Why MMP is right for Ontario

Proportional Representation (PR) is a bit of a pet issue of mine, so I've been reading the blogs with great interest. As usual, the arguments against PR are leaving me a bit unimpressed. One series of posts in particular has caught my attention today. I'd like to debunk some of the arguments made in this post.

"What is the goal of our legislature? There are two basic positions you can take on this: it is about electing the best representatives who will represent the views of the people or it is a means of reflecting the momentary popularity of various parties at a certain time."

This is a false dichotomy; it belies the fact that, in Canada, the vast majority of people base their votes on party affiliation first, the leader of a party second, and local candidates last. The casual democratic citizen in modern Canada does not sit down and carefully consider the pros and cons of each individual candidate in their riding - they pick the party they like, and vote for the candidate running under that party's banner.

So there is no dichotomy here; the two basic positions are in fact the same position. The goal of our electoral system should be to elect the best quality representatives who will represent the views of the people, as expressed by the party for which they vote. As long as we have political parties, this is the way it is going to be, so it seems to me that our goal ought to be to maximize the utility and proportional accuracy of our electoral system with respect to its reflection of the will of the voters. To this end, there are really two broad options:
1) Abolish political parties - all candidates run on their own merits;
2) Create a framework through which the amount of representation a party has is directly proportional to the amount of people who voted for it.

I don't think there is much support for the first option. The second option is clearly favoured by Canadians; when most people I talk to about this topic find out that this is in fact not how our system works, they are usually quite surprised.

There is a third option, wherein we keep the antiquated system we have now. This system does not reflect the will of the majority of the voters, and it certainly does not encourage the election of the best candidates. I am not going to make arguments to this effect right now, but rather I hope to do so throughout this post.

"Here are the 2003 results for Toronto-Danforth:

Marylin Churley (NDP) 47.14%
Jim Davidson (L) 31.63%
George Sardelis (PC) 16.95%
Michael Pilling (G) 3.53%

Here are the 2006 by-election results for Toronto-Danforth

Peter Tabuns (NDP) 47.8%
Ben Chin (L) 38.9
Georgina Blanas (PC) 10.0%
Paul Charbonneau (G) 2.1%

The seven percentage point jump in Liberal support [in Toronto--Danforth between 2003 and 2006) can only be explained by the candidacy of Ben Chin. It is not reflective of party popularity at that moment. The party was getting into a lot of trouble over the placement of the new natural gas plant so close to this environmentally conscious riding but Chin was able to get voters out to support him. In short, local candidates matter."

There are many things wrong with this statement. First of all, to say that a 7 point jump in support in elections held three years apart is entirely due to the candidacy of one man is assuming a great deal. It may be true, but proving this to be the case would require a lengthy polling process; in absence of this process, it is impossible to say anything for certain.

The results shown really prove my point: look at the NDP vote - two different candidates, three years apart, received the exact same share of the vote, precisely because they were running for the NDP. The Liberals' share of the vote increased by 7%, mostly at the expense of the PCs. Bear in mind that we're talking about Toronto--Danforth here, a very leftish riding - formerly PC support going Liberal is not unusual, nor is it surprising, and it is patently absurd to claim that such a phenomenon can only be explained by the candidacy of one man.

Let's grant that at least some of this increase is due to Ben Chin, and some of it probably is - we are not talking about a huge increase here. No one is denying that local candidates don't matter, and it's a red herring to argue against such a position - local candidates do matter. But they don't matter as much as party affiliation. This is precisely why there are some ridings that are considered "safe seats;" it is readily acknowledged that the chances of any party besides the dominant one in that riding winning are extremely low, and a good local candidate for an opposing party, most of the time, is not going to change that.

Moving on to part 2:

MMP entrenches power in the hands of Ontario's political elite and takes it away from the average citizen. How? Well, simple it adds 39 seats awarded to political parties who decide who will go into the legislature. Nearly a third of the Ontario legislature will be parachuted in by partisans.

This argument is a very common one; it sounds logical, until you realize that it is being made in defence of a system - that is, the current first-past-the-post electoral system - that is seeping with partisanship, and quite regularly adorns our legislatures with partisan hacks.

