Saturday, December 12, 2009

How would Canada look if Michael Fortier had won in Vaudreuil-Soulanges last year?

How would the current history of Canada differ if Michael Fortier had won in Vaudreuil-Soulanges last year?

And why didn’t he?

It must have looked like a winnable riding, only 72% francophone in 2006, yet part of Quebec’s fast-growing Montérégie region (87.4% francophone).

Provincially the PQ won it only once in 1976. Otherwise it was solidly Liberal, safely held by Daniel Johnson who was premier in 1994.

The Bloc could not take it even in 1993 when the Bloc won 54 seats, nor in 1997 with a different candidate, nor in 2000 with a third candidate.

It had been held by the Tories from 1958 to 1963, while Jack Layton (born in 1950) was growing up there. It was held by the PCs again in 1984 to 1993. (Jack's father Bob Layton was a Liberal when Jack was young, as was Jack, and Bob tried for the local Liberal nomination in Vaudreuil in 1972, but in 1984 Bob Layton ran and won for the PCs in Lachine.)

Yet the Bloc took it in 2004 and holds it today.

Fast-growing Montérégie is typical of what winner-take-all does for the Bloc. Its candidates won 10 of the region’s 11 seats last year, although it got only 45% of the vote. A democratic voting system would have let Conservative voters, with 18% of the vote across the region, elect two MPs, along with two Liberals and one or two New Democrats. And one of the two Conservatives would surely have been Michael Fortier.

Similarly, there would have been three Conservative MPs from Montreal Island (perhaps including Hubert Pichet and Andrea Paine), and two from fast-growing Laval--Laurentides--Lanaudière (no doubt including Claude Carignan).

In Vaudreuil-Soulanges in 2008 the incumbent Bloc MP whom Fortier would face, Meili Faille, must have looked like as much of an “accidental MP” as Quebec City’s three surprise MPs of 2006. (Sylvie Boucher, Daniel Petit and Luc Harvey were the “accidental Tories” as Phil Authier famously called them. Phil’s great story is no longer on the Gazette website, but can be found here.)

In 2004 the Bloc had run a new young candidate Meili Faille, 31, who turned 32 just 10 days before the election. Her father had been the defeated PQ candidate in Vaudreuil in 2003, as he had also been in 1998.

Her mother Feng-Chi and her father Yvon Faille had named her “Meili,“ a Québecois version of the Chinese name Mei Li which means “beautiful.” Her mother tongue had been Mandarin.

Meili had left the riding in 1989 when she finished high school, going to Ottawa for university, getting a Business Administration degree. She had worked at Employment and Immigration Canada from 1993 to 1995 as a Project Manager for the International Service group. Her son Jasmin was born in 1996, and she had then worked for an IBM affiliate LGS Group as a Project Manager from 1996 until her election. In her team at LGS Group she was its only Mandarin-speaker, which was sometimes very useful.

To her surprise, she had won in 2004 with 44% of the vote. In 2006 Liberal star Marc Garneau thought she was easy prey, but she had held on, although her vote had slipped slightly to 43%.

Yet in 2008 Michael Fortier crashed and burned against her, when she still got 41% but he got only 24% while the new Liberal woman candidate, 28-year-old Brigitte Legault, recently president of the Quebec Young Liberals, got 21%. (She had been appointed to run there as a consolation prize; like other Liberals she had hoped to run in Outremont.)

There was more to Meili, a double giant-killer, than Fortier had first thought.

Her father Yvon had been a Catholic priest, working as a missionary and teacher in Taiwan in the 1960s. Her mother Feng-Chi, 12 years his junior, had once been his student, and became a linguist. Their relationship began while she was working as a translator on a U.S. military base in Taiwan. He left the priesthood and they married. They moved to Quebec in 1970, where Meili was born June 18, 1972, in Montreal. A younger sister and brother followed.

Yvon got a job as a teacher at College Bourget in Rigaud, 7 km from the Ontario border, where he would become his daughter’s favourite teacher. Meili grew up on a farm just outside of Rigaud, just upstream from Jack Layton‘s hometown Hudson, on what has become the limit of Montreal‘s commutershed. (The AMT runs 13 trains a day to Vaudreuil but only one to Hudson and Rigaud.)

Her grandfather Émilien Faille had run for the Bloc Populaire in 1944 in Châteauguay, also in Montérégie, and lived until 1978 when he died in Valleyfield, Beauharnois, also in Montérégie. Her father Yvon first ran for the PQ in 1981 in Huntingdon, also in Montérégie.

(The various parallels with Jack Layton, whose grandfather was elected provincially in 1936 as a member of the Union Nationale and ran federally as an independent PC in Mount Royal in 1945, and who has a Cantonese-speaking wife Olivia Chow, cannot have escaped Meili.)

Her father had taken his eldest child to countless political events since she was nine, and she had stayed at her father’s side. An active PQ member since 1992, she had worked in election campaigns in Vaudreuil--Soulanges in 1994, 1997, 1998, and 2003, and was president of the BQ riding association when she took the nomination in 2004, a veteran at age 31.

Her mother ensured Meili, a multi-talented girl, excelled in piano and oil painting. She also excelled in judo, and played hockey (defence).

During university, she had worked as an intern in the office of two PC Ministers. For Pierre Cadieux, the local MP for Vaudreuil (born in Hudson), as Minister of Indian Affairs and Northern Development in 1989. For Pierre Cadieux again, as Solicitor General of Canada in 1990-1991. For Pierre Cadieux again, as Minister of State (Fitness and Amateur Sport and Youth) in 1991-1992. And for Bernard Valcourt, the Minister of Employment and Immigration in 1992-1993; and also for the Secretary General of the Canadian Human Rights Commission in 1993. Her resume even includes being a volunteer for the United Way campaign in the federal public service in 1991-1993.

Both Meili Faille and Michael Fortier deserved to be in the House of Commons. But winner-take all could let only one of them be elected, while leaving 55% of Montérégie's voters unrepresented.

Monday, November 30, 2009

What would Quebec’s National Assembly look like with a proportional voting system?

