Note: An earlier version of this article had a calculation error in regards to the Bloc Quebecois. I have corrected it and added additional information in italics regarding calculating voting power.
I have been posting about electoral reform for a while now. Mostly this has been concerning a new system I devised which I named the Bennett Method. I applied it to the results of the recent provincial election but the flaws were already becoming evident. The results just refused to move as close to proportionality as I would like. Re-tooling was in order it was clear.
After some thought I decided to fix what was a major inconvenience of the original idea; its complexity. The original idea had three main objectives. 1. Decouple voting power in Parliament from seat totals and align it with the popular vote.
2. Reward MPs who secured proportional higher vote totals in their ridings. 3. Keep as much of the current system in
place as possible. I have come to realize the second objective would have to be abandoned in the interest of simplicity.
So what does the Bennett Method 2.0 look like? So glad you asked.
Like the previous version this method keeps the First-past-the-post system largely intact. Voters still cast a single vote in a single-member constituency. The candidate who receives the most votes still becomes the MP. But in place of the next several convoluted mathematical steps there is but one: Give each MP a vote proportion to their party's share of the
popular vote.
The formula would be Popular Vote / # of MPs in party = Voting share of each MP.
I would like to draw your attention to the 2011 Federal Election. The Conservative Party won 166 seats (53.89%) with
39.62% of the popular vote. The NDP won 103 seats (33.44%) with 30.63% of the popular vote. The Liberals won 34 seats (11.03%) with 18.91% of the popular vote. The Bloc won 4 seats (1.29%) with 6.04% of the popular vote. Finally the Green
Party won a single seat (0.32%) with 3.91% of the popular vote. Using my system each Conservative MP would have a vote valued at 0.23, each NDP MP vote valued at 0.29, each Liberal vote valued at 0.55, each Bloc vote valued at 1.51, and the single Green vote valued at 3.91.
Voting power in Parliament under the proposed system:
Conservative voting power: 38.18
Liberal voting power: 18.7
New Democrat voting power: 29.87
Bloc voting power: 6.04
Green voting power: 3.91
As a result the FPTP electoral system will have been rendered mostly proportional.
Some eagle-eyed readers may note that the totals above do not equal 100%. This is true. For starters only 99.11% of the popular vote went to parties who won seats. So were did the remaining 3% that is missing go? I think this is due to not rounding numbers on my part. I will do so now. Below are the new voting power totals (with MP share rounded and in brackets):
Conservative voting power: (0.24) 39.84
Liberal voting power: (0.56) 19.04
New Democrat voting power: (0.30) 30.09
Bloc voting power: (1.51) 6.04
Green voting power: (3.91) 3.91
Again, slightly off at 98.92% which is better than the previous 96.7%. But in either case does this matter? As a percentage of 96.7 the Conservatives totals of 38.18 is 39.48%. If rounding is included 39.84 is 40.27. For the NDP it works out to 30.88% & 30.41% respectively. The Liberal totals come to 19.33% & 19.24%. In all three cases it would seem that rounding to the second decimal place provides little in the way of overall change to voting power. Only in the tightest of minority situations would it even matter. As such, not rounding the totals is acceptable.
Advantages of the system I have proposed:
-The voting system does not change at all for Canadian voters.
-It is as understandable and simple as FPTP is.
-It renders the House of Commons mostly proportional.
-It is no more expensive than FPTP is.
-It benefits (or at least does not harm) all of the main political parties in different ways.
-It makes votes for parties in ridings where they will never win matter.
Now, this system is not perfect. It should also be noted that votes for parties that do not win a seat still do not count.
What's more, large numbers of such votes can mess with the math unless explicitly excluded (which I did not do
here). At the federal level this should not be an issue since such parties generally receive less than 1% of votes cast. The
Green Party's high of 6% being somewhat acceptable as well.
So how would things like confidence votes and picking a Speaker work? Surprisingly well as it turns out.
Picking a Speaker (and other scenarios)- Under the current system picking a Speaker lowers the vote total of his or her party. However, under this system I foresee a rule that states the voting power is recalculated at the beginning of each day. In short a Speaker would simply have to drop his party affiliation and his party's total would not be affected. Likewise, re-calculating the numbers each day means vacant seats would not affect a party's total (which they currently do). I would not extend this rule to MPs who end up as independents for one reason or another due to potential loophole abuse. At the same time, poaching MPs from other parties will no longer be advantageous as having more MPs will not mean more voting power.
Forming a Government- While voting power should be based on popular vote I feel the formation of governments should remain based on seat count. The reasons are simple: regional representation and a larger talent pool to draw on.
Confidence Votes- I would keep explicit confidence motions and the Speech from the Throne as being based on seat totals but have all other legislation based on voting power.
Concluding Remarks
As I have repeated often, I don't think this is a perfect system. I do however think I am making it better. As a means of distinguishing this system from the original idea I have decided to call it Single Member-Proportional Vote.
