The MPP Tallying Project

The truth is, free time is not something I have a lot of these days. When I do find some, I’m more inclined to do nothing of great consequence, including putting off cleaning my apartment which is a total disaster zone. Instead I prefer catching up on my blog reading, renting a few movies, or visiting my “nephew” Jackson, BeeGoddessM and Stephanie‘s new puppy. (I also pay attention to what BeeGoddessM and Stephanie have to say when Jackson allows me. 🙂 )

However, the Harper government’s announcement recently that it would pass legislation to have fixed election dates have rekindled my interest in electoral reform. At this time, I’m particularly interested with the accusation of nay-sayers that the chief reason proponents want reform so badly is because it would heavily favour the party they support. However, an amount of number-crunching quickly reveals that a mixed member parliamentary (MMP) system would not only better represent the people’s will but also equally affect all parties favorably and negatively.

In order to demonstrate this point, I have been collecting riding-by-riding results of elections held federally and in Nova Scotia, New Brunswick and QuĂ©bec in the last 20 years, entering these results in Excel spreadsheets, and applying Gregory Marrow’s “Option A” projection formula of an MMP system. It’s important to point out that his formula leads only to a projection, as it doesn’t account for individuals who would split their vote for a candidate from Party A on their local ballot but for Party B on their regional ballot. Indeed, the formula assumes that a vote for Party A under the current first-past-the-post system (i.e., the one and only local ballot voters get) would translate into a vote for Party A at the regional level as well. However, if given two ballots — one local and one regional — there would likely be less strategic voting (i.e., voting for a second-choice candidate because his or her party is more likely to win locally and thereby block the advance of a last-choice candidate who is equally likely to win). For instance, I could choose to vote for the candidate from Party B locally on the grounds that I like the candidate personally, but regionally I would vote for Party A, which is my “traditional” party.

My number crunching has shown that, while it is true that the NDP (federally) would have consistently emerged with more seats federally in the last 20 years, both the Liberals and the Conservatives (and the Parti Québécois provincially in Québec) would not have fallen out of grace as spectacularly as they have. The current system yields some truly perverse results.

  • The most striking recent instance at the federal level was the 1993 election that saw the Conservatives fall from a solid majority to a paltry two seats in Parliament (or 0.68% of the seats in the House of Commons), despite earning 16.06% of the popular vote nationally.
  • In 1993, again at the federal level, the Liberals picked up 60.34% of the seats with only 41.19% of the popular vote.
  • In 1984, it was the federal Conservatives’ turn to obtain 74.82% of the seats on the weight of 50% of the popular vote.
  • Provincially in QuĂ©bec in 1998, the Parti QuĂ©bĂ©cois formed a majority government despite getting 1% less in popular vote than the opposition Liberals.
  • And perhaps the most remarkable outcome in Canadian history was the 1987 general election in New Brunswick, which saw the Liberals snatch all 52 seats in the provincial Legislative Assembly.

But what I have found even more interesting is how the raw numbers can tell untold stories. For example:

  • Many were those who were surprised by how the Conservatives picked up 10 seats in QuĂ©bec after being unable to pick up any 18 months earlier. Granted, Conservative support in that province exploded from 8.8% to 24.6% from one election to the next. However, had an MMP system been in place both times, the Conservatives could have earned 7 regional seats in 2004 and a total of 17 seats in 2006. In other words, the surprise of 2006 would have been considerably lessened.
  • Also in QuĂ©bec, we tend to think that the federal NDP has never had a base of support in that province, but that isn’t true. In the 1988 “free trade” election, it got nearly 14% of the popular vote …but not a single seat. Under MMP, it would have obtained 11 of the province’s 75 seats. In the last decade, much of the left-of-centre vote has migrated to the Bloc QuĂ©bĂ©cois, whose positions are very much like the NDP’s except on the matter of QuĂ©bec’s position within Confederation.
  • Over the years, at both the federal and provincial levels, there has been significant pockets of support for parties we don’t associate with certain provinces, like the NDP in Alberta or New Brunswick, or the Liberals in Saskatchewan.

For political junkies like myself, these numbers mean something. And I’m sure I’m not the only one to whom they mean something. So if I ever find some spare time, what I’d like to do is import this raw data into an online database that could be easily queried, with or without the theoretical MMP projections. All the information is already publicly available, but it can’t be analyzed easily from event to event. I find it fascinating to discover how some ridings, either federally or provincially, as just not as blue or red or orange as we tend to believe them to be. Rather than simply looking at the number of consecutive times a party has won a riding and being shocked when the riding suddenly switches, it’s much more informative to investigate if another party has consistently come in second during that time, and within what margin.

For in the end, I am not among those who believe that parties are interchangeable, that they’re all equally bad. I think those who have come to take this view are disaffected because, for too long, we’ve allowed a system to yield governments that do not represent our will. It’s all too easy for any party to overestimate their political capital when they manage to get 60, 70, 80 or even 100 percent of the available seats in an assembly. Fair representation would ensure that no single party could emprint an agenda that a vast majority does not embrace.