The decennial readjustment of the number of MP per province and establishing the boundaries of electoral districts for the federal House of Commons began in October 2021 and February 2022, respectively. Canadians often like to congratulate themselves for having eliminated gerrymandering from our political system, in contrast to the Americans who have honed it into a dark art. While we eliminated gerrymanding in Canada by delegating the task of re-establishing electoral districts to a series of independent federal electoral boundaries commissions in the 1960s, Parliament has deliberately maintained a significant malapportionment of MPs per province through three consecutive iterations of the Representation Formula have since 1985 which contain many exceptions to the principle of Representation by Population.

The Senatorial Clause guarantees that a province must hold as least as many Members of the House of Commons as Senators, and the Grandfather Clause as of June 2022 guarantees that no province shall hold fewer MPs than it had during the parliament elected in 2019. The Representation Rule would still grant Quebec (but not British Columbia, Alberta, and Ontario) a variable number of compensatory seats so that its share of the House of Commons matches its share of the population of the provinces if the Grandfather Clause did not already bridge the gap. Parliament alone created the Grandfather Clause and Representation Rule and could therefore repeal them both unilaterally under the Section 44 Amending Procedure, but only a constitutional amendment under the Unanimity Procedure could repeal the Senatorial Clause – and the legislative assemblies of the Atlantic Provinces will never agree to it.

In general, redistribution in the lower houses of federal parliaments in Australia, Canada, and the United States broadly occurs under two steps:

1. allocating the number of MPs per polity based on their populations
2. establishing the boundaries of the electoral districts themselves within each polity

Plugging in figures from the census into a mathematical formula determines the number of MPs per province or state. All three countries profess to guarantee what we typically call in Canada “Representation by Population,” which simply means that the proportion of MPs per province or state should mirror the proportion of the population held by each province or state. An uneven allocation of MPs under this step results in malapportionment. Nick Seabrook in his recent One Person, One Vote: A Surprising History of Gerrymandering in America cites the infamous “rotten boroughs” of the United Kingdom as the historical archetype of malapportionment.[1] In federations, malapportionment pertains to problems with the mathematical formula that favour some provinces as a whole over other provinces, rather than to how electoral districts within each province are drawn up. In contrast, gerrymandering arises from step 2 of this process when partisan majorities in state legislatures in the United States deliberately design congressional districts that will favour their party over the other within a state in future federal elections. Gerrymandered electoral districts therefore usually take bizarrely serpentine, arbitrary, and artificial forms instead of following geographical boundaries and otherwise hewing mainly to simple squares and rectangles. Yet these gerrymandered electoral districts within a state could nevertheless still contain equal populations.

Malapportionment and the systemic over-representation of some provinces makes the ten provincial electoral quotas under the Electoral Boundaries Readjustment Act diverge from a low of 38,583 in Prince Edward Island to highs of around 116,000 in Ontario and British Columbia. The electoral quota is obtained by dividing the population of the province by its number of MPs and comes to the average number of people per MP, and thus what should be the population of each electoral district within that province. The Federal Electoral Boundaries Commission of each province can deviate from its electoral quota by up to ±25% in the normal course of events and even lower than -25% in sparsely populated northern regions. In 2022, the Commissions for Alberta and Manitoba have decided to establish electoral districts (even the northern ones) all within around ±6%, which Saskatchewan’s has re-drawn a northern riding with a significantly smaller population.

In other words, Members of the House of Commons in the over-represented provinces represent fewer people than their colleagues from under-represented provinces. This, in turn, means that the votes of Saskatchewanians, Manitobans, New Brunswickers, Nova Scotians, Prince Edward Islanders, and Newfoundlanders and Labradoreans carry more weight on average than the votes of British Columbians, Albertans, Ontarians, and Quebeckers even though our electoral districts follow objective criteria and are not designed to favour one party over others. And within a province, voters in electoral districts with populations significantly lower or higher than the province’s electoral quota corresponding hold voters worth more or less relative to other voters in their province.

We can measure the malapportionment within the latest iterations of the Representation Formula from June 2022 by comparing the proportion of the province’s total number of MPs in the House of Commons to the proportion of the province’s population out of the total population of the ten provinces. The difference in percentage represents the distortion; a negative percentage signifies over-representation, and a positive percentage indicates under-representation. The Representation Formula under section 51(1) of the Constitution Act, 1867 only pertains to the provinces, so these calculations therefore show the percentage of each province’s population compared to the population of all ten provinces, not to the population of Canada as a whole. Parliament provides each territory one MP under section 51(2) of the Constitution Act, 1867 instead, separate from the Representation Formula. These figures show that Newfoundland & Labrador, Prince Edward Island, Nova Scotia, New Brunswick, Manitoba, and Saskatchewan are all over-represented in the House of Commons relative to their proportions of the population of the ten provinces combined. In contrast, British Columbia, Alberta, and Ontario remain under-represented. Surprisingly, Saskatchewan – not Prince Edward Island – is the most over-represented province of them all, while Ontario remains far and away the most under-represented.

The Globe and Mail published an editorial on 30 August 2022 on this very subject, arguing that Parliament should amend the Representation Formula once more to “bring Rep by Pop back to Parliament.” The editorial board acknowledges that we will never repeal the Senatorial Clause under the Unanimity Amending Procedure but argues that Parliament should repeal the Grandfather Clause altogether (though it doesn’t mention the Representation Rule at all). But since merely repealing the Grandfather Clause alone would reduce Quebec’s representation from 78 to 71, Parliament should also repeal the current formula for calculating the electoral quotient and replace it with a new formula that uses Quebec’s representation at 78 MPs as the baseline to determining the number of MPs for the other provinces. This would build on earlier precedents in Canadian history. The first iteration of the Representation Formula (1867 to 1946) and the fourth Representation Formula (1974 to 1985) similarly used Quebec’s number of MPs (65 and 75, respectively) as the baseline for calculating the electoral quotient and the number of MPs for each of the other provinces. The Globe continues:

“Parliament should scrap the grandfather clause, and return to a system that starts with a fixed number of seats for Quebec and allocates seats to the other provinces in direct proportion to their populations.

