The sensitivity of the Ontario election results to the Green Party vote
Oct 16, 2011

The sensitivity of the Ontario election results to the Green Party vote

One of the interesting results of the Ontario general election was the collapse of the Green Party from its 2007 peak of 8% down to just over 3% in 2011. In this post I run the math on various counterfactual scenarios to see what would have happened had this drop not occurred, or only partially occurred, and what the result might have been if some portion of this vote wasn't transferred to the Liberals or NDP.

This post follows from my previous discussion surrounding the politics and significance of the Green Party in the 2011 Ontario election, and attempts to quantify to what extent this role might have been. A five point change between Liberals and PCs would have undoubtedly resulted in a very different election, but the question is how significant this five point drop in the Greens was?

Of the 5% drop between 2007 and 2011 for the Greens, I considered various cases where x% of the total vote was taken from the 2011 Liberal vote and given back to the Green party, and y% was taken from the NDP vote and given back to the Green party, for various x and y where x%+y% was smaller or equal to the 5% total drop. The outcome is listed as the number of Liberal/PC/NDP seats respectively. For reference, the actual outcome of the election was 53/37/17. Fifty-four seats are needed for a majority.

The results:

5% Lib                    44/43/20
4% Lib                    45/42/20
4% Lib, 1% NDP     45/42/20
3% Lib                    46/41/20
3% Lib, 2% NDP 49/41/17
2% Lib                    49/39/19
1.5% Lib                 51/39/17
1% Lib, 1% NDP 52/38/17
1% Lib                    52/38/17
4% NDP                 54/38/15
5% NDP                 55/38/14


The above indicates that this change does matter, but isn't enormous. In the most extreme scenario where we assume the entire 5% Green vote went to the Liberals, and consider what would happen if this were not the case, the Liberals and PCs are in a statistical tie for a minority government. This is the category where the Liberals would have to make a considerable strategic alliance with the NDP in order to form a government (or possibly some other scenario such as Lib-Con temporary alliance).

However, even if only 2 or 3 percent of the total 5% went to the Liberals, without that we quickly get out of the "major minority" category of 53 seats and down into the mid or high forties. At the rather unlikely number of 53 seats when 54 makes a majority, the Liberals can more or less govern as a majority requiring convincing of only a single member of either party to go with them. It will take a fully whipped vote of both opposition parties without a single dissension to bring down the government. However, at even 49 votes (for 2% shift), we quickly get into the territory of needing to significantly listen to the other parties and appeal to their leadership for support opposed to just singling out some backbench person to dissent or switch parties. That changes the politics considerably.

One interesting thing that comes out is the narrow influence of the NDP. Based on political considerations, my guess is that more Green voters that shifted ended up voting Liberal and not NDP. Before the election, poll aggregators like were indicating a very narrow range of possible seat outcomes for the NDP while the Liberals and PCs had a very wide range. This is reconfirmed here where even had a full 5% shifted to the NDP, were that not to happen the Liberals would only gain 2 seats, and 1 seat at 4% (it makes no difference below that). Ironically, this is the crucial seat to give a majority but that is merely coincidence.

Ultimately, I think the largest influence of the 2007 Green vote was in pushing the Liberals to try and appeal to this demographic. They did move this way in terms of policy and it is reasonable to assume succeeded in attracting some portion of this vote. While it would likely have been a Liberal minority government regardless, the circumstances of that minority government do depend to a reasonable degree on the extent to which they secure this Green vote and undoubtedly would result in a different realpolitik.

Comments on the methodology: 

Firstly, in the interest of simplicity, I ignored changes in parties outside of the Liberals, PCs, NDP and Greens; any difference from this is largely inconsequential compared to the changes I am imposing. Secondly, I assumed that zero previous Green voters would vote for the PCs and that every 2007 Green voter that didn't vote Green in 2011 either voted Liberal or NDP.

The most significant issue with the above is that I only compared the 2011 results to themselves, and not to the 2007 results. As such, my x% and y% for the proportion of that ridings total vote shifted to the Liberals and NDP, respectively, are constants across every riding, including ridings where the Green vote is either very significant or entirely insignificant. If I compared to 2007, I could have looked within each riding to determine the percentage of the Green vote drop, and then give a fixed proportion of that riding specific drop to each party respectively. The reason for not doing this was simply due to difficulties with the 2007 data as presented by Elections Ontario. I don't expect substantial changes in outcomes, however.

One other issue is the drop in the voter turnout from 52.8% to 49.2%. If supporters of all parties stopped voting with equal likelihood, this wouldn't matter. However, I think there is a reasonable case to be made that Green voters are the type of people who are somewhat less likely, as an aggregate, to come out to vote than other parties. This would make the influence described above marginally less influential.


Spreadsheets of the election data can be found here. This data only involves the four major parties; it also shows the various other columns I added to do the above computations. The raw data originally came from Elections Ontario. I want to extend my thanks to the good folks at RealSpace3D for doing the technical work to get the data into a usable format (something that Elections Ontario really ought to have done themselves in this spirit of accessibility and transparency).

Similar analysis of counterfactuals for the 2011 Canadian Election can be found here and here

Thoughts on this post? Comment below!

Share this post:

Tweet It! Facebook Add Feed Reddit! Digg It! Stumble Delicious Follow


Mark Francis said...

The Green Energy Act helped the Liberals attract green vote, I think. Also, Hudak campaigning to scrap the Act may have driven nervous greens into Liberal's waiting hands.

bazie said...

I definitely agree. I also think the NDP had an opportunity in 2009 to attract the Green vote during the leadership election where it might have been Peter Tabuns, but after electing Andrea Horwath, the average Green was more likely to vote Liberal than NDP.

Post a Comment

Frequent Topics: