No Solution

The reasons people don’t vote suggest a mandatory voting law would be futile

Statistics Canada has released the results of a survey, conducted in conjunction with the November 2015 Labour Force Survey, to inquire into Canadians’ “Reasons for not voting in the federal election, October 19, 2015.” These results are interesting, albeit not quite accurate. To my mind at least, they are further evidence for the proposition that mandatory voting is not the solution to what ails Canadian democracy.

A word, first, about accuracy. Elections Canada put turnout in the last election at 68.49%, though that doesn’t include voters who registered on Election Day. Adding the number of voting day registrations from the Elections Canada “Report on the 42nd general election of October 19, 2015,” I get to a turnout of 70.4% ― though the report itself actually gives a lower figure, 68%. Anyway, these discrepancies don’t matter for my present purposes. What does is that according to Stats Can, 77% of Canadians “reported that they had voted in the 2015 federal election.” So unless the survey’s sample was unrepresentative (which is unlikely though not impossible), anywhere between 7 and 9% of the respondents lied about having voted.

Of course, this suggests that, for these people anyway, voting already is a duty ― albeit one that they might shirk. Which way this cuts in the debate about mandatory voting, I’m not sure. On the one hand, these people don’t need the law to tell them that there is a duty to vote ― they already believe there is one. On the other, the probably need a relatively small nudge to act on their belief, so a mandatory voting law setting a small penalty for not showing up might be effective at getting them involved in the political process.

What about the quarter of the population who actually admit to not voting? Stats Can has a detailed breakdown of their reasons for not voting. Almost a third say they are not interested in politics. Would the threat of a (small) penalty get them to the polls? Quite possibly, though surely not all of them. But what would they do once they get there? The threat of a fine won’t make them develop an interest that they now lack. At best, they will be honest enough to spoil their ballots. At worst, they will cast reluctant, uninformed votes, which will surely not improve our democratic process.

A relatively small number ― only 7% of the non-voters abstained for “political reasons” other than a lack of interest. I suspect that most of these were people who ― like me ― did not find a candidate or party to their liking. If voting were mandatory, most of us would presumably spoil our ballots (or vote “none of the above” if that’s an option). As I’ve observed here, a spoiled ballot doesn’t really add anything to the democratic process either, and even a small number of “none of the above” votes (7% of 23% is just over 1.5% of the electorate) would not be taken as a serious message by the political actors).

Almost half of the self-confessed non-voters invoked what Stats Can terms “everyday life or health reasons” ― being too busy (almost a quarter of the abstainers), being out of town, or being ill. Quite a few of them, especially though surely not only those who were ill, would be excused under a mandatory voting regime, after an inquisition into their circumstances ― which doesn’t strike me as something that the state should be engaging in, but I suppose the defenders of mandatory voting see things differently. Others, those who consider themselves too busy, may well regard a small fine as a cost worth incurring. Even if the fine does tip their utility calculus in favour of voting, it is difficult to imagine that they would be willing to expend the much more substantial amount of time and effort it would take for them to become reasonably informed about the issues. They would show up at the polls and, like those uninterested in politics (whom, I suspect, they resemble more than they care to admit), cast an uninformed ballot.

Finally, 8% of non-voters said that they stayed home for “electoral-process related reasons” ― such as inability to prove their entitlement to vote, or to get to the polling station, excessively long lines, or lack or information about the process. I find it difficult to believe that the threat of a fine would change anything to situation of these people, most of whom would anyway be excused.

Making voting mandatory will not improve our democracy. It will not make people who cannot be bothered to take the political process serious invest their time in it. While it will doubtless force some ― though not all ― of them to the polls, they will not be good voters, whatever one’s definition of “good” in this context. Nor will mandatory voting make those who simply don’t like the options on offer change their mind. And it will certainly not cure the sick or provide identification or transportation to those voters who lack one or the other. Even assuming for the sake of argument that abstention is a problem, mandatory voting is not a useful solution.

