed comment until I read the English text. For the moment, suffice it to say that they agree both with the critics who called for larger districts, and with the Citizens’ Committee who called for province-wide proportionality (making district size largely irrelevant.) Which do they recommend? Unanimously undecided. Either way, the government’s draft bill needs to be improved. A unanimous step forward. And they also agreed the incentives for the election of women need to be strengthened. They were silent on the issue of one-ballot or two-ballot MMP, perhaps tacitly favouring the one-ballot model of the draft bill.

  13. I was part of the Citizen’s committee and just stumbled across this discussion. Thanks for reading our report so attentively. I find the various comments extremely interesting. Unfortunately, there was no real debate in Quebec – after our report or after the “official” ones. And so few people seemed to understand what we were proposing! Perhaps if a real reform proposal is tabled the media will remember our existence…

    This particular thread might be dead now, but if anyone has further comments – especially on the differences between our report and the commissions’, I’d be interested.

  14. Yohanna, no planting at F&V ever dies! (That’s the purpose of the “propagation bench” at the right sidebar: to keep topics going.) Thanks for your comment, and I hope others will have more to say about the Quebec process. I rely on my readers in Quebec for almost all of my information about this process.

  15. I have some (not very encouraging) news about the subject. Benoît Pelletier, the minister for Electoral Reform, has announced that he would put off any reform of the voting system for a few years (article in French only). I believe all the work done will end up being for naught, just like all the previous attempts to reform our electoral system. Still, there are elections coming up in the next few months, so maybe one of the parties will revive this discussion. Not that I’m expecting it, since it really isn’t a current issue in Quebec.

    [MSS note: Thanks for the (not encouraging) news, Marc. You had started it with the statement: “I know that this thread hasn’t been active for a while, but I just stumbled on it…” Not a problem, Marc. One reason for having the “propagation bench” show at the top of the right sidebar of the main page is so that readers can bring up topics from older plantings and continue an updated discussion. No planting in this orchard ever dies, though many do go into extended dormancy!]

  16. Electoral reformers just got a nice Christmas present from Quebec’s Chief Electoral Officer.

    A full year in the making, it’s a detailed technical report which does not overtly recommend anything. It assumes MMP because the government’s draft bill and the Citizens Committee both recommended it. In fact nothing but MMP is really on the table in Quebec.

    It then discusses how to apply it in Quebec, without overt recommendations, but with apparent leanings.

    It discusses the number of regions and regional calculation methods. It finds that Quebec-wide calculation, with parties’ seats then allocated to their regional lists in proportion to the regional distribution of the party’s votes (the federal German method), with nine regions, provides the best representation. This seems to me to be a solution that almost everyone in Quebec will accept.

    As the press release states “this nine-region scenario would be a good compromise for those people looking for both proportional results and a territorial base for list members.” It notes the risk that some regions with low turnouts, or with lots of wasted votes cast for parties getting less than the threshold, could end up slightly under-represented as has happened in Germany, but it says this risk is reduced when “overhang” seats are not added, and when only nine regions are used. It notes that the risk of a region losing a seat through a low turnout should produce higher turnouts.

    On open lists, closed lists, or flexible lists, it notes that no MMP model but Bavaria uses open lists. It relegates no-list MMP to a small footnote. It notes the Citizens’ Committee recommended closed lists, as did most submissions from groups (union and women’s groups) for the usual reasons: to allow parties to balance the lists by gender and by cultural minorities. It notes that the Select Committee received 31 briefs favouring closed lists and only 11 favouring open lists.

    However, it notes that the ACE article on the topic states the disadvantages of closed lists: voters have no way to influence the choice of their regional MNA, and no one can react to developments during the course of the election campaign. It notes that, in 2000, a survey of New Zealand voters showed the majority wanted open lists. (It fails to mention the similar surveys in Scotland noted in the Arbuthnott Report. Remarkably, in 260 pages it never mentions the OCA, not the BC CA, nor the referendum results in PEI, BC or Ontario.)

    Finally, it makes a point of noting the Jenkins Commission in the UK and the Law Commission of Canada both recommended flexible lists. I infer that flexible lists sound like the best of both worlds. (It does not mention the recent report on an elected House of Lords for the UK which also recommended flexible lists.) Even without making inferences, one can say that flexible lists are clearly on the table in Quebec.

    Since many of us had expected the Ontario Citizens’ Assembly to design a regional flexible-list model, this report may help to remind people across Canada that the OCA’s MMP model is not the only one on offer.

  17. I don’t think I’ve heard the expression “no-list MMP” before. Does this refer to systems that award compensatory seats to “best losers” instead of having a second (list) vote? Thanks!

  18. There are many names for the Baden-Wurttemberg model: “best loser,” “best runner-up,” “near winner” (the name invented by Ontario Citizens’ Assembly staff), “list-free” and “no-list.” My ancient journalism training makes me use the shortest.

    I feel it has legs. It was remarkable, during the Ontario referendum campaign, to see dozens of bloggers and other voters, reacting to criticism of closed lists, spontaneously inventing this model. I’m certain none of them had ever heard of Baden-Wurttemberg. This may be unique to Ontario, since Quebecois are far more aware of MMP variations than Ontarians, and they seem to have no interest in no-list. However, if we ever have a Citizens’ Assembly at the federal level in Canada, I expect to see them choosing between flexible list MMP and no-list MMP.