Below is a simulation based on the votes cast in 2008 and the new electoral map just announced. With 133 MNAs, Quebec would have 58 Liberals, 48 PQ, 23 ADQ, and four Quebec Solidaire.

Quebec has been debating proportional representation since before 1980. The latest step in the debate is the new electoral map announced Nov. 25, 2009.

Why a new map?

Population growth has been concentrated in the Montreal suburbs. Several no-growth regions of Quebec stood to lose seats if the number of MNAs was kept at 125. Without an increase, the ridings of Gaspé, Kamouraska-Témiscouata and Beauce-Nord would have been scrapped.

(This is a familiar issue in Ontario. In 2003, the no-growth North stood to lose up to three of its 11 MPPs. McGuinty’s Liberals promised to let the North keep them, and they kept that promise. But as the Conservative opposition critic Norm Sterling told the House: “Our Constitution has been interpreted by the Supreme Court of Canada to say that you can't give more electoral power to one segment of our population than others. . . if you're going to have 11 ridings in the north, you're probably going to have to have, not 96, but 105, or maybe even 110, in the south.” Ontario needed more MPPs, but left that issue to the Citizens' Assembly which recommended a return to 129 MPPs, inserting that issue into the debate on proportional representation. Quebec may be wiser in tackling the issues separately.)

Quebec’s Bill 78 will protect the four ridings of Bas-St-Laurent, the three of Gaspésie, and the eight in Chaudière-Appalaches, as well as all the other regions, totalling 123 seats. As well, three exceptional ridings are spelled out: Îles-de-la-Madeleine, Ungava and Nunavik. After dividing the population of the 123 normal ridings by 123, any region with a shortfall is given more seats. By my calculation this means seven extra ridings, for a total of 133.

A unanimous consensus

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."

The DGE Report

Quebec’s Chief Electoral Officer (DGE) reported on mixed compensatory models for Quebec in December 2007: “Systèmes Mixtes Avec Compensation (SMAC),” what the rest of Canada calls the Mixed Member Proportional system.

See MMP Made Easy.

Based on his report, my simulation uses the nine regions he uses, with the open list system called “flexible lists,” a 3% threshold, and region-by-region calculation.

In referendums in PEI and Ontario, voters turned down a Mixed-Member system with closed province-wide lists. BC voters recently turned down an STV model. That leaves a Mixed-Member system with regional open lists.

Open regional lists

You have two votes. You vote for your local MNA -- whoever you like best locally, and this vote won't count against your party, for a change -- and you also have a vote for your favourite out of your party's candidates for regional MNA. Your regional vote counts for your party.

A regional MNA, who faced the voters in the region, will represent voters in the region whose votes didn‘t elect a local MNA. Unrepresented and under-represented voters will finally have a voice. And all voters will then have a choice after the election: you can go to your local MNA for service, or to one of your regional MNAs. Instead of having to vote for your party's single candidate, and then having to go to your single MNA, you have competing MNAs! What a concept!

Simulation of 2008 results

In the following simulation, Quebec has nine regions, based on its established 17 regions but grouping some smaller ones. Overall, 61% of the MNAs are still from local ridings, 81 of them, while 52 MNAs are from the nine regions. Montreal Island has 28 MNAs as it does today: 17 local (from larger ridings) and 11 regional. The three regions of Outaouais–Abitibi-Témiscamingue–Nord-du-Québec (West-and-north Quebec) have 11 ridings under the new map, up from the current nine since Nunavik gets a special seat and the Outaouais gets an extra MNA due to growth. With PR this becomes seven local MNAs, four regional.

Take the votes as cast in 2008. (This isn’t real, since many voters in safe ridings don’t bother to vote today, while others have no hope of their vote counting and also stay home. So with a Mixed-Member system more voters would vote, and we’d expect more choices to vote for. But take 2008 as an example.)

Liberals

In Montreal’s northern suburbs of Laurentides--Lanaudière, Liberal voters elected only one MNA last fall, despite casting 30% of the region’s ballots. They would have elected three more. Which three? The ones who got the most votes on the regional ballot (after skipping over anyone who won a local seat.) I’d bet on the three who were the best runners-up: Monique Laurin, director of the Collège Lionel - Groulx, who missed election by only 403 votes. Then Isabelle Lord, Political Assistant responsible for the Laurentians Region to the one Liberal MNA previously elected. Then Johanne Berthiaume, former municipal councillor for the City of Boisbriand.

Under-represented Liberal voters would have elected two more MNAs from the suburban Montérégie south of Montreal. Maybe two rising young Liberal stars: Chambly lawyer Stéphanie Doyon, and Longueuil lawyer Isabelle Mercille who is also Director of Public Affairs for the well-known comedy festival “Just for Laughs.” (Polls show 90% of Canadians want to elect more women, and if a good woman candidate is on the ballot, we’ll elect them.)

In Saguenay–Lac-Saint-Jean–Côte-Nord Liberal voters again elected only one MNA last fall, despite winning 37% of the vote. They would have elected at least one more: the best runner-up was Joan Simard, a former Chicoutimi Councillor.

The government would have fewer MNAs, but a more representative caucus.

Parti Québecois

Two more MNAs from Montreal, where PQ voters were underrepresented. Maybe Frédéric Isaya, a teacher whose father was born in Côte d'Ivoire and whose adoptive father is of Congolese origin, and Martine Banolok, a marketing professional of Nigerian origin.

Two more from Outaouais–Abitibi-Témiscamingue–Nord-du-Québec. Maybe Dr. Gilles Aubé of Gatineau, plus former Val-d'Or MNA Alexis Wawanoloath (first aboriginal elected to the National Assembly), former Rouyn MNA and teacher Johanne Morasse, or Gatineau teacher Thérèse Viel-Déry.

One from the City of Laval. Maybe legal aid lawyer Donato Centomo, real estate agent Rachel Demers or retired police office Marc Demers.

One more from Capitale-Nationale--Mauricie. Maybe Congolese-born Neko Likongo who just finished his LL.M., or lawyer and United Way Campaign Chair Françoise Mercure.

