Monday, December 1, 2008

A readily available proportional representation model for Canada.

In the last election more than half the votes cast elected no one. This is not fair. Our voting system gives regional grievance parties a bonus, but no bonus for compromising. Many members of all parties and none want a fairer voting system.

Do we have a readily available model for a proportional representation system? Yes. (I have modified the model described below similar to the model presented by Prof. Henry Milner Feb. 21, 2009.)

The starting place would be the Law Commission of Canada Report tabled in March 2004. It proposed a mixed member proportional (MMP) model with two-third of MPs elected locally as today, and one-third of MPs to be 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. Like the right-hand part of this ballot. 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.

What have we learned since then?

The Law Commission was right to propose an open-list model. All MPs would face the voters. The voters have twice rejected MMP models with closed province-wide lists. Ontario voters would have voted for even that model if they had understood it, polls have shown, but that is not easy to explain, and anyway electoral reformers have concluded the Law Commission’s model would have been more acceptable on that point: the regional MPs would be more accountable. Voters are concerned with effective local representation, having an MP that they voted for and not having to vote strategically. With open regional lists, every voter will have a local MP, and if they did not vote for him or her, at least they will mostly have a regional MP they voted for.

This open-regional-list model was recommended by Scotland's Arbuthnott Commission in 2006 as an improvement to their regional model. A similar model is used in the German province of Bavaria.

In Germany and Scotland they nominate local candidates first, then rank them on regional lists as well. With open-list in smaller accountable regions the rank doesn't matter as much.

The Report says it is inspired by the systems currently used in Scotland and Wales, which have 16-MP regions (9 local MPs, 7 regional MPs) or 12-MP regions (8 local MPs, 4 regional MPs). With 2/3 local MPs, a 14-MP region would have 9 local MPs and 5 regional MPs. This would mean seven regions in Ontario, five or six in Quebec, two in BC, and two in Alberta. They also show a sample calculation with larger regions, because they provide more proportional results than smaller districts, and make it easier for smaller parties to win representation in Parliament, but this is not their recommendation.

First, Northern Ontario showed last year that they wanted Northern voters to elect Northern MPPs, and a guarantee that the North would not lose MPPs. The North must be a separate region. Jack Murray’s Ontario NDP PR Task Force reached the same conclusion in 2002.

Second, regions of 36 or so MPs would mean an effective threshold of only 3%. In Ontario’s referendum last year, the 3% threshold was widely criticized. Smaller regions with an effective threshold of 5% (or even higher in some cases) would be less controversial.

Third, ballots with long lists of candidates for as many as 12 regional MPs would be easy targets for criticism. Smaller regions would have more accountable MPs. As Lord Jenkins said: regional "members locally anchored . . . are, we believe, more easily assimilable into the British political culture and indeed the Parliamentary system than would be a flock of unattached birds clouding the sky and wheeling under central party directions."

So we need medium-sized regions.

The preferred model for Quebec has also been refined since 2004. Last December the Quebec Director-General of Elections tabled the latest report, which suggests a model with 127 MNAs in nine regions, an average of 14 each. Typically a region would have five regional MNAs, a workable number, and nine local MNAs, with an effective threshold averaging 7.1%. In the two most urban and largest regions it would be lower, giving new parties a chance for a foothold. When parties nominate five or more regional candidates, they will naturally tend to nominate a balanced group including women and minorities. And 85% of voters support electing more women, while only 11% oppose this.

In the model described here, in most cases, three present ridings become two larger ridings. Easily organized with an expedited Boundaries Commission process. Local ridings are usually 50% bigger than today. Regions have at least one-third regional MPs; a typical region would have 14 MPs, 9 local, 5 regional.

Yet would such a model, with separate smaller regions and only 35% regional MPs (compared with more than 40% in other models), produce fair results?

Yes, fair enough. A spreadsheet of the 2008 results shows that this model would (if voters voted the same) give quite proportional results. (See below.)

Does the model have distracting math problems? Not at all. If a party gets 38% of the vote in a 15-seat region, that's 5.7 seats, round up to 6. If it won 5 local seats already, that's one regional "top-up" MP. The only possible tricky math would be this: suppose one party gets 7.4 seats, one gets 6.3, and one gets 1.3, if you round them all down, that's only 14 seats; who gets the 15th? The one with 7.4: the "largest remainder" rule.

One of Jack Layton’s objectives in the 2008 election campaign was, as he put it, "the very early implementation" of proportional representation. This model would fit the bill.

What would a proportional House of Commons look like?

As compared with a perfectly proportional model, this model gave the Bloc three extra seats as a result of the low 35% ratio, at the cost of the Greens (two) and the Liberals (one). In the rest of Canada the 35% ratio costs the NDP two seats and the Greens two more, to the benefit of the Liberals (three, outweighing their Quebec loss) and the Conservatives (one). The result Canada-wide would have been 119 Conservatives, 83 Liberals, 56 NDP, 31 Bloc, 17 Greens, and two Independents: a Liberal-NDP government could rely on either the Greens or the Bloc for a majority. (In Germany they don't use the term "minority government" in that situation, they call it "governing with shifting majorities," a rather more accurate term. But in the situation where the minority government has only one partner, that's a "minority government with external support.")

This model gives the Greens one MP from Quebec, where they got only 3.5%. If there was a legal threshold of 5% in each province, the Greens would have no MP from Quebec.

Note that these numbers assume voters voted as they did on October 14. 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.

No comments: