There is some debate whether a referendum is a good or bad instrument for making public policy choices in a democracy. The question came into sharper focus following the Brexit referendum. The debate has direct relevance in Canada as we consider whether a referendum is required on reform of our electoral system.
Let us begin with the premise that all forms of democracy are flawed. As Churchill said, “No one pretends that democracy is perfect or all-wise.” A referendum is one instrument by which the democratic will can be expressed. Representative democracy is another instrument. Structured community consultation (sortation) is a third, though rarely used.
This discussion begins with a comparison of the referendum instrument with the representative democracy instrument. The discussion continues with an assessment of two uses of the referendum instrument. It concludes with an opinion on which democratic instrument is more suitable to select some reform, or no reform, on the electoral system.
Comparisons between democratic instruments
Certain practical realities apply regardless of the instrument used:
• Voters are often ill-informed. In the 1995 Québec referendum, a substantial number of voters believed that they could retain Canadian passports if Québec were an independent country. In the 1974 federal election campaign, a substantial number of voters were influenced by the front page newspaper image of the Conservative leader dropping a football. Indeed, the whole premise of ‘dog-whistle’ politics is to convince the voter to make a decision on an issue other than the main issue of the campaign. In a democratic system, we accept the right of voters to make their decisions based on whatever reasons they choose.
• Sometimes politicians will do whatever it takes to influence voters. In the 2006 election campaign, the Liberals suggested that a Harper Government would put soldiers on Canadian streets. In the 2015 election campaign, the Conservatives used the ‘niqab issue’ to attack the NDP in Québec. Politicians do not always help voters understand the important issues at play.
• Issues are sometimes divisive. Some public policy choices, particularly those involving moral or religious values, are contentious. Debates about such issues are likely to be acrimonious, regardless whether the debate takes place in parliament or in a referendum. Canada has, in effect, no law regarding abortion, because parliamentarians have determined that such a debate would be too divisive. Parliament did have a debate to outlaw capital punishment in 1976, and considered the issue again in 1987; both debates were difficult, with those on the losing side (i.e., in favour of capital punishment) suggesting that the matter would be better settled by referendum. Even the recent legislation on assisted dying raised passions in the house, provoking our pugilistic Prime Minister into grabbing and elbowing to force a vote. That issues are divisive is not a valid argument for avoiding discussion on the issue.
• Accountability is often blurred. Some argue that voters in a referendum are not accountable for their decision, whereas legislators can be held accountable for their decisions. It is fair to say that voters cannot be held accountable by someone else for a referendum decision, but since they must live with the result voters are accountable to themselves. Californians voted against taxes in various Propositions, and have lived for decades with poor services as a result. As to whether legislators can be held accountable, in practice this seems dubious. Voter decisions in elections are made for a number of reasons, some based on policies, others based on personalities. Liberal governments in Ontario were re-elected despite breaking a 2003 promise not to raise taxes (and imposing a health tax) and despite a long-running scandal on cancellation of gas power plants. The fact is that parliamentary governments often break promises and are often not held accountable. Accountability, or the lack of it, is not a clear basis upon which to distinguish referenda from representative democracy.
Notwithstanding these differences, one can distinguish issues or questions where one would prefer to use one instrument rather than the other. As examples:
• Hundreds of decisions are required in democratic governance. Representative democracy allows voters to delegate responsibility for most of those decisions, with periodic votes in judgement of how well the representatives have done or are likely to do. That said, there is insufficient time in parliamentary government to give adequate consideration to all issues. Omnibus bills are sometimes used to group policy decisions together, not always on related matters. Time allocation and closure are often used to limit debate.
• Many decisions are complex and involve public policy trade-offs: fiscal policies, trade policies, cultural policies etc. Representative democracy allows legislators to familiarize themselves (not that they necessarily do so) with the intricacies of public policy in a way beyond the realistic capacity of voters who have their own familial and work responsibilities.