One more from Chaudière-Appalaches–Bas-Saint-Laurent–Gaspésie–Îles-de-la-Madeleine (Est-du-Québec). Maybe Annie Chouinard from Gaspé, who teaches social work techniques and is president of the teachers union at the Cégep de la Gaspésie et des Îles. She was Québec Solidaire’s candidate in 2007 but came back to the PQ in 2008.

Action démocratique du Québec

The 2008 election was a disaster for Mario Dumont’s Action démocratique du Québec, which dropped from 41 MNAs to only seven, losing official party status. However, its voters cast more than 16% of the votes, and deserved to elect 23 MNAs. That would have included:

Three from Laurentides--Lanaudière, where the ADQ was wiped out. Maybe François Desrochers MNA for Mirabel 2007-8, a Vice-Principal and Shadow Minister of Education. Maybe Pierre Gingras, MNA for Blainville 2007-8, former mayor of Blainville and ADQ Caucus Chair. Maybe Linda Lapointe, MNA for Groulx 2007-8.

Three more from Montérégie, where the ADQ was reduced to only one MNA. Maybe Richard Merlini, MNA for Chambly in 2007-8 and vice-president of the ADQ in 2006. Maybe Simon-Pierre Diamond, MNA for Marguerite-D'Youville from 2007-8, the youngest MNA ever elected, a law student who was ADQ Youth President. Maybe Lyne Denechaud; she was political assistant to ADQ MNA André Riedl until he crossed the floor to the Liberals.

Two from the Island of Montreal: maybe Diane Charbonneau, lawyer, vice-president of the ADQ, and three-term president of the Association of Businesspeople of Ahuntsic-Cartierville; and maybe 45-year-old lawyer Pierre Trudelle.

Two from Estrie-Centre-du-Québec. Maybe Jean-François Roux, MNA for Arthabaska 2007-8, and Sébastien Schneeberger, MNA for Drummond 2007-8.

Two more from Capitale-Nationale--Mauricie. Maybe Sébastien Proulx, MNA for Trois-Rivières 2007-8, lawyer and House Leader of the ADQ, and Catherine Morissette, lawyer, MNA for Charlesbourg 2007-8 and Vice-President of the ADQ.

One from the City of Laval: likely Tom Pentefountas, who was then President of the ADQ.

One from Outaouais–Abitibi-Témiscamingue–Nord-du-Québec. Maybe Gilles Taillon, MNA 2007-8 and party President 2006-7.

One from Saguenay–Lac-Saint-Jean–Côte-Nord. Maybe Robert Émond, a union militant in the CSN for 20 years, or Baie-Comeau chiropractor Dr. Louis-Olivier Minville.

One more from Chaudière-Appalaches–Bas-Saint-Laurent–Gaspésie–Îles-de-la-Madeleine (Est-du-Québec). Maybe Christian Lévesque, MNA for Lévis 2007-8, or Claude Morin, MNA for Beauce-Sud 2007-8.

The ADQ would have a broader base and wider perspective.

Québec Solidaire

To add to Dr. Amir Khadir’s solitary win in Mercier, PR would let Québec Solidaire voters elect at least three more MNAs.

One more in Montreal, no doubt co-leader Françoise David.

One from Montérégie. Maybe young Longueuil trade unionist Sebastian Robert who is responsible for QS internal communications at the national level, or Longueuil community development worker Manon Blanchard.

One from Capitale-Nationale--Mauricie. Maybe Serge Roy, president of the Quebec Public Service Union (SFPQ) from 1996 to 2001, or Martine Sanfaçon, community activist.

Power to voters

What’s the point? First, voters everywhere would have real choices, for both candidates and parties.

Second, all voters would have an MLA they trusted. Competing MLAs would be more accountable.

Third, you’d be sure that the system gave fair results. Supporters of all political parties would be fairly represented in proportion to the votes they cast.

Rural and urban voters would be fully represented. All regions would be sure of effective representation.

MNAs would have real control, not be rubber-stamps for a powerful Premier. MNAs and their parties would have to work together, like a real democracy. Bring it on!

Does anyone else use this particular model? Lots of countries use a Mixed-Member system. This particular open-regional-list model is used in the German province of Bavaria, and has been recommended as an improvement to Scotland's similar system.

Quebec-wide proportionality

The Citizens’ Committee in 2006 insisted on Quebec-wide proportionality. (Original French text here.) However, their model was not complete. The DGE’s Report seemed to favour the German model of national calculation and regional allocation, within parties, by the weight of the regional votes within that party’s total, even though this can cause a region to gain or lose “a seat or two” said the DGE’s report.

So I calculated my simulation both ways. With province-wide calculation, some regions lose more than two seats.

My 2008 province-wide projection shows the East losing all three seats that the government just decided to give back to them. Montreal also loses three seats due to lower turnout. Ouest-et-Nord-du-Québec loses the two seats the new map would have given them, and one more. The winners are:

- Capitale-Nationale--Mauricie with three more than even the government wants to give them (they now have 16, the new map gives them 17, and regional allocation gives them 20)
- Montérégie with two more than even the government wants to give them (they now have 21, the new map gives them 23, and regional allocation gives them 25)
- Laurentides--Lanaudière also with two more than even the government wants to give them (they now have 14, the new map gives them 16, and regional allocation gives them 18)
- Saguenay–Lac-Saint-Jean–Côte-Nord with one more than they have now (the new map left them unchanged)
- Estrie-Centre-du-Québec with one more than they have now (the new map left them unchanged)

In return for all this, would Quebec get more proportionality?

Well, Québec Solidaire would get a fifth MNA, in Laurentides--Lanaudière. Maybe Lise Boivin, a teacher at Cégep de St-Jérôme, and former full-time coordinator of the Women’s Committee of the Quebec Teachers Federation.

The model of regional calculation works better than many had feared. The model of Quebec-wide calculation works much worse than I had expected.

