Would a different electoral system change Canada's Parliament?

Crunching the numbers for different ways to elect MPs

As of now (morning of May 1), the seat count for the Canada’s next Parliament is:

• 169 Liberal

• 144 Conservative

• 22 Bloc Quebecois

• 7 NDP

• 1 Green

• 0 PPC (or anyone else).

What would the election results look like with a different electoral system?

A number of different models have been proposed. Let’s look at three alternatives to our constituency-based, single-member, “first-past-the-post” system.

Ranked-order preference

With this system, voters rank their choice in order of preference. If one candidate does not get over 50 percent of the first-choice votes, the lowest vote getter is dropped from the count. Their supporters are re-allocated to the remaining candidates based on their second preference. This process continues until one candidate has over 50 percent support.

We cannot predict what the result would have been with a ranked-order preference because we don’t know what people’s second or third choices would have been. However, I’ve divided each party’s seat total into three categories:

· Seats the party would HAVE won since the winning candidate got more that 50 percent of the vote. In this category, second and third place choice were irrelevant since they would not have been counted.

· Seats that the party would LIKELY HAVE one since the winning candidate was close to a majority and likely to get it. I’ve arbitrarily set the parameters such that the winning candidate (a) got more than 45 percent of the vote, and (b) was at least 10 percent higher than the second-place finisher.

· Seats that are UNPREDICTABLE given our lack of information about second or third choice preferences. These were ridings with a three or four way split that allowed the winner to claim victory with less than 45 percent of the vote or where the two leading candidates were close enough to each other that we can talk of horseshoes and hand grenades. We’ll call these “coin-flip” wins.

With this categorization, the results are:

Liberals (169 “first past the post” seats):

• 126 guaranteed wins.

• 14 probable wins.

• 29 coin-flip wins.

Conservatives (144 “first past the post” seats)”:

• 100 guaranteed wins.

• 12 probable wins.

• 32 coin-flip wins.

Bloc Quebecois (22 “first past the post” wins):

• 0 guaranteed wins.

• 10 probable wins.

• 12 coin-flip wins.

NDP (7 “first past the post” wins):

• 0 guaranteed wins.

• 1 probable win.

• 6 coin-flip wins.

Green (1 “first past the post” win):

• O guaranteed wins.

• 0 probable wins.

• 1 coin-flip win.

The bottom line: we have no idea who would have on the election with a ranked order electoral system. Growth potential for the smaller parties was very limited. The Bloc finished second in only seven close ridings won by either the Liberals or Conservatives. The NDP finished second in only four close ridings.

Proportional Representation

With this system, each party would submit a national list of candidates in rank order of their preference. A party would then be awarded the number of seats based on their total vote share.

Under a national proportional representation system, the results would have been:

• 150 Liberal.

• 142 Conservative.

• 22 Bloc.

• 22 NDP.

• 4 Green.

• 2 People’s Party of Canada.

There would also be one seat left over because of rounding issues.

The bottom line: we’d still have a Liberal minority. More NDP, a few more Greens, and Maxine Bernier would not only be in Parliament but he’d have a seatmate.

Multi-Member District Proportional Representation

This is a variation of proportional representation to account for regional differences. The country would be divided into electoral districts with each party submitting a list for each region. The seats in the districts would be allocated by the vote share in each location.

I’ve worked the numbers using each province as an electoral district. This produces bottom line result of:

• 150 Liberal.

• 140 Conservative.

• 22 Bloc.

• 21 NDP.

• 1 Green.

If you’ve got your calculator out, you can see a problem. The total number of seats awarded drops to 334. The more (smaller) districts we have, the more rounding issues are created. What’s more, as compared to national proportional representation, the smaller parties get squeezed since one or two percent of the vote is not enough to generate seats in most provinces – but it adds up nationally.

The NDP platform called for the implementation of this kind of system. They do not specify the size of the electoral districts. This matters. The smaller the districts, the more rounding issues are created and the more that smaller parties are squeezed out.

What’s it all mean?

Elections are about aggregating choice. There is no perfect system.

The first-past-the-post system we use in Canada today was inherited from the British. It’s two great advantages are its simplicity and its ability to be sensitive to both local and candidate impacts. Political scientists argue that it is the system best suited for generating stable majority governments but given that six of the last eight Parliaments have been minorities I am not sure this really applies as much as it used to.

The rank-order system generates the most unpredictability. Some argue that it pushes parties to be more “moderate” and “compromising” since being second choice matters a lot. We don’t have a lot of experience with this system in general elections. Alaska has used it for statewide office for about a decade. It seems to have had the predicted moderating effect thus far. In 1952, British Columbia used the system in an explicit attempt by the Liberals and Conservatives to keep the CCF out of power. It resulted in a surprise win for the Social Credit Party. The province then returned to a first past the post system.

The rank-order system has been used more often in party leadership races. If there are a lot of candidates, one effect has been that candidates finishing third on the first ballot have ultimately won more often than you’d expect.

The proportional representation system tends to cause the creation of more political parties since a winning a small number of votes over a wide area can result in winning some seats in Parliament. One result is that the actual formation of governments occurs through post-election negotiations between parties rather than being reasonably clear on most election nights. Another is that it gives much more power to party leadership since placement on the list is the biggest determinator of whether an individual will end up in Parliament. Finally, the system removes all connection between voters, their communities, and any particular elected representative. On the other hand, extent of the “regional divide” in Canada would be smoothed out with a provincially based proportional representation system.

Representation by Population?

19,597,674 Canadians voted in the 2025 election. That means there were, on average, 57,136 voters for every Member of Parliament elected.

But this varies from province to province.

The provinces that can make a legitimate claim to be under-represented in Parliament are:

• Ontario (62,103 votes per MP).

• British Columbia (61,356 votes per MP).

• Alberta (61,149 votes per MP).

Quebec is the province that has the closest alignment between population and the number of MPs with 57,150 votes per MP. Nova Scotia comes close at 52,520 votes per MP.

Provinces over-represented in Parliament are:

• The three territories (15,655 votes per MP).

• Prince Edward Island (24,260 votes per MP).

• Newfoundland and Labrador (39,650 votes per MP).

• Saskatchewan (40,055 votes per MP).

• Manitoba (45,663 votes per MP).

• New Brunswick (46,310 votes per MP).

Many of my friends in Alberta complain that their province seems under-represented in Parliament. They usually point to Ontario and Quebec as having “to much power”. In reality, however, the people in the three Territories, most of the Maritimes, and Saskatchewan are the most privileged voters in Canada.