• Parliament often grants the Executive authority to make administrative decisions – which can themselves be substantive (e.g., foreign investment approvals, telecommunication regulation).
• However, some decisions avail themselves of clear ‘Yes’ or ‘No’ answers. Such questions often pertain to moral questions, rights issues or constitutional matters. Referenda might well be appropriate here, subject to certain conditions:
There should be some indication that the people want a voice in the matter.
There needs to be legal rules under which the question is to be decided: e.g., length of campaign, campaign spending limits, who has the right to vote, etc.
There must be a clear understanding of what the result means: e.g., whether binding on the executive or the Parliament.
There can be no referendum which would result in a change to Charter rights or basic legal rights (habeas corpus, etc.).
Two Uses of the Referendum Instrument
1. In 1995, the government of Québec proposed the question:
Do you agree that Quebec should become sovereign after having made a formal offer to Canada for a new economic and political partnership within the scope of the bill respecting the future of Quebec and of the agreement signed on June 12, 1995?
It is immediately apparent that this is a complex question, and not only due to its length.
a. It can be argued that the intent of the question is to mislead. Initially it appears that any mandate to declare Québec sovereignty would be contingent on another event; it implies an accord with Canada on the terms of sovereignty association. However, no accord is required: only an offer is required. In the event that Canada refused to negotiate on the offer, the wording of the question would have allowed a proclamation of sovereignty. Indeed, the leader of the Parti Québecois, M. Parizeau, stated that this was his intention. Moreover, the reference to a signed agreement implied some existing participation by the federal government, which was simply not the case: the agreement was between certain political leaders in Québec
b. However, even if the offered contingency was false, it was necessary for the voter to understand the offer that would be made, which would have required the voter to understand a bill under consideration in the National Assembly and an accord signed between three Québec political leaders. Both documents have multiple sections.
This should not have been a question voted on a referendum, as there is little doubt that voters could have had a clear understanding of what a ‘Yes’ vote would mean.
Following the referendum, the federal government passed the Clarity Act, which declared that the federal House of Commons had the right to determine whether any sovereignty referendum question was sufficiently clear. A contingent question, such as that used in 1995, was specifically ruled out.
2. In 2016, the Government of the United Kingdom posed the question
“Should the United Kingdom remain a member of the European Union or leave the European Union?”
This is a more simple question than the Québec question, not least because it is less half the length of the Québec question. Indeed, the original draft of the question was even shorter: “Should the United Kingdom remain a member of the European Union?” That original draft was found to be biased by the electoral commission in the UK, as it did not express the balance between the ‘Remain’ and ‘Leave’ options. Even with the improvement, the question is problematic:
a. The question presupposed that voters have some idea: what it means to be part of the European Union; and what it would mean to leave the European Union. The lack of awareness among most voters opened the door to irresponsible claims by campaigners. However, even if the debate had been conducted in a responsible fashion, the issue would have continued to be at a level of complexity beyond all but the most engaged voters.
b. There was no clear understanding what the ‘exit’ result would mean. Indeed, after the vote, one of the Brexit campaigners, Boris Johnson, argued that having left the European Union ”there will continue to be free trade, and access to the single market”; such a result might be possible, but appears more like the same magical thinking that had some Québecois believing they would retain a Canadian passport in a sovereign Québec.
The Brexit referendum question appears less problematic than the 1995 Québec referendum question, but asked voters to decide on a matter about which they understood relatively little.
Considerations on the electoral reform decision
Decisions about how our electoral system works are qualitatively different from any other public policy decisions taken by elected representatives. The argument for this position follows:
Our political system of representative democracy requires an electoral system to elect the individuals who will represent us in the legislature.
What makes the political system democratic is the right of every individual (over a certain age) to vote for a representative.
How that vote is applied to determine the representative is thus at the core of representative democracy.
The representative is the result of the vote: the vote is the cause; the representative is the effect.