Can the DGE’s 9-region model be improved upon? Certainly. The smallest of their nine regions is Laval, which is hardly a remote region. It does not need to be a separate region. Combine it with Laurentides--Lanaudière, and you will have eight regions which work even better. For example, QS got only four seats with the 9-region model, when it deserves five; but with the 8-region model it gets a seat in Laval--Laurentides--Lanaudière.

Sunday, November 15, 2009

Why doesn't New Zealand have open regional lists?

Since the consensus among Ontario electoral reformers now is that closed province-wide lists will not fly here -- as some founding FVC members had warned from the start -- we should examine why New Zealand has closed nation-wide lists for its MMP system. (Open lists require regional lists, since voters need a manageable number to choose from, and most candidates don't have nation-wide reputations.)

This very issue was examined in New Zealand's last review of their MMP model, by Parliament in 2001:

“Survey information shows significant majority support for the principle of open lists, the idea that closed party lists deprive voters of choice has wide currency. Those favouring open lists suggested that MMP would not be fully accepted until voters had the opportunity to exercise some influence over which candidates were to be elected from party lists.

The United party suggested the use of open lists would provide voters with a means to signal to parties how they rated the performance of particular members. The party submitted “there is an understandable adverse reaction [when a member is defeated in an electorate] if an MP defeated in this way returns to Parliament subsequently because of a high place on the list.” The party also submitted list MPs were “effectively beyond public sanction” and, as long as they retained the confidence of their parties, were likely to be re-elected because of their place on the list."

But all parties except the small United Party liked it that way.

The original Royal Commission had considered the issue of open or closed lists at some length. Did subsequent public opinion matter? Apparently not.

"It (the Royal Commission) noted that while the idea of voters having some influence over lists was attractive in principle, there were considerable difficulties in practice with combining open national lists with constituency contests, particularly with dual candidacies. Although supportive in principle of the idea of open regional lists, in the end the Royal Commission recommended that a system of closed national lists be adopted.

The parties that supported the status quo agreed it was good for democracy when political parties had the ability and a strong electoral incentive to present a balanced list.

These parties also saw the ability to control the party lists as an important means to encourage party discipline.

• a national list enables parties to ensure balanced representation among its candidates

• regional lists may lead MPs and electors to concentrate unduly on local or regional issues to the detriment of national issues

• since New Zealand does not have clearly defined regions and is not a federal state, it may be unnecessary and unwise to artificially create such divisions

• with regional lists and each party’s entitlement determined nationally, there is no obvious correlation between list position and the likelihood of election

• in order to make it clear that the party vote is a choice between parties and their leaders, all voters should have the same key names in front of them.

• parties should be able to retain those they regard as talented even if the public did not appreciate these talents to the same degree.

• open lists would undermine the effectiveness and legitimacy of political parties by providing for an outside influence that might rank candidates on a superficial basis."

Good arguments, but hardly democratic ones.

In Scotland surveys also showed significant majority support for the principle of open lists, so their Arbuthnott Commission reviewed the model and recommended a change to the open-list variation of MMP. There is one available for inspection in the German province of Bavaria, with seven regions, that might suit Scotland, and Canada's larger provinces, nicely.

The open list method was also recommended by the Jenkins Commission in the UK. Their colourful explanation accurately predicted why closed lists would be rejected in Canada: additional members locally anchored are “more easily assimilable into the political culture and indeed the Parliamentary system than would be a flock of unattached birds clouding the sky and wheeling under central party directions.”

Does New Zealand have, say, four natural regions? Perhaps not, but Canada's four large provinces certainly have natural regions. As for ensuring balanced representation among parties' candidates, all polls show that most Canadians want more women in parliament, and if parties give us women we can vote for, we'll elect them. Open regional lists will do that. MMP with open regional lists is the Ontario model the Citizens’ Assembly almost chose.

Sunday, October 25, 2009

What would Saskatchewan's legislature look like with a proportional voting system?

Today the ten MLAs from Yorkton-Melfort-Humboldt are all from the Saskatchewan Party. Although 27% of those voters voted NDP, they elected no representatives. Conversely, only three of Regina's 11 MLAs are in the government caucus, although 37% of Regina voters voted SP.

With a regional open-list Mixed Member Proportional (MMP) system such as the Law Commission of Canada recommended (but with smaller regions), if Saskatchewan voters voted as they did in 2007 they would have elected 30 Saskatchewan Party MLAs, 22 New Democrats, and six Liberals.

See MMP Made Easy.

That's using a model with at least one-third of the MLAs elected regionally, in five regions. Three local ridings would generally become two larger ones. You might have 37 local MLAs and 21 elected regionally.

One interesting difference would be the 12 MLAs from Moose Jaw-Swift Current-Estevan-Kindersley: instead of a SP near-sweep, my spreadsheet projects three New Democrats and a Liberal, once NDP votes and Liberal votes count equally with SP voters. That would include the two regional NDP candidates and one regional Liberal candidate who got the most votes across the region. Maybe NDP voters would have elected Glenn Hagel and Sharon Elliott or Ken Crush, and Liberal voters Colleen Christopherson-Cote or Tim Seipp or Michael Klein. The 12 MLAs in that region would be eight local, four regional. The SP would no doubt have won seven of the eight local seats, so those SP voters would even elect one of the regional MLAs.

Another change would be the 10 MLAs I mentioned from Yorkton-Melfort-Humboldt: instead of an SP sweep, we'd see three New Democrats and a Liberal. That would be the three regional NDP candidates who got the most votes across the region (maybe Randy Goulden, Marlys Knezacek and Jordon Hillier) and the top-voted Liberal (perhaps Brent Loehr). The 10 MLAs in that region would be six local, four regional. Those SP voters would no doubt have elected all six local MLAs.

Of course, this projection simplistically assume voters would have cast the same ballots they did in 2007. The reality would be different. When every vote counts, we typically see around 8% higher turnout.