I have been posting about electoral reform for a while now. Mostly this has been concerning a new system I devised which I named the Bennett Method. I applied it to the results of the recent provincial election but the flaws were already becoming evident. The results just refused to move as close to proportionality as I would like. Re-tooling was in order it was clear.
After some thought I decided to fix what was a major inconvenience of the original idea; its complexity. The original idea had three main objectives. 1. Decouple voting power in Parliament from seat totals and align it with the popular vote.
2. Reward MPs who secured proportional higher vote totals in their ridings. 3. Keep as much of the current system in
place as possible. I have come to realize the second objective would have to be abandoned in the interest of simplicity.
So what does the Bennett Method 2.0 look like? So glad you asked.
Like the previous version this method keeps the First-past-the-post system largely intact. Voters still cast a single vote in a single-member constituency. The candidate who receives the most votes still becomes the MP. But in place of the next several convoluted mathematical steps there is but one: Give each MP a vote proportion to their party's share of the
popular vote.
The formula would be Popular Vote / # of MPs in party = Voting share of each MP.
I would like to draw your attention to the 2011 Federal Election. The Conservative Party won 166 seats (53.89%) with
39.62% of the popular vote. The NDP won 103 seats (33.44%) with 30.63% of the popular vote. The Liberals won 34 seats (11.03%) with 18.91% of the popular vote. The Bloc won 4 seats (1.29%) with 6.04% of the popular vote. Finally the Green
Party won a single seat (0.32%) with 3.91% of the popular vote. Using my system each Conservative MP would have a vote valued at 0.23, each NDP MP vote valued at 0.29, each Liberal vote valued at 0.55, each Bloc vote valued at 1.51, and the single Green vote valued at 3.91.
Voting power in Parliament under the proposed system:
Conservative voting power: 38.18
Liberal voting power: 18.7
New Democrat voting power: 29.87
Bloc voting power: 6.04
Green voting power: 3.91
As a result the FPTP electoral system will have been rendered mostly proportional.
Some eagle-eyed readers may note that the totals above do not equal 100%. This is true. For starters only 99.11% of the popular vote went to parties who won seats. So were did the remaining 3% that is missing go? I think this is due to not rounding numbers on my part. I will do so now. Below are the new voting power totals (with MP share rounded and in brackets):
Conservative voting power: (0.24) 39.84
Liberal voting power: (0.56) 19.04
New Democrat voting power: (0.30) 30.09
Bloc voting power: (1.51) 6.04
Green voting power: (3.91) 3.91
Again, slightly off at 98.92% which is better than the previous 96.7%. But in either case does this matter? As a percentage of 96.7 the Conservatives totals of 38.18 is 39.48%. If rounding is included 39.84 is 40.27. For the NDP it works out to 30.88% & 30.41% respectively. The Liberal totals come to 19.33% & 19.24%. In all three cases it would seem that rounding to the second decimal place provides little in the way of overall change to voting power. Only in the tightest of minority situations would it even matter. As such, not rounding the totals is acceptable.
Advantages of the system I have proposed:
-The voting system does not change at all for Canadian voters.
-It is as understandable and simple as FPTP is.
-It renders the House of Commons mostly proportional.
-It is no more expensive than FPTP is.
-It benefits (or at least does not harm) all of the main political parties in different ways.
-It makes votes for parties in ridings where they will never win matter.
Now, this system is not perfect. It should also be noted that votes for parties that do not win a seat still do not count.
What's more, large numbers of such votes can mess with the math unless explicitly excluded (which I did not do
here). At the federal level this should not be an issue since such parties generally receive less than 1% of votes cast. The
Green Party's high of 6% being somewhat acceptable as well.
So how would things like confidence votes and picking a Speaker work? Surprisingly well as it turns out.
Picking a Speaker (and other scenarios)- Under the current system picking a Speaker lowers the vote total of his or her party. However, under this system I foresee a rule that states the voting power is recalculated at the beginning of each day. In short a Speaker would simply have to drop his party affiliation and his party's total would not be affected. Likewise, re-calculating the numbers each day means vacant seats would not affect a party's total (which they currently do). I would not extend this rule to MPs who end up as independents for one reason or another due to potential loophole abuse. At the same time, poaching MPs from other parties will no longer be advantageous as having more MPs will not mean more voting power.
Forming a Government- While voting power should be based on popular vote I feel the formation of governments should remain based on seat count. The reasons are simple: regional representation and a larger talent pool to draw on.
Confidence Votes- I would keep explicit confidence motions and the Speech from the Throne as being based on seat totals but have all other legislation based on voting power.
Concluding Remarks
As I have repeated often, I don't think this is a perfect system. I do however think I am making it better. As a means of distinguishing this system from the original idea I have decided to call it Single Member-Proportional Vote.
RSS Feed