So the objection it seems is not so much to a legislature filled with partisan hacks, but to a legislature filled with partisan hacks who are selected by the party they represent. But aren't local candidates selected by the parties they represent also? Of course they are - and not only that, but local candidates are not allowed to run unless they have the personal, signed approval of the leader of the party. Talk about entrenching power in the hands of the political elite!

So, then, it seems the objection is not so much to a legislature filled with partisan hacks who are pawns of the party elite (and is that not a bit redundant anyway?) but rather to a legislature filled with an amount of partisan hacks proportional to what the people voted for. That is the real objection; that is the motivation at the core of all the arguments against proportional representation. Once you get right down to it, and strip away all the obfuscations and red herrings from the arguments against PR, all your are left with is a fundamentally anti-democratic instinct - that little things like the will of the people shouldn't really have much bearing on how the legislature is constructed. I am not saying that the people who oppose PR are anti-democratic. Many of them are - and I have debated with many people who are not shy about those view - but most arguments against PR are so veiled by cant and bullshit that their true implications, as simple as they are, are sufficiently and so skillfully masked that reasonable, democratic-minded people end up making them without even realizing the implications of their arguments.

"Even assuming that parties use the most democratic means at their disposal, a one member one vote of their party membership, how many people are choosing are MPP's? How many members does the Green Party of Ontario actually have? I sincerely doubt it's anywhere close to the 100,000 people that choose each and every MPP in the legislature today."

I was rather stunned to see this argument, because it completely ignores that our current system works in almost the exact same way! Let's compare the two systems:

Current System:
Local candidates are decided by the party. The voters then decide on those local candidates.

Proposed System:
Same as above, except an additional step is added; 39 extra seats are distributed. The candidates for these seats are decided by the party. The voters then decide which party they want to vote for.

To argue against parties selecting their own candidates, given that our current system is based on that very principle, is absurd.

One good point is made however:

"By asking political parties to put forward their lists before the election you deprive them of the most democratic way of making their lists i.e. using the election results and putting in the candidates who lost close races."

I actually agree with this; my preference would be that local candidates, defeated by only small margins, should be on the top of the party lists from which the parties draw their apportionment of the 39 extra seats distributed based on a party's share of the popular vote. But this is a highly technical argument that can only really be appreciated by hardcore political geeks, and since I already agree with the subject of this Fisking, I'm not going to go any further with this, except to say that once we get the proportional system in place, it will be much easier to make small changes here and there, like improving the way the lists are written and such.

On to part 3:

"A party receives seats based not purely on the percentage of the vote it receives but instead based on its popularity in general minus its popularity locally. A party's allocation is determined by figuring out how many seats out of 129 it 'should' get and subtracting the number of seats won locally. A party that does well locally could get ZERO seats from the lists. This scenario is described by the assembly itself. Why is this a problem?"

Indeed, why is this a problem? If, according to the percentage of the popular vote, a party is supposed to receive 40% of the vote, and they win 65 of the 90 local seats (which would give them 65 of the 129 in total, ie: 50% of the seats) why should they get additional seats on top of this? They are already over-represented; they won 40% of the vote and have 50% of the seats. The purpose of this system is precisely to avoid parties being grossly over-represented in the legislature. Giving more seats to a party that is already over-represented doesn't make any sense at all.

"[The proposed system] favours parties who do poorly at the local level and disadvantages parties with strong local organization. Parties with good candidates that voters actually vote for get nothing, parties that have poor candidates that could never win locally get seats."

This is a gross mischaracterization of what the proposed PR system actually does. It does not "favour" parties who do poorly. There would still be 90 MPPs elected at the local level - that's over 3/4 of the total. Parties who do poorly locally would not receive any of these MPs. That is hardly to their advantage. In fact, in the example I gave above, the hypothetical party managed to win a bare majority government (50% of the seats) entirely by doing well locally. Parties with good candidates that voters mark their ballots for would not get "nothing" - they would get elected MPPs, the most tangible form of political currency, as a reward.

