Monday, December 15, 2008

What would a proportional Alberta Legislature look like?

In the last provincial election, the majority of Alberta voters stayed home. And in the 2008 federal election, while BC had a 60% turnout. Alberta had only 54%. Why? Because, outside of two or three ridings in Edmonton, the Conservatives had safe seats. Why bother voting?

Many Alberta provincial Liberals have favoured PR for some time. The party's website used to say "An Alberta Liberal government would organize a Citizen’s Assembly on Electoral Reform, to determine if other voting systems — including proportional representation — could improve participation and representation in our democracy."

So what would the Alberta Legislature look like under a PR model?

Note that, if voters knew every vote would count, more would have voted, and some would have voted differently. Alberta would have had different candidates - likely more women, and more diversity of all kinds. However, all I can do is project the votes cast in 2008 into a reformed voting system.

I used a Mixed Member Proportional model with regional open lists. You still have 83 MLAs. The 50 local MLAs are elected from districts larger than today's (about five of today's districts become three larger districts.) The 33 regional top-up MLAs are elected from five regions.

You, the voter, have two votes: one for your local MLA, the other for the party you want in government and for your favourite regional MLA candidate of your party (like the right-hand part of this ballot.) So you are free to vote for the best candidate locally; only your regional ballot counts for your party. The regional MLAs top up the local results, so the total result matches the vote shares in the region. Every vote counts equally. And the regional seats are filled by the party's regional candidates who get the most votes on the regional ballot (unless that person was already elected to a local seat.)

See MMP Made Easy.

Overall, PC voters would still have elected a majority: 44 of the 83 MLAs. Liberal voters would have elected 22 MLAs, NDP voters seven, Wildrose Alliance voters six, and Green Party voters four.

Liberal voters would have elected MLAs across Alberta, not just in Calgary, Edmonton and Lethbridge. In the 12 ridings of Northern Alberta, Liberal voters would have elected two regional MLAs: maybe Lisa Higgerty from Hinton and Ross Jacobs from Fort McMurray. In the 13 ridings of Central Alberta, another two MLAs such as Diane Kubanek and Richard Farrand of Red Deer. In the 13 ridings of Southern Alberta, two more regional MLAs such as Karen Charlton of Medicine Hat and Bal Boora of Lethbridge. In the 23 Calgary ridings, three more MLAs such as Craig Cheffins, Mike Robinson and Avalon Roberts. In the 22 Edmonton ridings, four more MLAs such as Rick Miller, Mo Elsalhy, Bruce Miller and Weslyn Mather.

New Democrat voters would have elected a regional MLA from Northern Alberta such as Adele Boucher Rymhs from Peace River. In Central Alberta, a regional MLA such as Lisa Erickson from Leduc County. In Calgary, a regional MLA such as Julie Hrdlicka. In Edmonton, two more MLAs such as David Eggen and Ray Martin.

Wildrose Alliance voters would have elected two regional MLAs from Southern Alberta, no doubt leader Paul Hinman, plus another such as Kevin Kinahan from Coaldale. From Calgary, two regional MLAs such as Chris Jukes and Bob Babcock. From Central Alberta, a regional MLA such as Dean Schmale from Winfield. From Northern Alberta, a regional MLA such as Dale Lueken from Fairview.

Green Party voters would have elected a regional MLA from Calgary such as leader George Read or Susan Stratton, one from Edmonton such as Glen Argan or Kate Harrington, one from Central Alberta such as Joe Anglin from Rimbey, and one from Southern Alberta such as Dan Cunin from Cochrane.

All MLAs would have faced the voters, and all votes would have counted. Democracy, eh?

Voters for all parties would be represented in all regions, except where they had too few voters to elect even one regional MLA: NDP voters in southern Alberta, Wildrose Alliance voters in Edmonton, and Green voters in Northern Alberta.

Of course, that was with a miserable turnout of only 40.6 percent. Fair voting systems usually boost turnout by an average of about six percent. Likely more in Alberta, where turnout in 1993 was more than 60 percent. Funny thing: if there's no point voting, lots of people don't. Or as Elections Alberta stated "In an election where there appears to be a clear front runner, electors may be less motivated to vote since the outcome is perceived to be predetermined and their vote may not be needed or may not make a difference." So add a whole lot more votes to the picture, and who knows what would have happened?

