On November 11, 1947, Winston Churchill said to the British House of Commons,
Many forms of Government have been tried, and will be tried in this world of sin and woe. No one pretends that democracy is perfect or all-wise. Indeed it has been said that democracy is the worst form of Government except for all those other forms that have been tried from time to time.…
The concept has been attributed to Churchill himself, but it’s clear he was quoting another unknown source. In the aftermath of the collapse of the Soviet Union, social scientists have explored with wonder how numerous societies, when provided the choice, typically choose democratic forms of governance (at least in our time).
The imperfections of our democracy compel us to constantly revisit and reexamine how we operate our own form of governance. We are experiencing record low voter turnout, public confidence in Parliament is one of the lowest among our public institutions, and our Upper House is rocked by scandals.
The current Liberal majority was elected on a platform clearly outlining a commitment to transform the way we elect our political representatives:
We are committed to ensuring that 2015 will be the last federal election conducted under the first-past-the-post voting system. We will convene an all-party Parliamentary committee to review a wide variety of reforms, such as ranked ballots, proportional representation, mandatory voting, and online voting. This committee will deliver its recommendations to Parliament. Within 18 months of forming government, we will introduce legislation to enact electoral reform.
Following the election there appear to be more divided feelings on electoral reform, with 25% of the population indicating they would prefer that the government does not follow through with this promise. Although some parties are demanding for a referendum on the issue, the government has indicated they will not do this, instead forming an all-party committee with broad consultation.
From a constitutional perspective, it might be a little more tricky. We explored these challenges on Friday at the 19th Annual Constitutional Cases Conference at Osgoode Hall. Peter Hogg chaired a panel with Yasmin Dawood of UofT, Michael Pal of University of Ottawa, and Emmett Macfarlane of the University of Waterloo.
Although some academics have questioned the very constitutionality of our current First Past The Post (FPTP) system, it was upheld by the Quebec Court of Appeal in Daoust c Quebec. FPTP is not mentioned anywhere in the Constitution explicitly, or spelled out in any constitutional documents. However, there are long standing practices in place which provide meaning to the constitutional text and any gaps in the provisions, which could make it part of constitutional convention.
All of the panelists looked into the complicated question of whether provincial consent would be required for electoral reform, especially in light of Part V of the Constitution Act, 1982, which requires several procedures under ss. 38, 41, 43, 44 and 45. Pal cited Richard Albert to suggest that provincial consent would make changes to the electoral system highly unlikely in the current circumstances.
The constitutional question is further complicated by the Court’s 2014 decision in the Senate Reform Reference. Prior to this decision there was an assumption that Parliament already had the power to unilaterally change our electorate model, as long as the system maintained relative parity of voting power, played a meaningful role in the democratic process, and promoted the right of each citizen to participate in the electoral process.
Prior to this decision, Dawood indicates that electoral reform could occur based on the wording and structure of the Constitution Act, as long as it did not infringe constitutional limits. The problem with the Senate Reference is an allusion by the court to unwritten principles which appear to inform the constitution,
 These rules and principles of interpretation have led this Court to conclude that the Constitution should be viewed as having an “internal architecture”, or “basic constitutional structure.” The notion of architecture expresses the principle that “[t]he individual elements of the Constitution are linked to the others, and must be interpreted by reference to the structure of the Constitution as a whole.” …In other words, the Constitution must be interpreted with a view to discerning the structure of government that it seeks to implement. The assumptions that underlie the text and the manner in which the constitutional provisions are intended to interact with one another must inform our interpretation, understanding, and application of the text. [citations omitted]
Emmett Macfarlane argues that the architecture language used by the Court in this decision is highly ambiguous, but could be broad enough to block any unilateral electoral reform. The lack of clarity here over the role of provincial consent has led Pal to concludes that its possible that only the general amending formula can be used for electoral reform, and the matter should be referred to the Court on this issue specifically.
Dawood provides 3 ways in which Senate reform is not analogous to electoral reform:
- There are several constitutional references to the Senate, and far less references to the electoral model
- The constitutional language itself is markedly different in reference to Senate selection than it is to any particular electoral model
- The nature of the reforms contemplated in the Senate Reference would fundamentally alter its function, which would not necessarily be true with electoral reform
She cites Yaakov and Jonathan E. Roth to suggest that alternative electoral system would not be permitted under the wording of s. 40 of the Constitution Act. They also suggest that any reform may be prohibited because our preamble makes reference to the UK House of Commons. Dawood responds to the Roth argument by citing Dennis Pilon, who notes that the UK used multi-member ridings and semi-proportional limited vote in 1867. Single member districts were not used universally until 1885.