And we would see different candidates. Note that, when the SP members from Moose Jaw-Swift Current-Estevan-Kindersley met in a regional nominating convention, they would have not only voted to put the eight local nominees on the regional ballot, but would have added several regional candidates. With only one or two women from the eight local ridings, when they nominated several additional regional candidates, they would have naturally wanted to nominate a diverse group: more women. And 90% of Canadian voters say that, if parties would nominate more women, they'd vote for them.

Conversely, SP voters across Saskatchewan would also count equally. In the 12 ridings of Regina plus Indian Head - Milestone, instead of four SP MLAs we'd see five, and a Liberal (maybe Michael Huber.) If the SP had won three of the eight larger local ridings, who would Regina voters have chosen as the two regional SP MLAs?

The 13 ridings of Saskatoon plus Martensville were less skewed. Instead of seven NDP and six SP we'd see five NDP, six SP and two Liberals: perhaps David Karwacki and Zeba Ahmad?

The 11 ridings of Prince Albert - Battlefords & North would have an extra NDP MLA (perhaps Maynard Sonntag) and a Liberal (perhaps Ryan Bater).

The exact numbers might be different if Sakatchewan had four regions rather than five. But this is only an exercise in projection: the real results would have been different when more voters turned out to vote in what are now "safe seats."

As noted in previous posts, I prefer regional "top-up" MLAs elected personally under the "open list" model. You would have two votes, and more choice. "Open list" means that voters can vote for whoever they like out of the regional candidates nominated by the party's regional nomination process. The party would win enough regional "top-up" seats to compensate for the disproportional local results we know all too well. Those regional seats would be filled by the party's regional candidates who got the highest vote on the regional ballot. Canadian voters have twice rejected models with closed province-wide lists. The open-regional-list mixed-member model is used in the German province of Bavaria, and was recommended by Canada's Law Commission and by Scotland's Arbuthnott Commission.

Tuesday, October 20, 2009

Imagine that 1.5 million fraudulent votes had been stuffed in Canada's ballot boxes

Larry Gordon wrote:

“Imagine this election scenario. A party wins 155 seats in a 301-seat Parliament and forms a majority government. But after the election, officials discover that 1.5 million fraudulent votes had been stuffed in the ballot boxes, giving the winning party 38 seats it didn't deserve and majority power that it didn't earn.

That would be electoral fraud on a breath-taking scale. Fortunately the scenario is imaginary, but the following one is real.

In 1997 federal election, the Liberals won just 38 per cent of the votes, but the voting system —- not the voters —- gave them 51 per cent of the seats, or 38 more seats than warranted by the popular vote. If Canada had a fair voting system that treated all votes equally, the Liberals would have needed another 1.5 million votes to capture a majority of seats.

The imaginary scenario would be criminal because individuals manipulated results to give an undeserved 1.5 million vote advantage to one party. The real-life election in 1997 also produced an undeserved advantage equal to 1.5 million votes. The only difference is the fantasy fraud was perpetrated by individuals, whereas the culprit in real life is a voting system that distorts what we say with our ballots.”

In the same vein, let’s look at The Bloc Bonus, and other chronic bonuses.

In the 2008 federal election, the Bloc won just 38 per cent of Quebec’s votes, but the voting system gave them 65 per cent of those seats, or 21 more seats than warranted by the popular vote. If Canada had a fair voting system that treated all votes equally, the Bloc would have needed an extra 2.8 million votes to capture 65 percent of Quebec’s seats. (More precisely, an extra 2,843,986 votes to capture 65.333 per cent of those seats.)

In the 2008 federal election, the Liberals won just 46 per cent of the City of Toronto’s votes, but the voting system gave them 91 per cent of those seats, or 10 more seats than warranted by the popular vote. If Canada had a fair voting system that treated all votes equally, the Liberals would have needed an extra 4.6 million votes to capture 91 percent of Toronto’s seats. (More precisely, an extra 4,604,061 votes to capture 90.909 per cent of those seats.)

In the 2008 federal election, the Conservatives won 65 per cent of Alberta’s votes, but the voting system gave them 96 per cent of those seats, or 8 more seats than warranted by the popular vote. If Canada had a fair voting system that treated all votes equally, the Conservatives would have needed an extra 11.3 million votes to capture 96 percent of Alberta’s seats. (More precisely, an extra 11,301,192 votes to capture 96.429 per cent of those seats.)

Wow!

No wonder some Toronto Liberals and some Alberta Conservatives are willing to put up with the Bloc Bonus.

Who cares if the equivalent of 2.8 million fraudulent votes had been stuffed in Quebec ballot boxes, when you’re benefiting from the equivalent of 4.6 million fraudulent votes stuffed in Toronto ballot boxes, or from the equivalent of 11.3 million fraudulent votes stuffed in Alberta ballot boxes.

Proportional representation would be good for Canada.

Tuesday, October 6, 2009

Democratic nominations: why is Germany more democratic than Canada?

You can't turn on the television without hearing of a candidate being appointed to run for parliament somewhere in Canada.

In Germany, this would be illegal.

But isn't Germany the place, you may ask, where half the MPs are elected on a party list? Aren't they appointed? And anyway, can't parties do whatever they like?

No, and no. Germany has laws to guarantee democratic nominations.

Why can't Canada have laws making nominations democratic?

Germany's Law on Political Parties states "The nomination of candidates for elections to parliaments must be by secret ballot:" Sec. 17. Their Federal Elections Act states "A person may only be named as a candidate of a party in a constituency nomination if he or she has been elected for this purpose at a members' assembly convened to elect a constituency candidate:" Sec. 21(1).

The Law Commission of Canada designed a democratic voting system for Canada. The nomination system was outside the mandate of that study, but their model was the German system.

What about nominations at provincial conventions for candidates to be on province-wide lists of party candidates for the federal parliament? Doesn't the party leader decide which candidates get the top ranking, almost guaranteeing them a seat?

No. Again, Germany's Federal Elections Act requires that the order of names of the candidates in the provincial list must be laid down by secret ballot: Sec. 27(5). (This matters for federal elections in Germany, where half of the MPs are elected from closed party lists. By contrast, in provincial elections in Bavaria the list order doesn't matter, since voters vote for the candidate on the regional list they prefer, as well as for the local candidate they prefer.)