Voting is thus logically, politically and morally prior to the elected representative and any actions that representative might take.
Therefore, the legislature (made up of our representatives) does not have the moral right, and should not claim the political right, to change the rules under which our representatives are elected.
Only voters have the right to change the electoral rules; the mechanism is a referendum.
That granted, there are significant challenges to establishing a fair and meaningful referendum on electoral reform. Chief among these is the difficulty in explaining what an alternative mechanism means and how it would work. Herein lies a major advantage for the existing first-past-the-post (FPtP) system. Whatever its flaws, voters understand what happens to their vote under FPtP. They can see the result when ballots are counted and know what happened to their vote in their constituency. In order to make a fair referendum question, voters would need to understand the workings of alternate systems, such as ‘single transferable vote’ (STV), ‘proportional representation’ (PR), ‘mixed-member proportional’ (MMR), etc.
Although changes to the electoral system were proposed by provincial citizen’s assemblies, in British Columbia (2009) and in Ontario (2007) referenda results on whether to change the electoral system showed strong support for the existing FPtP system. Lack of clarity about the workings of the proposed alternate systems, plus a lack of dissatisfaction with the status quo, were probable factors in the defeat of the possible changes.
Given that history, one can understand the reluctance of the current federal government to propose a referendum on changes to the electoral system. However, the probability of defeat is not an argument against a referendum; it is on fact an argument for a referendum.
The government may have identified electoral reform as a priority, but it is by no means assured that this priority is shared by the majority of voters. What is first needed is a significant national consensus on the need for change. It is also absolutely essential that voters trust the process by which change is implemented. Lack of trust in this matter does irreparable damage to the democratic ideal.
Some have argued that Parliament has a better capacity than individual voters to identify and implement reforms to the electoral system. These arguments presume a Panglossian perspective on Parliament. Indeed, serious and careful consideration by Parliament is far from a certainty. Parliament has often given short shrift to due process and careful consideration: witness the use of omnibus bills, of time allocation, of closure and, indeed, of physical intimidation by the Prime Minister on the floor of the House of Commons.
We already know that current or proposed election systems appear to benefit one political party or another. Conflicts of interest abound. Citizens would rightly object if one political party tried to set the rules to benefit itself, which is why we watch for the Gerrymandering of riding boundaries. Citizens would rightly object if one or more political parties tried to do the equivalent malevolent practice in setting electoral boundaries. I doubt we can presuppose that citizens have the requisite trust in legislators to design a system without favouritism or some manipulative technique. And without that trust, any electoral reform loses its moral and political legitimacy.
Is it possible to reconsider and perhaps change the electoral system while retaining the confidence of voters in the integrity of the democratic system? One proposal worth consideration would be the establishment of up to two referenda:
• The first referendum would pose a question something like: “Should Canada consider replacing the current first-past-the-post electoral system?” Such a question would ascertain whether Canadians see the need for electoral reform in the first place, but would not commit them to any specific alternative. If more than 50% of voters said they wished to consider alternate electoral systems, then:
• The second referendum would offer a menu of the principal choices for electoral reform: e.g., FPtP, PR, STV, MMP. The menu of choices could be developed by Parliament, Elections Canada or an external commission; it is in the interest of those seeking change to limit the number of alternative choices. It is important to note that the existing FPtP system remains on the ballot as, although voters may not be entirely happy with it, it may still be more popular than any of the alternatives. Consideration is required as to what threshold of voter support would be required to institute a new electoral system. If no alternative emerged above the threshold the existing system would remain.
The use of two referenda has a precedent in our constitutional history. The entry of Newfoundland into confederation also required two questions, as there were initially up to four alternatives available.
1. Both referenda and legislative decisions have significant flaws as determinants of the democratic will. However, distinctions between the two methods give a guide for the circumstances under which one method might be more appropriate than another.
2. Referenda recognize the primacy of voter rights. But referenda organization and questions must enable informed voter decisions.