7September
A Vote for MMP campaign has set up an office in Toronto and claims a dozen chapters in the province; its opponents have managed to raise just $500 (€350) to print a leaflet.
Grrliz has written about her opposition to a mixed-member proportional voting system. Mat GB has responded to her points. Some weeks late and a couple of euro short, we're going to stick our oar in, discussing the psephology alone. Quotes are from Grrliz unless specified otherwise.
1. Your vote always counts. Always. They can't not count your vote just because it's for the person who did not win. I know this is shocking, but not everyone can win.
2. You can never waste your vote unless you choose to purposely spoil it in an act of protest, in which case it's your fault entirely that your vote has been wasted.
True, but such are the problems of the pro-MMP camp using unclear and emotive terms such as wasted vote. It's easier to discuss votes that make a difference, as in the following scenarios:
* A vote for the winning candidate, up to the point where the winner has received more votes than the runner-up. (So if Red wins with 30,000 votes, Blue has 25,000, and Yellow has 10,000, then 25,001 Red votes made a difference.)
* All votes for the candidate in second place, as the winner had to secure more. (In the example above, all 25,000 Blue votes made a difference.)
* In elections where this applies, votes sufficient for a candidate to retain their deposit or qualify for relief on their expenses. In the UK, candidates securing less than 5% of the total poll lose a £500 deposit; in Canada, it appears that 10% of the vote secures reimbursement of some expenses. (In the three-party example, under UK rules, the first 3250 votes for each candidate all made a difference.)
* It can be argued that other competitions can make a difference. Often, a close battle for second and third can determine the clear opposition party at the next election. Sometimes, lower places count - we're reminded of the sniggering against the Conservatives when they were beaten into fourth place at Hartlepool by the UIP, or the time the SDP was beaten by the Monster Raving Loony Party into seventh place at Bootle and promptly gave up.
In the example here, the votes that affect the outcome of something are:
Red: First 3250, then 21,751 of the remaining 26,750.
Blue: All 25,000
Yellow: First 3250.
Some votes make more of a difference than others, and it's common - pace Mr. GB - to ignore the deposit and minor placing conditions, and concentrate only on victory.
3. Do not confuse "my candidate didn't win" with "the system is unfair". My candidate still got votes, they simply did not get enough votes to win our riding. Although I did not vote for the winner, proportionally more people in our riding did and thus the winner best represents the pervading popular opinion of the people in my riding. I may not agree with their choice but, well, you can't always get what you want.
In his response, Mr. GB mentioned Duverger's Law. It's always worth explaining terms:
(1) a majority vote on one ballot is conducive to a two-party system; (2) proportional representation is conducive to a multiparty system; (3) a majority vote on two ballots is conducive to a multiparty system, inclined toward forming coalitions.
First-past-the-post (FPTP) is the largest single vote on one ballot. In any given riding, this will tend to form two large minorities and a small group floating between the two.
Mixed member proportional (MMP) is, as the name suggests, a proportional representation system, allowing many parties to prosper. In Scotland, where we've recently had the third MMP election, there's always been representation for the Green party - they could not possibly win a single riding, but attract enough party preferences to consistently win area seats.
For completeness, the two-ballot majority system is the one used in French elections - all candidates appear on the first ballot, then assuming no-one won half the vote, the top two progress to a run-off a week or two later.
Now, under FPTP, the candidate of the larger minority gets to sit in parliament. The smaller minority and any fringe opinions are completely unrepresented. Under MMP, the candidate of the larger minority still gets their place, but is joined by representatives of the smaller minority, and of any fringe opinions attracting significant support. Proponents of MMP reckon that this system allows for a plurality of views, and almost everyone can say that they elected one or more members.
4. Under the new proposed system, the number of ridings will go from 107 to 90. The remaining ninety ridings are now larger than ever, both geographically and by the number of constituents they represent. Your vote will count less since a greater population voting in a riding dilutes the relative value of your own vote.
5. With a new total of 129 elected representatives, the province will now have to pay twenty-two new members of parliament, plus their staff.
This appears to be an argument against the details of the specific implementation proposed by Ontario, and doesn't really pertain to the merits of MMP and FPTP. There are still something over 90,000 Ontarians per member, a very large figure in international terms.
6. The 39 members of parliament chosen based on the popular vote are not beholden to any one riding, and thus have no constituents and no accountability to the province's voters. The proposal for MMP also says nothing about how these 39 representatives will be distributed across the province.
The 39 top-up members do not belong to any single riding, true. They do have constituents, each and every Ontario resident. Rather than having one MPP, everyone will have 40 of 'em, and of all political hues. The accountability to the province's voters is less clear than the FPTP members, but that's point 7. The top-up members will be elected across the province, without the sub-regions that bedevil the process in Wales and Scotland.
Two other matters of interest to psephologists: Overhang top-ups will not be used; this is less than optimal, but on a 7:3 ratio with global lists, overhang seats shouldn't arise regularly. The distribution will be by the Hare method, using largest remainders, rather than the d'Hondt method used in the UK. This is welcome, and within hailing distance of optimal.
7. The 39 bonus members get chosen from lists submitted by the political parties. These lists are basically Top Ten lists of each party's candidates; if it is determined that the NDP get six extra representatives, then they'd take the Top Six from the NDP list. That's 39 unelected representatives out of a total of 129.
The proposal is to use closed lists, determined by parties. It's generally considered a less than ideal method of ranking top-up members, as it restricts choice to members of the party rather than the whole electorate. These top-up members owe their primary allegiance to the party, but must also use their position in parliament to advance their party's cause, and hence their own individual career. It's relatively simple to move from closed lists to open lists, if that's required in the future.
It is completely inaccurate to say that these representatives are unelected. They owe their election to the success of their party, rather than being part of the largest minority in any given riding.
8. Even under MMP, the candidate you vote for could still remain unappointed to parliament after the 39 bonus members are chosen.
Voters will have two votes. One for the individual candidate in their own riding, and one for the party in the province-wide top-up list. It is almost impossible for both ballots to go to waste; the one voter for the Eating Puppies For Breakfast Party will have no effect on the outcome of riding or provincial election, but almost everyone else will have a bearing in one or both contests.
The key point, almost eclipsed in both Grrliz and Mat GB's comments, is the final outcome. FPTP almost guarantees that one party will have an majority of seats in parliament, even though they've only got about 40% of the votes. Under MMP, it's only possible to govern alone if a party wins 65 of the local ridings, or half the party vote. That would mean they had support of more than half the voters, and could truly be said to be the people's choice. Assuming that that doesn't happen, it will be necessary for the largest party to govern in coalition with one or more smaller parties. Voters for the NDP, voters for the Greens will, eventually, find themselves in government.
Our preferred method, Probabilistic Representation (get all the ballot papers in one place, pull one out, that person is the winner), is not under consideration here. We agree with Mr. GB that the Single Transferrable Vote, as used in Ireland, is the least bad alternative. MMP is probably the best that the rest of the English-speaking world can do, given its preference for parties rather than people. It's certainly more democratic and inclusive than FPTP and list-PR. Alternative Vote Plus seems to have died a quiet death; it adds a small top-up element to a system that vastly favours the local two parties, and we do not mourn its passing.