But aren't those provincial conventions controlled by the party brass?

No. Even the election of convention delegates is democratic: "The elections of the delegates to delegates' assemblies (party conventions) shall be secret" says Sec. 15(2) of the Law on Political Parties. And if the party allows executive members to be automatic ex-officio delegates at conventions, that Law states that the number of them eligible to vote must not exceed 20% of the total number of delegates: Sec. 9(2). And the usual practice is that the provincial convention to elect list candidates is held only after local constituency candidates have been elected; most good list positions go to candidates who have already won a local nomination. (The SPD, for example, makes sure at least 40% of each group (5 or 10) of candidates on the list are women. Occasionally the list includes a "list-only" minority or female candidate not nominated locally.)

But if an incumbent MP loses the nomination, can't the party brass protect him or her? Not much. The provincial executive may object to the decision of a membership meeting. "If such an objection is raised, the ballot shall be repeated. Its result shall be final:" Sec. 21(4) of the Federal Elections Act. An interesting example from the recent German election was the nomination of Bärbel Bas for the SPD in Duisburg I, defeating an incumbent MP for the nomination. At the first nomination meeting she won by only five votes. A second meeting was called: she increased her margin to 17 votes, and then won the seat in the election.

A notorious instance of breach of these provisions arose in Hamburg in 1993, when the Hamburg constitutional court ruled that the CDU had not abided by the provisions of the law in its selection procedures for the Land election of 1991, and a new election had to be held.
http://ecpr.eu/Filestore/PaperProposal/7aea87aa-da0e-44c0-aea3-eb40a5b3132f.pdf

Saturday, September 12, 2009

The Bloc Bonus, and other chronic bonuses

In 2008 it took 86,203 federalist voters to elect one Quebec MP, but only 28,163 Bloc voters. Bloc Québecois voters cast 38.1% of the votes in Québec, so they deserved 28 of the 74 MPs won by parties. But they got 49, a bonus of 75%.

Back in 1993 the Bloc Québecois formed the Official Opposition despite getting fewer votes than either Reform or the Progressive Conservatives.

In 1993 Bloc voters cast 49.3% of the votes in Québec, so they deserved to elect 36 MPs of Québec’s 74 seats won by parties. But they elected 54, a bonus of 50%.

And they did it again, and again, and again, and again, and again.

Over six elections they got an average bonus of 53%. (Details below)

Why do conservatives accept this?

Why do some Conservatives accept such an undemocratic voting system? A system that cheats Québec federalist voters? A system that also cheats Toronto Conservative voters, who deserved to elect six MPs in the last election but have elected no one since 1993?

Is it because their party has been run from Alberta?

The Alberta conservative bonus

In 2008 it took 449,013 non-Conservative voters to elect one Alberta MP, but only 30,450 Conservative voters. Conservative Party voters cast 64.7% of the votes in Alberta, and deserved to elect 18 of Alberta’s 28 MPs. But they elected 27, a bonus of 50%.

In 1993 Reform Party voters cast 52.3% of the votes in Alberta, so they deserved to elect 14 of Alberta’s 26 MPs. But they elected 22, a bonus of 57%.

In fact, over six elections Alberta conservative voters got a bonus of an average of 57%, even a bit worse than the Bloc’s 53% bonus in Quebec. (Details below.) Meanwhile, in six elections Toronto's Conservative voters have elected no one: 244,732 of them in 2008.

Why do Liberals accept this?

So why do some Liberals accept such an undemocratic voting system? A system that cheats Québec federalist voters? A system that also cheats Liberal voters in the West and in Ontario outside the GTA?

Is it because the party has been run from Toronto?

The Toronto Liberal bonus.

In 2008 it took only 21,887 Toronto Liberal voters to elect an MP, but it took 252,090 non-Liberal voters to elect one MP. Toronto Liberals keep getting a big bonus of their own, again for the last six elections in a row. (See this post.)

These numbers assume voters voted as they did in 2008. In fact, if voters knew every vote would count, more would have voted, and some would have voted differently. We would have had different candidates - more women, and more diversity of all kinds.

Are Canadians slow learners?

In a country of solitudes, where parties are comfortably entrenched in their strongholds, will nothing change?

When will Conservative activists outside their Party’s strongholds, and Liberal activists outside their Party’s strongholds, be more vocal? They must be thinking “what are we, chopped liver? These regional bonuses are bad for Canada. And the Bloc’s bonus keeps paralyzing Parliament.” When will they say it in public?

Imagine that 1.5 million fraudulent votes had been stuffed in Canada's ballot boxes.

Details of the Bloc Bonuses:

In 1997 Bloc voters cast 37.9% of the votes in Québec, so they deserved 28 of Québec’s 75 MPs. But they got 44, a bonus of 57%.

In 2000 they cast 39.9% of the votes in Québec, so they deserved 30 MPs. But they got 38, a bonus of 27%.

In 2004 they cast 48.9% of the votes in Québec, so they deserved 37 MPs. But they got 54, a bonus of 46%.

In 2006 they cast 42.1% of the votes in Québec, so they deserved 31 of the 74 MPs won by parties. But they got 51, a bonus of 65%.

Details of the Alberta bonuses:

In 1997 Reform Party voters cast 54.6% of the votes in Alberta, and again they deserved to elect 14 of Alberta’s 26 MPs. But they elected 24, a bonus of 71%.

In 2000 Canadian Alliance voters cast 58.9% of the votes in Alberta, and deserved to elect 15 of Alberta’s 26 MPs. But they elected 23, a bonus of 53%.

In 2004 Conservative Party voters cast 61.7% of the votes in Alberta, and deserved to elect 17 of Alberta’s 28 MPs. But they elected 26, a bonus of 53%.

In 2006 Conservative Party voters cast 65.0% of the votes in Alberta, and deserved to elect 18 of Alberta’s 28 MPs. But they elected all 28, a bonus of 56%.

