Dennis Pilon's book The Politics of Voting raises several important points.
Here's a little I want to add to what he wrote:
Pilon writes that Western countries of the late 1900s were ranked in order of proportionality. (He provides two different rankings. Complied by different people they show some variation.)
Denmark and Netherlands are the most proportional.
When Pilon wrote the book, four major countries were using pure single-member plurality - U.S., UK, Canada and New Zealand. They all are among the ones listed as the least proportional.
Since then all but on of these have diversified thier elections through reforms.
(New Zealand now uses MMP (SMP with PR top-up)).
UK uses a variety of PR systems as well as single-member plurality.
US now elects some of the members of the House of Representatives using Alternative Voting (IRV). It now uses Alternative Voting in two of the 50 states.
Canada is only major country in the world that uses FPTP and only FPTP to elect its legislators.
The U.S. was ranked as the least disproportional of the four, but this seems to me to be due to voters in that country self-regulating themselves - a vast majority of them cast their vote to for one of the two main parties, something not done in Canada.
(In fact many in the U.S. think that that only two candidates, one of each major party, run for the U.S. presidency, while actually ten parties compete in organized fashion for the president's seat and more than 200 candidates in all ran for the seats. In fact in 2016, no candidate - not even Hilary Clinton, the most popular candidate - took a majority of the votes. In 2016, neither major party received a majority of the presidential election votes, and in fact the most popular candidate was not elected.)
In Canada, in the last federal election, 67 percent of the votes went to the two most popular parties, a large but not vast majority.
And in fact in the 2019 federal election, the most popular party was not elected to the most seats. The Liberals took more seats than the Conservative party, which took more votes.
This wrong-winner gave minority government to the Liberals, which was in fact the choice of 36 percent of the voters (the most popular choice) in 86 percent of the country (all of the country outside AB/SK). The most popular party (the Conservatives) took a large part of its seats in only one region (the AB/SK sub-region), where it was vastly over-represented.
Because Canadian voters spread themselves over more than five parties, our dis-proportionality looks so bad despite the fact we elected members of five parties to the House of Commons - a more proportional result than in the U.S. where only two parties are represented.
Alongside Canada in the rankings was France, which elects through a majority system with no proportionality. The hallmark of France's two-round election system is that a majority of the voters (at least the system guarantees that a majority of those who cast votes in the second vote) are represented by the winner.
Canada's system does not have this attribute - a third or more of our MPs and MLAs are elected with only the support of a minority of the voters in their districts.
Maps (such as in the Wikipedia election reportage) that merely record percentage of support for party is not useful for showing which party or candidate wins. A candidate with 34 percent of the vote could be a winner or could be unsuccessful. A candidate with 49 percent of the vote could be a winner or could be unsuccessful.
Only a candidate with more than majority of the vote is guaranteed to be known to win a seat under our distorted system, while many do win with much less.
And total vote tallies vary so much from single-member district to single-member district so there is no known number of votes that ensures a winner either. Hence the term "First Past The Post' is a misnomer as there is actually no post at all.
And the non-proportional single-member plurality system creates apparent regionalism, where voters in each region are generalized and made into stereotypes, while their actual voting behaviour is much more shared than it is different.
Regionally votes looked something like this
ON Qu AB/SK BC
Cons 33 10 65 34
Lib 42 35 13 26
NDP 17 11 15 24
Greens 6 5 3 13
BQ 32
In no provinces except Alberta and Saskatchewan did any one party take a majority of the vote. And this is a generalization as well - in many districts within these two provinces not even the Conservative candidate took a majority of the vote.
The same-ness is that in no province do the:
-Conservative voters make up less than 16 percent of the vote.
-Liberal voters make up less than 11 percent of the vote.
-NDP voters make up less than 7 percent of the vote.
-Green voters make up less than 2.5 percent of the vote.
So something like about 40 percent of the voters are the same across the country. (This is based on the provincial flattening of the results, but likely stands up well on riding by riding basis.)
I am not sure this has been calculated in any other place in the literature. [Maybe someone out there can tell me if they have seen this elsewhere.]
if we look at just urban ridings in BC, AB, ON and QU, there will be even more commonality, with Liberal and NDP doing better consistently.
Or if we look just at rural ridings across the country, there will be great commonality. with Conservatives doing better consistently.
So apparent artificially-created regionalism would be shattered by a system that allocates seats proportionally based on provincial vote tallies, or in multi-member city-wide districts. (My "province-based PR system," proposed in a different blog, would show this commonality among the votes by ensuring minimum provincial representation to each major party.)
Any system that allocates seats based on multi-member districts or other large-district (provincial or regional) systems would dispel the artificial regionalism that so divides the country.
As Pilon (The Politics of Voting, p. 58) puts it:
"in the 1993 federal election, such distortion led to results that mapped the country as completely polarized on regional and ethnic lines.
A PR system would have reflected the real diversity that existed within all regions of Canada, thus laying the basis for some accommodation across this regional issue. With PR, parties could campaign across the country, even in areas where they had limited support because they would know that any votes they received would contribute to their overall results. No longer would certain parts of the country appear one-dimensional in their political interests (e.g., Alberta, Quebec). No longer would minority views in a particular locale be orphaned by the way the votes are added up (e.g., Conservatives in Toronto, Liberals in Calgary). And the resulting transparent relationship between votes and seats would allow voters to better understand the various political divisions in the country, and the political necessity of working with others to get things done." (p. 58)
Great point.
Thanks for reading.
========================================
Comments