    By the way, although Baden-Wurttemberg uses the old one-vote model, it would be quite feasible to have a two-vote no-list model. In fact it would be an improvement on Baden-Wurttemberg’s model, giving voters more choice and making local MPs more accountable, the same reasons most of Germany shifted to the two-vote model.

  19. Wow, this is a welcome surprise! Perhaps I simply don’t have the right connections, but everybody I know thought the government had put the Citizens’ Committee’s report on a shelf and forgotten about it. Nobody seemed to have any idea that such a detailed study was being prepared.

    The results of the simulations aren’t particularly unexpected: Larger districts yield more proportionality, especially for small parties. Forcing parties to alternate between genders on their lists causes more women to be elected. I guess the most significant result is that there is not much of a difference between nine regions and none at all. I’m a bit disappointed with their simulations overall, to be honest. They seem to be based on a theoretical model which assumes three large parties, two small ones, and a bunch of minuscule ones–essentially what Quebec has now. I’d rather have seen a more pragmatic look at all sorts of different vote distributions, since one might expect a shake-up in vote shares were the voting system to change.

    I’m also a bit puzzled by how the authors divided Quebec in regions. The simulations apparently considered a division into nine regions of equal population, but no mention of this is made in the main document. Instead all three divisions considered in the main text have significant size variances, and include some regions with only one or two compensatory seats. (Here’s a copy of just the relevant part of the text, so you don’t have to slog through 200 pages in a foreign language to see what I mean.) I think perhaps the authors try too hard to keep each “administrative region” intact, when few Québecois really care about these particular groupings–as is reflected by the way the current electoral districts often span regions. It would be nice to see a more equal division, with no region having fewer than eight seats.

    But these are just minor gripes, on the whole it’s great to see a government agency taking electoral reform seriously and putting real effort into studying it.

  20. I just realized that I don’t quite understand how regional seat allocation à la Germany would work without overhangs. With overhangs, the calculation is relatively simple: First you do the national seat allocation by votes to find the seats per part. Next you divide each party’s seats by region. Finally you deal with overhangs by assigning each party MAX(calculated-seats, constituency-seats) per region.

    Without overhangs, you could do the first two steps the same. But suppose some party has N more constituency seats than would be proportional to vote share in some region. We now have to take away N seats from somebody, but who? There’s no fixed number of seats per region, and the number of seats per region is not used in the calculations, so we can’t just subtract from that. The only option I can think of that makes any sense at all is removing the votes and seats of the given party and region from calculation, and starting all over–then adding them back at the end. Is this the obvious method, or is there something I’m missing?

    Does anybody know of other methods aside from the German one for national compensation with regional distribution? I think the “regional calculation with national top-off seats” method has already been discussed on F&V. I could certainly imagine more ways of doing it, but it would be nice to know what’s actually in use.

    Also, thanks to Wilfred for mentioning this report. It’s been reported locally, but not all that conspicuously and certainly not in the English papers. I wouldn’t have wanted to miss it!

    PS: Just to add more options for what to call “overhang seats”, the report uses “sièges en surnombre” meaning “excess seats”.

  21. How would regional seat allocation à la federal Germany work without overhangs (temporary excess seats)? I agree with Vasi.

    The Quebec Chief Electoral Officer’s report is silent on the point, but it says “It is possible that a party can obtain more constituency seats than the total number of seats to which it would be entitled. When such a situation arises, the party retains its surplus seats and the compensation is recalculated by removing from the calculation the over-represented party.” No doubt the same method would be used to remove from the regional calculation any region where the party had won too many local seats.

    What other two-tier calculation methods are in use, other than Sweden’s? Austria’s. Maybe others.

  22. La Presse reports that the Minister for Democratic Institution Reform seems to be putting electoral reform on the back-burner (French). His reasons are:

    * The failure of the referenda in BC and Ontario
    * Disagreement among reform proponents over the preferred new system
    * The difficulty of getting all three parties to support a new system

    He claims that he still personally supports reform. The cynical part of me wonders if this is connected to new polls showing Liberal support on the rise.

  23. Vasi, are there declared (or even widely assumed) positions of each of the parties on reform? If all want a reform, but disagree on the preferred specific form, that’s a pretty bad reason to privilege the status quo. But if the Liberals really prefer the status quo, then it’s easy to see why they would use lack of consensus as an excuse to put the issue on the back burner. And good poll numbers have a way of giving ruling parties under FPTP cold feet about PR!

  24. Just to prove that this thread is not dead, I will note that the Qebec Liberals still need MMP — because they are disadvantaged by the huge federalist strongholds in west Montreal piling up unneeded majorities — but still can’t seem to agree on how to do it.

    Meanwhile, it should be noted that, on Nov. 11 and 12, 2009, Quebec’s National Assembly unanimously voted for Amir Khadir’s motion that reform of the Elections Act ensure “fair representation of political pluralism.” As the government spokesman said “Everyone can agree on the principle, the problem is the how. . . We want a proportional voting system, we must specify which method of proportional representation we put forward. We have resumed the debate, the issue is not dead, the issue is evolving, the question is before us, the issue moves forward.” Slowly.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.