Open list

As noted in previous posts, I prefer a mixed member proportional system with regional "top-up" MPs elected personally under the "open list" model. You would have two votes, and more choice. "Open list" means that voters can vote for whoever they like out of the regional candidates nominated by the party's regional nomination process. The party would win enough regional "top-up" seats to compensate for the disproportional local results that cause these chronic bonuses. Those regional seats would be filled by the party's regional candidates who got the highest vote on the regional ballot. Still, 65% of MPs would be elected from local ridings as we do today. Each province and region would keep the same number of MPs it has today. This is the model recommended by the Law Commission of Canada, used in the German province of Bavaria, and recommended for Scotland by its Arbuthnott Commission.

Sunday, September 6, 2009

What would Manitoba's legislature look like with a proportional voting system?

With a regional open-list MMP system such as the Law Commission of Canada recommended (but with smaller regions), and using the Scottish Parliament's "highest average" calculation method, if Manitoba voters voted as they did in 2007 my spreadsheet projects an NDP majority of three: 30 NDP, 23 PC, four Liberal.

See MMP Made Easy.

That's using a model with at least one-third of the MLAs elected regionally, in five regions. In most cases three local ridings would become two larger ones. You might have 36 local MLAs and 21 elected regionally.

One interesting difference would be in South West and Central Manitoba: instead of the sole Brandon New Democrat MLA, I project four. That would be the three regional NDP candidates who got the most votes across the region. Maybe Denise Harder from Ste. Rose, James Kostuchuk from Portage La Prairie, and Harvey Paterson from Minnedosa? Instead of the PC near-sweep of the region, when NDP votes count equally the PCs get six seats, not nine. The ten MLAs in that region would be six local, four regional. The PCs would no doubt have won five of the six local seats, so they even get one of the regional MLAs.

Of course, this projection simplistically assume voters would have cast the same ballots they did in 2007. The reality would be different. When every vote counts, we typically see around 8% higher turnout. And you would see different candidates. When any party's regional nomination process nominates five regional candidates at once, you can expect them to nominate a diverse slate.

Conversely, Conservative votes across Manitoba would also count equally. In the 18 ridings of north Winnipeg, instead of one lonely PC we'd see five, and two Liberals instead of only one. Maybe Linda West, Chris Kozier, Kelly de Groot and Brent Olynyk would be PC regional MLAs, and Wayne Helgason a Liberal regional MLA?

The 13 ridings of south Winnipeg were less skewed. Instead of three PCs we'd see four; perhaps Jack Reimer? Instead of only one Liberal, we'd see two: perhaps Paul Hesse?

The six Northern ridings would have a couple of regional Conservative MLAs along with four local New Democrats. That might be Maxine Plesiuk and David Harper?

The 10 southeast ridings around Winnipeg actually, by a fluke, saw a fair result over all. It would still be four New Democrats and six PCs.

The exact numbers might be different if Manitoba had four regions rather than five. And they would certainly be different if Manitoba used the old German "highest remainder" calculation (which Germany has just moved away from): maybe 28 NDP, 22 PC and seven Liberal. I'm assuming the Manitoba government would prefer the Scottish model.

As noted in previous posts, I prefer regional "top-up" MPs elected personally under the "open list" model. You would have two votes, and more choice. "Open list" means that voters can vote for whoever they like out of the regional candidates nominated by the party's regional nomination process. The party would win enough regional "top-up" seats to compensate for the disproportional local results we know all too well. Those regional seats would be filled by the party's regional candidates who got the highest vote on the regional ballot. Canadian voters have twice rejected models with closed province-wide lists. The open-regional-list model is used in the German province of Bavaria, and was recommended by Canada's Law Commission and by Scotland's Arbuthnott Commission.

Friday, July 17, 2009

Why didn’t more Liberals speak up in 2008?

As noted in a previous post, with a democratic voting system, a proportional House of Commons elected as voters voted in 2008 would have given Liberal voters 26 more MPs from regions where they were unrepresented or under-represented.

That‘s nine more from the West, ten more from Ontario outside the GTA, and seven more from Quebec outside Montreal. Liberals need the recommendation of the Law Commission of Canada.

Chronic Liberal under-representation

Was this a one-time problem? No, a chronic one.

In 2006, with a democratic voting system Liberal voters would have elected 18 more MPs from regions where they were unrepresented or under-represented. That‘s eight more from the West, five more from southern Ontario outside the GTA, and five more from Quebec outside Montreal.

Chronic federalist under-representation

In 2008, 49 of Quebec’s MPs were Bloc members, and only 26 were federalists (14 Liberals, 10 Conservatives, 1 NDP, and 1 independent). With a democratic voting system in 2008 Quebec voters would have elected 17 more federalists. That‘s 43 federalists (17 Liberals, 16 Conservatives, 8 NDP, 1 Green, 1 independent) and only 32 Bloc MPs.

Was this a one-time problem? Again, not at all.

In 2006, Quebec federalist voters would have elected 17 more MPs: seven Conservatives, six New Democrats, three Greens and one Liberal. And that's chronic. See the Bloc Bonus and other chronic bonuses.

So why don’t more Liberals speak up?

Many Liberal activists in the West like Anne McLellan know all this very well; they've been dealing with it since 1972. So do Liberal activists in Quebec.

So why don’t more Liberal activists promote electoral reform?

Because in 2006 Liberal voters would have elected seven fewer MPs from Toronto and five fewer from Peel/York. Just as, in 2008, Liberal voters would have elected eight fewer MPs from the City of Toronto and four fewer from Peel/York.

Now, the best Toronto Liberal reformers have a national vision.

However, others think 12 fewer Liberal MPs from the GTA are more important than 17 more federalist MPs from Quebec. They think 12 fewer Liberal MPs from the GTA are more important than 18 or 26 more Liberal MPs from regions like Alberta where Liberal voters were unrepresented or under-represented.

Do these Toronto-centred folks really run the Liberal Party?