Pal doesn’t go as far as the Roths, and has indicated that any major electoral reform such as proportional representation (PR) or mixed-member proportional (MMP) systems would likely require the general amending procedure. If the changes only affect the provincial interests in a trivial manner, such as preferential balloting, Pal suggests it could be more strongly constitutional.
Pal finds that the Court’s interpretation of provincial interests in the Senate Reference was very generous, and will readily deem provincial interests to be engaged. This is strengthened by the 2014 Nadon Reference. Even though most of the essential features of the Court are not listed in Part V of the Constitution Act, Parliament can only alter it with provincial consent.
Although Pal agrees that the role and the nature of the electoral process may not be transformed through reform, he points to Opitz v. Wrzesnewskyj and Henry v. Canada to highlight that courts appear to provide some constitutional significant to the connection between Members of Parliament and specific geographic ridings.
Macfarlane cites Leonid Sirota to support the proposition that even with electoral reform, the “nature” of the House of Commons would not be changed, as it would continue its primary role of responsible government.
The most significant case Pal has Figueroa v. Canada, dealing with funding to smaller political parties unable to achieve the threshold of 50 electoral districts for registered party status. The applicant claimed his s. 3 rights under the Charter were violated, and the Court found ss. 24(2), 24(3) and 28(2) of the Canada Elections Act to be unconstitutional.
The FPTP system is referred to by Justice Lebel at para 157 of Figuera as “one of Canada’s core political institutions” which has “the virtue of fostering a strong political centre and reducing factionalism.” However, much of this statement can be considered obiter, given the subsequent qualifiers in the following paragraph, and the context of a majority-concurring decision.
Yet Pal finds the language in Figuera strong enough to resemble the constitutional architecture language found in the Senate Reference. PR or MMP systems would both fragment the traditional party system, and reshape our notions of majority government.
The argument for government making unilateral reforms under s. 44 advanced by Macfarlane is that Parliament has previously used this procedure in 1985 with the Representation Act and in 2011 with the Fair Representation Act. Both of these changed the composition of the House, which arguably had a direct provincial interest. In fact Pal has argued that the latter Act may have been unconstitutional in light of the Senate Reference.
Macfarlane also points to Campbell v. Canada, dealing with this first exercise of power under the Representation Act. The B.C. Court of Appeal affirmed the ability to allow for deviations from absolute proporationality given smaller provinces already enjoy guaranteed representation in relation to their respective populations.
If proportionality was understood in a literal sense, protecting the seats for smaller provinces would require an enormous and unworkable increase of seats in the rest of Canada. John Ibbitson estimates it would require increasing our House to 890 members.
Instead, Macfarlane calls for a contextual reading of s. 44,
Provincial consent for changes to the formula for apportioning seats in the House should thus only be required when a proposed change departs significantly from historical margins or constitutionally-mandated norms.
Further examination of Canadian history also demonstrates that the role of Parliament in electoral reform has hardly been static. The very wording of the Electoral Districts under s. 40 of the Constitution Act is preceded by “Until the Parliament of Canada otherwise provides…”.
Although Pal points to this section as further evidence of an intent to create single-member geographic districts in 1867, Macfarlane points to these spent provisions which were replaced by the Canada Elections Act, Parliament of Canada Act and the Constitution Act, 1982, as further evidence of Parliament’s unilateral s. 44 powers.
He refers to a Supreme Court of Canada case in McKay et al. v. R. and notes the majority’s adoption of the following statement from 1879 by the Court in Valin v. Langlois,
the Parliament of Canada has the exclusive power of legislation over Dominion controverted elections. By the lex Parliamentaria, as well as by the 41st, 91st, and 92nd sections of the British North America Act, this power is as complete as if it was specially and by name contained in the enumeration of the federal powers of section 91, just as promissory notes, Insolvency, &c., are.
The Court of Appeal in Dauost rejected the proposition that effective representation is dependent on the electoral system. The court acknowledged at para 57 that every system has its shortcoming, and referred to expert evidence to that effect. Instead, the appropriate threshold is the function and the degree of representation,
 …in order to be valid, every electoral system must confer on the electorate or assure it of a minimal, albeit significant, degree of representation. [emphasis added]
These statements, originally made in defence of FPTP, can also be used to justify the flexibility afforded to electoral reform and the existing alternatives for FPTP. The law does not assume or presume any single type of electoral system, even if this is what has worked for us – for now. Our effective representation, and our Parliament, is not dependent on any particular system, including FPTP.
This is exactly why it should not be a constitutional issue. Fortunately we live in a democracy which is, for all of its failings, still a functional democracy. If electoral reform can enhance that function, then it really shouldn’t require a constitutional amendment at all.