This projection is based on five regions: Calgary with 23 MLAs (14 local, 9 regional); Edmonton with 22 MLAs (13 local, 9 regional); Central Alberta with 13 MLAs (8 local, 5 regional); Southern Alberta with 13 MLAs (8 local, 5 regional); and Northern Alberta with 12 MLAs (7 local, 5 regional).

The same kind of model works well federally too. Alberta's provincial Liberal party is quite separate from their federal cousins, and more popular, so federal Liberals need proportional representation even more than provincial Liberals.

This model is based on the recommendation of the Law Commission of Canada. An STV model like BC-STV would likely have had much the same result.

2 comments:

Jennifer Jilks said...

Great to see another blog to add to my 'following' list. I love reading them and I'm glad to see another blogger over the age of 20! :-)

I am afraid, Wilf, that you're going to have to go back to basics for me! I will happily follow your blog, I love discourse, it helps us move towards truth, but I need clarification.

When you say that one-third of MPs to be regional "top-up" MPs elected personally do you mean that I cannot elect those topped up? They are appointed by the party?

I cannot accept that this is democracy if I do not vote directly for the 'top-ups'. We are trying to get away from an appointed Senate, and I cannot see that this is any different.
Can you help enlighten me?!

Wilf Day said...

By "elected personally" I mean just that. On the regional ballot, you vote for the person. The regional candidate with the most votes from the regional ballot wins the party's first regional seat (unless that person already won a local seat, in which case the party's regional candidate with the second highest votes is elected to that regional seat.)

This is the "open list" MMP model used in the German province of Bavaria. It is also the MMP model designed by the British Columbia Citizens' Assembly, which designed two models, and then made a final choice for the Irish STV system instead of open-list MMP.

There are other MMP models. Scotland uses a regional closed list model. So do a couple of German provinces. New Zealand uses a national closed list model. So do the majority of German provinces. That means that the party's provincial convention nominates the list of provincial candidates for the provincial "top-up" seats, and the voters do not choose between them. In that model, the voters choose the local MPP from the candidates nominated locally, and their second vote is simply for the party they want to see in government.

The Ontario Citizens' Assembly chose province-wide lists rather than regional lists. This is because they were trying to give priority to keeping local ridings as small as possible, not make the legislature too large.

The ideal model, from the point of view of getting proportional results, would have been to keep the present 107 ridings and add another 71 "top-up" MPPs, for a legislature of 178 MPPs of which 40% would be "top-up" MPPs, which is a good ratio to get proportional results. But no one was bold enough to propose 178 MPPs.

So those 103 Citizens were looking at expanding the House to 129 or 139 MPPs, and cutting the number of local ridings to 97 or 103, so that the percent of "top-up" MPPs would be 25% (or maybe 30%). But 25% is quite often not enough, especially if the province is broken down into seven or eight regions, some of which will have more disproportional results than others. So for that reason they decided on province-wide lists. It would be silly to expect voters to choose between at least 32 candidates, most of whom they had never heard of, on a province-wide ballot, so they chose the closed-list model.

But then they finally refined the numbers to 90 local ridings, 39 "top-up" MPPs, 129 total MPPs, 30% "top-up." At that point they should have gone back to review their previous decision: they could have chosen to use seven regions, with only five or so regional MPPs each, and then they could have used the Bavarian open-list model. But they had no time to review their previous choices. Their schedule was too tight, partly because they started a year later than the government's Democratic Renewal Secretariat had wanted.

That's why critics complained last fall about the parties "appointing" MPPs, like Senators. Actually that was unfair; they would have been nominated well before the election, and all parties agreed they would do so democratically. But province-wide lists didn't sound very accountable.

The main reason MMP lost in the referendum was poor public knowledge. A survey soon afterwards showed that, among voters who understood the proposal, the majority voted for it. Still, the closed province-wide lists were not a great feature, I agree. So I'm not suggesting that model.