Perhaps not. When John Gerretsen was elected MPP for Kingston in 1995, he found himself the only Liberal elected between Toronto and Ottawa, facing a very conservative majority government elected by a minority of voters. A familiar position for Ontario Liberals, who had faced fake-majority governments for 42 of the previous 53 years. Proportional representation was in Ontario Liberals' interest, and Gerretsen started working for it. Unfortunately, by 2005 some of them had started to forget this. It is also in Quebec Liberals' interest, where the skewed demographics give the PQ the same bonus the Bloc gets; that's how the PQ won the 1998 election with fewer votes than the Liberals.

Open list

As noted in previous posts, I prefer regional "top-up" MPs elected personally under the "open list" model recommended by the Law Commission of Canada. You would have two votes, and more choice. "Open list" means that voters can vote for whoever they like out of the regional candidates nominated by the party's regional nomination process. Like this ballot, designed by Elections PEI, which PEI voters chose as the winning system in their plebiscite in November 2016. The party would win enough regional "top-up" seats to compensate for the disproportional local results we know all too well. Those regional seats would be filled by the party's regional candidates who got the highest vote on the regional ballot. Each province would keep the same number of MPs it has today.

Monday, May 18, 2009

What would a proportional representation model for Nova Scotia look like?

Nova Scotia’s Premier Rodney MacDonald stated in the May 18 Halifax Chronicle-Herald:

“One of the concerns I have . . . is the issue of proportional representation. As a rural Nova Scotian, that scares me . . . because that means less of a voice for rural parts of Nova Scotia.”

That’s why, when the Ontario NDP took a position on proportional representation in 2002, they decided on a regional model. Perhaps Nova Scotians have not yet looked at how a decent proportional system would work in Nova Scotia.

They're not alone. Outside BC, journalists routinely say things like, with winner-take-all, "you get representatives who fight tooth and nail for the good of their communities. I'm not sure how you get a system of proportional representation that doesn't somehow require the appointment of MPs or MPPs."

So let’s look at a typical regional proportional system.

In referendums in PEI and Ontario, voters turned down a Mixed-Member system with closed province-wide lists. BC voters recently turned down an STV model. That leaves a Mixed-Member system with regional open lists.

You have two votes. You vote for your local MLA -- whoever you like best locally, and this vote won't count against your party, for a change -- and you also have a vote for your favourite out of your party's candidates for regional MLA. Your regional vote counts for your party. Like this ballot.

See MMP Made Easy.

A regional MLA, who faced the voters in the region, will represent voters in the region whose votes didn‘t elect a local MLA. Unrepresented and under-represented voters will finally have a voice. And all voters will then have a choice after the election: you can go to your local MLA for service, or to one of your regional MLAs. Instead of having to vote for your party's single candidate, and then having to go to your single MLA, you have competing MLAs! What a concept!

The province has four regions. Urban Halifax has 18 MLAs as it does today: 11 local (from larger ridings) and seven regional. Cape Breton still has nine: six local, three regional. South-West still has 14: nine local, five regional. North-East still has 11: seven local, four regional. Overall, 63% of the MLAs would still be from local ridings, 33 of them, while 19 MLAs would be from the four regions.

Mr. MacDonald would like what this does in Halifax.

Take the votes as cast in 2006. (This isn’t real, since many voters in safe ridings don’t bother to vote today, while others have no hope of their vote counting and also stay home. So with a Mixed-Member system more voters would vote, and we’d expect more choices to vote for. But take 2006 as an example.)

The NDP swept Halifax in 2006. With fewer local seats, NDP voters would have elected five fewer Halifax MLAs. But PC voters would have elected three more MLAs than they did. Which three? The ones who got the most votes on the regional ballot (after skipping over anyone who won a local seat.) I’d bet on Bill Black (hmm -- might he have been Premier today?), African-Canadian educator Dwayne Provo (would Nova Scotia have a black cabinet Minister today?), and former Caucus Chair Gary Hines.

Halifax Liberal voters would have elected one more MLA than they did. Maybe their leader Francis MacKenzie wouldn't have lost his seat?

Halifax Green voters would have elected one MLA. Maybe their leader Nick Wright? Then again, Amanda Myers got more votes than he did.

But in the South-West, where NDP voters were short-changed, they’d have elected two more MLAs than they did, and PC voters two less. Maybe Wolfville Councillor David Mangle and Mahone Bay councillor Chris Heide would have been elected NDP regional MLAs?

In the North East, almost 16,000 Liberal voters elected no one in 2006, when they deserved two MLAs. I’d bet Antigonish lawyer Daniel MacIsaac would have won a regional seat, and maybe Danny Walsh from Pictou County.

Sometimes our winner-take-all system happens, by accident, to work about right. In 2006 it did in Cape Breton. PC voters elected four MLAs, Liberal voters three, and NDP voters two, and those numbers wouldn’t change.

In fact, the overall result wouldn’t be so very different from 2006, which accidentally worked out about right. You’d still see a PC minority government, with one less MLA, but better Halifax representation. Three fewer NDP MLAs, three more Liberals, one Green.

What’s the point? First, voters everywhere would have real choices, for both candidates and parties.

Second, all voters would have an MLA they trusted. Competing MLAs would be more accountable. Scotland's similar model has 16-member regions (nine local members, seven regional), while Wales has 12-member regions (eight local, four regional), allowing reasonable accountability.

Third, you’d be sure that the system gave fair results. Not like 1999, when PC voters elected 58% of the MLAs with only 39% of the vote. Supporters of all political parties would be fairly represented in proportion to the votes they cast.

Rural and urban voters would be fully represented. All regions would be sure of effective representation.

Would this system always mean minority governments? Well, in 1993 Liberal voters outnumbered the PC and NDP voters combined, so they got an honest majority, and that wouldn‘t change. Same as the PC win in 1984.

But otherwise, MLAs would have real control, not be rubber-stamps for a powerful Premier. MLAs and their parties would have to work together, like a real democracy. Bring it on!

Does anyone else use this particular model? Lots of countries use a Mixed-Member system. This particular open-regional-list model is used in the German province of Bavaria, and has been recommended as an improvement to Scotland's similar system. It was recommended for Canada by the Law Commission of Canada.