If we fix Quebec at its current 78 seats – 110,000 people per riding – and apply that quotient to the other provinces, we’d get a House of Commons with 359 seats. That’s 21 more than today. Manitoba, New Brunswick and Newfoundland and Labrador would each lose a seat and Saskatchewan would lose three. Ontario would gain 14, Alberta would add seven and B.C. six.

The Atlantic Provinces would still get slightly more seats than under rep by pop, thanks to a Dingwall entitlement known as the Senate floor rule, whose removal requires an (impossible) constitutional amendment. Leaving that aside, a couple of small changes can bring Canada a lot closer to equality of votes, and fairness, without adding more than a handful of MPs.”

Even here, the Globe still can’t quite fathom a Representation Formula that caps the number of MPs and instead proposes expanding the House of Commons by 21 MPs in one fell swoop. And Quebec’s electoral quota is more like 109,000 than 110,000 – but who’s counting?

Canada’s Representation Formula incorporates so many exceptions to Representation by Population that it produces systemic malapportionment and over-represents six of the ten provinces in the House of Commons; the formula therefore cannot cap the number of MPs without utterly destroying Representation by Population and instead has to allow the House of Commons to grow unchecked. The second problem depends on the first. The Representation Formula guarantees the systemic over-representation of Newfoundland and Labrador, Prince Edward Island, Nova Scotia, New Brunswick, Manitoba, and Saskatchewan by ensuring that these provinces can never lose seats even as their populations decline outright or grow more slowly than average. This therefore necessarily means that Ontario, Alberta, and British Columbia can only achieve some semblance of their proportionate representation relative to their shares of the population by granting them more MPs upon each decennial readjustment. The House of Commons therefore continues to expand in size every ten years because we cannot properly redistribute the MPs whom we already have between the provinces.

Indeed, we should not even call what we do a “redistribution” in Canada but instead something more like a “readjustment.” Logically, a redistribution can only happen when some polities lose MPs and others gain those MPs. In contrast, the House of Representatives of Australia and the United States House of Representatives properly undertake redistributions: these legislative bodies maintain fixed numbers of MPs and therefore rely solely on changing the ratio of MPs per state in accordance with changes in population between the states after a census. Of the three federations, the United States and Australia do well on apportionment or allocating MPs per state, while Canada has instead entrenched a permanent malapportionment. Canada and Australia do well on eliminating gerrymandering by allowing independent commissions headed by judges re-establish the electoral districts. We ensure that federal commissions draw federal boundaries and that federal electoral agencies administer federal elections. In the United States, each state legislature determines how to establish the boundaries of federal congressional districts, and the majority party in most state legislatures engages in overt gerrymandering; however, some states like California and Michigan have finally started delegating this responsibility to commissions. Based on Seabrook’s analysis, I would argue that these commissions do not follow the Australian-Canadian model because they consist of self-identified and registered Democrats and Republicans and independents instead of simply non-partisan experts. The United States can only ever achieve bipartisanship rather than non-partisanship. Only Australia does well on both allocating MPs and on establishing the boundaries of electoral districts.

Canadian politicians and even most political scientists have become so obsessed with the number of MPs per province that they have forgotten that “Representation by Population” refers to the ratio or proportion of MPs per province and not to numbers. The original Representation Formula assigned Quebec 65 MPs and calculated the electoral quotient and the number of MPs for each of the other provinces simply because Quebec and Ontario each held 65 MPPs in the Province of Canada’s Legislative Assembly. If a province contains 25% of the population, then it will likewise hold 25% of MPs in the lower house; the numbers themselves don’t matter provided that the proportion or ratio of MPs between the polities remains intact. In a House of Commons of 100 MPs, this hypothetical province would hold 25 MPs; in a Commons of 200 MPs, it would get 50 of them, etc. The Globe and Mail demonstrates this same obsession with numbers over ratios in its editorial. Pure Representation by Population will never exist in Canada because the Atlantic Provinces will never agree to repeal the Senatorial Clause, but Parliament alone could come close by repealing the Grandfather Clause and Representation Rule and capping the number of MPs. Saskatchewan, Manitoba, Quebec, Nova Scotia, and Newfoundland and Labrador would lose MPs to British Columbia, Alberta, and Ontario.

The Globe and Mail also neglected to mention in its editorial that repealing and replacing the Representation Formula at this time would necessarily nullify all the work that the current crop of Federal Electoral Boundaries Commissions have done in 2022; we would then have to re-calculate the number of MPs per province and re-establish a new set of Commissions to work on the same census figures from 2021. This happened in the 1980s, but Parliament ought not to follow that precedent in the 2020s or any other decade.

Similar Posts: 

• Decennial Readjustment of the House of Commons
• “The Origins of Canada’s Electoral System and the Constitutional Considerations of Electoral Reform” (2020)
• The New Democrats’ Anti-Constitutional Stance on Electoral Redistribution (September 2011)
• My Column in the National Post on the New Democrats’ Unconstitutional Bill to Give Quebec a Fixed Proportion of Seats (October 2011)
• Section 52 and Over-Representing Quebec in the House of Commons (October 2011)

Notes

[1] Nick Seabrook, One Person, One Vote: A Surprising History of Gerrymandering in America (New York: Pantheon Books, 2022), 15.