First of all, thank you for the invitation. It's always nice to come to Ottawa when the weather is nice.
I teach political philosophy and war, not political parties and elections, so my remarks will reflect this approach to political reality.
Changing the electoral system changes one of the fundamental attributes of the regime, which in Canada we usually refer to as responsible government. Because fundamentals are involved, I'll have to commit a little political science, for which I apologize in advance.
First of all, changing fundamentals means you can't simply change the electoral system and everything else stays the same. A lot of things will change if we move from the majority-plurality system we have today to one or another form of proportional representation, or PR, as I'm sure you've heard it referred to. The kind of PR matters as well, but I won't go into that. The important thing is this: changing the electoral system is not just applying a new coat of paint.
Second, politics involves forming coalitions—of interests, of ideologies, of personal patronage, whatever—in order to govern. This is as true for chimps and bonobos as it is for human beings.
Parties part people in the sense of dividing them, but they also bring them together as they search for winning coalitions. Whatever the electoral system, whether plurality or PR, parties exist to form governments; that is their rational purpose. It is not just to promote interests and ideologies. Interests and ideologies are often promoted by organizations other than parties and are, within parties, often subordinate to forming a government, but here matters get more complicated, because different electoral systems incentivize people to form different kinds of coalitions.
Specifically, plurality systems incentivize people to form coalitions within a party to win a parliamentary majority. Brian Mulroney famously did so by forming a coalition of Quebec nationalists and westerners. Under PR, small parties, which may well represent ideologies or interests—or in Canada, a region—have no possibility and no ambition to form a government. They want to be part of a larger coalition in order to advance their ideology or interests that way. Governing with a PR system still involves forming a coalition, but now parties form them openly in parliament rather than within a big-tent party. That is, all PR systems incentivize persons to create single-issue parties to run on comparatively narrow agendas and do the coalition-building after the election.
None of this is news. I first learned about the effects of electoral systems as an undergraduate at UBC during the 1960s. The material we were reading was even older.
I have one last bit of political science. We all know that majority rule in free elections is the basis of democracy, but when there are more than two choices, there may be no majority for any particular rank order of choices. This “voters' paradox,” as it's called, is the electoral equivalent of “rock, paper, scissors”. More formally, it is the basis of Kenneth Arrow's impossibility theorem, which he described in 1951 and which helped secure him the Nobel Prize in economics 20 years later. The implication of this very complex argument is significant: there is no best electoral system.
To repeat, there are different incentives provided by different electoral systems, and these incentives are distinct from questions of establishing the franchise, homogeneity of electoral districts or constituencies, and what Canadian courts call “communities of interest”.
Let's look again at the consequences that a PR incentive system extends to a political party that is more concerned to advance a particular interest, ideology, or agenda than in ruling.
First, it leads to a proliferation of small parties. Even in a plurality system, the disintegration of the big-tent Mulroney coalition resulted in two little parties, Reform and the PQ. It took a decade for Stephen Harper to recreate at least part of the coalition in the Conservative Party of Canada. Outside Canada, between 2000 and 2015, 17% of PR elections resulted in single-party majorities. In contrast, 85% of majoritarian or plurality elections resulted in single-party majorities.
So what? That seems to be a pretty benign consequence.
The problem, however, is that when coalitions are formed in Parliament rather than within the party, large majority parties have to make concessions to small ones, including fringe parties. This is reason enough for small, intense, single-issue parties to favour PR.
The logic is obvious. If small-party ideologies or interests are supported by most electors, they become big parties. If not, they are in a position to leverage their small but intense support in exchange for supporting a big party in Parliament and getting the big party to legislate what they want—but notice that they do so against the wishes of the majority.
Democratic theory is rightly concerned with the tyranny of the majority. PR practically invites tyranny of the minority or minorities. In short, PR does not encourage the foremost political virtue, namely moderation, to say nothing of institutional stability.
The most obvious practical result is that PR elections lead to increased government spending as large parties acquiesce in the requests or demands of small ones in exchange for their support. One study has shown that increases in government expenditures are in the order of 25%. Moreover, PR countries tend to cover their increased expenditures by borrowing money, thus increasing debt. The generalized effect, therefore, is to increase the size of government, which increases the effective power of bureaucrats, and bureaucrats are not elected by anybody.
I should say in passing that the most obvious forgone benefit of instituting a PR system is that it becomes much more difficult to vote the government out of office. This practical advantage of plurality systems was clearly in evidence in the 2015 federal election and in the 2016 Alberta provincial election.
In that connection, so far as changing the federal electoral system is concerned, winning 39% of the popular vote does not constitute a mandate, especially when you ask how many of those who voted for the federal Liberals did so because that party promised to change the electoral system but didn't say how. I would suggest the answer is “precious few”, which leads to a final practical issue: the growing suspicion of, not to say cynicism with respect to, the motivations of government. Like all parties, the Liberals are rational actors and so will likely design an electoral system from which they expect to benefit. Whether they do is, of course, another matter.
A couple of lawyers wrote in the Toronto Star not too long ago, which usually supports the present government, “To allow a one-off parliamentary majority to unilaterally alter” the foundations for distributing political power “would be fraught”. Madison said something quite similar in “Federalist No. 10”, and so did de Tocqueville and many other democratic theorists.
Let me conclude by reiterating my first observation, that changing the electoral system is a change in fundamentals. In legal language it amounts to changing a constitutional convention, or what we now call the constitutional architecture. I need hardly remind you that constitutional conventions are the customs, practices, and maxims that are not enforced by courts but nevertheless constitute a practical political ethics. We were recently reminded of their importance by the spectacle of Senator Duffy.
What is more important is that we might anticipate a court challenge from one or more of the provinces on the grounds that changing the electoral system violates a constitutional convention that has been in place since 1791. It clearly changes the internal architecture of the Constitution, which invites scrutiny by the Supreme Court of Canada.
Changing the electoral system obviously affects provincial interests. Just think of the constitutional requirement of four MPs for P.E.I. If that constitutional requirement were carried forward, one of my colleagues at the University of Calgary calculated that the House would contain over 600 MPs, which implies another kind of architectural change to the configuration of the chamber.
More to the point, the decisions by the court in the Nadon reference and the Senate reform reference of 2014, and going back to the patriation reference of 1980, are pretty good indications that the government would lose.
In short, thinking about a PR system for Canada is fine for political science undergraduates in a bar on a Friday evening. For Parliament seriously to consider this constitutionally suspect change is politically imprudent, to use no stronger language. A lot can go wrong and likely will.