But in addition to that, the local candidates of smaller parties would receive increased attention. Moreover, the smaller parties would be even further encouraged to run the best candidates possible locally, because a strong local campaign, even if it doesn't result in winning that local seat, would still be very likely to increase the party's share of the popular vote, which would likewise increase their overall seat count due to proportional representation. And, for an additional benefit to local organizations, think star candidates: instead of a party leader bumping a favoured local candidate in favour of a parachuted star candidate, such a star candidates could just be given a high spot on the party's list. To add even further to that (the benefits just don't stop!) by giving voters two votes - one for a specific candidate and one for a party - voters will not be forced into a dilemma of having to pick between the party they support, or a candidate for an opposing party that they nevertheless like; they can vote for the candidate they like, while still voting for their preferred party - so it's a win-win situation for locally-minded voters.

So you see, far from diminishing the importance of the local campaign, the proposed system of PR would actually encourage and be a boon to strong local organizations.

Wasted votes. Fair vote and their ilk are always complaining about wasted votes. Under this system my vote would, in all likelihood, be wasted. As a Liberal in an NDP riding, I still have no say. I will admit this is not a big problem for me personally, but it is one of the reasons we are having this referendum. If the proposed system doesn't fix the problem, what's the point of tearing up our democratic traditions.

Very dramatic language which masks the fact that this argument also doesn't make any sense, because it is simply not true. Yes, Fair Vote Canada does rail against wasted votes. Under the current system, any votes which are not cast for the winning candidate are cast aside; they count for nothing. So a Liberal living in an NDP riding votes Liberal, the NDP wins, and that is the end of it. The writer of the above post apparently doesn't have a "big problem" with this, which is his right, but he propagates an outright falsehood by saying that the proposed PR system would not remedy this situation, because it quite obviously would.

Under the new system, a Liberal living in a strong NDP riding would case their vote for the Liberal candidate, and the NDP candidate would probably win; but that would not be the end of it. The Liberal voter also gets to mark a box indicating preference for the Liberal Party. These latter votes are the ones which would determine the apportionment of the 39 additional seats after the 90 local MPPs were elected. So even if your preferred local candidate doesn't win, your vote could still help yourparty of choice win additional seats.

Then again, it might not; in some cases, a party may do so well in local campaigns that they don't need any additional seats. But if this were the case, that would mean that the voter's preferred party must have done extraordinarily well in the election, hardly a reason to feel disenfranchised - rather a reason to celebrate.

"It assures perpetual minority governments... I don't mind the odd minority government but having an election every 18 months for the rest of my life is not an appealing prospect."

This is a specious argument, because the Canadian experience with minority governments is not in fact what minority governments are usually like. The fact of the matter is, minority governments are very, very, very common in the world. In fact, in almost every democratic country in the world, they are the norm. Take the example of Germany, which has an electoral system that is very similar to the one being proposed for Ontario (except in Germany's system, there are even more list seats, making minority governments even more likely). Germany almost always has minority governments - it has one right now in fact; Angela Merkel's Christian Democrats don't control a majority of seats. And before Merkel, Gerhardt Schroeder headed a Social Democrat minority government. Governments in Germany - almost all of them minorities - tend to last 3 or 4 years, hardly an election every 18 months.

So, then, why are Canada's minority governments so unstable? Precisely because of the current electoral system! The major parties capable of forming government are well aware that it is possible to win a majority government with only 40% of the vote. They do not see minority governments as a chance to actually govern, but rather as an extended pre-writ election campaign. They spend the entire time watching the polls, biding their time, and waiting for the most opportune moment to trigger an election, usually after about 18 months. This is usually once either the minority government itself cracks that magic 40% number, or once the opposition is ahead of the government in the polls. The goal in our system is always to win a majority, because it is so easy to do so.

Under PR systems, however, minority governments are not the exception, but the rule. Provoking an election every few months in such a system is pure electoral folly - since the resulting government will more than likely be a minority government anyway. Also, The People (particularly Canadians) tend to get pissy when they're dragged to the polls every few months (personally I love voting, but I'm weird like that); Canadians would surely punish a party if it insisted on continuously provoking unnecessary elections.

So the key difference is: under our current system, parties have an incentive to defeat minority governments and trigger elections. Under the proposed PR system, parties have an incentive to try to compromise in order to keep the legislature alive.

"It seems to place a higher respect for the list vote than the local vote."

Finally, this point is countered by simple mathematics: 90 local MPPs vs 39 list MPPs. Which do you think is more important? Remember, numbers don't lie.