De la formation du gouvernement

Juste avant les élections fédérales en Octobre, j’avais participé (en compagnie de Hoi Kong) à une mini-conférence à l’Université de Montréal, intitulée « Gouvernements minoritaires et/ou de coalition : Legality and/or Legitimacy ». La chose m’avait échappé à l’époque, mais ma présentation est disponible en ligne. Évidemment, elle ne s’est pas avérée très pertinente vu le résultat du vote du 19 octobre, mais elle pourrait l’être dans quelques années, d’autant plus si le système électoral est modifié entre temps. La voici.

Let Them Vote

I have a new post up at the CBA National Magazine’s blog, arguing that, with one significant qualification, a private member’s bill that would lower the voting age at federal elections to 16 is a good idea and should be enacted. I have already made the case for lowering the voting age, to 16 if not lower still, here and here. So I am happy to see that an MP, Don Davies, has taken up this cause ― and I hope that the government endorses it too, which would make the passage of the bill much more likely.

The one reservation I have about Mr. Davies’ bill as it now stands is that it makes no separate provision for, and indeed no mention of, a minimum age for running for Parliament. As I explain in the National Magazine post, under the Canada Elections Act, almost all eligible voters are allowed to be candidates. But it is not obvious that the minimum age for being an MP and for voting should be the same. At the very least, I think the issue deserves to be debated.

Subject to that, I wish Mr. Davies’ good luck with his bill. Its enactment would make our democracy more inclusive, and thus better.

The Core of the Case against Electoral Reform

Why the concerns of those who want electoral reform do not move me

When I wrote about the constitutionality of electoral reform, I did not want to discuss its merits, beyond saying that I did not believe it had many, or the process by which it should be determine upon, beyond saying I favoured a referendum. While I thought it important to mention my positions in order to avoid my constitutional opinion being interpreted as self-serving, I did not mean to argue for them, not least because I did not think I had any arguments to make that would be very interesting or original. Prompted by some of the comments I received, I have thought about these arguments again, and come to the conclusion that one of them may be of some interest, because it is seldom if ever made.

The two main problems that electoral reform is usually said to address are those of “disproportionality” and “wasted votes.” Disproportionality refers to the discrepancy between the percentage of the votes received by a party and its percentage of seats in Parliament, any such discrepancy being, in the view of those who make this argument, undemocratic. Wasted votes are those that apparently do not serve to elect anyone ― that is, under the first-past-the-post system that we now use, those cast for candidates who do not win the riding in which the votes are cast. Proposed reforms ― ranked ballot and mixed-member proportional representation are the main options, if I understand correctly ― address these issues in different ways and to different degrees, but I take it that those are who support reform are usually motivated by these concerns. (Or at least, those who support reform on grounds of principle rather than partisan interest.) There are other arguments for electoral reform, of course, such as that ranked ballot is supposed to make for kinder, gentler politics (like in Australia, I guess), or increased voter turnout (like in Switzerland, I suppose) but they strike me as speculative at best.

Here’s why concerns about disproportionality and wasted votes do not move me. Imagine a simple election between two candidates ― like the run-off stage in a French presidential election. Say the winner gets 55% of the vote, and the loser gets 45%. But the winner gets the prize, the French presidency for example, 100% of it, and the people who voted for the loser go home and their votes have no apparent effect at all. On the account of our would-be electoral reformers, this election results in substantial disproportionality, and a large fraction of the votes cast in it is wasted. Its democratic character should, therefore, be regarded as very questionable. Except that, of course, nobody actually thinks that. Similarly, nobody thinks it undemocratic that a party with, say, 55% of the seats in Parliament gets to govern 100% of the time while that Parliament exists, and nobody seems to say that the votes of the opposition parties are “wasted.”  The disproportion between 55% of the seats and 100% of the power is actually more significant, both in sheer numbers and, more importantly, in effect, than, say, that between getting 40% of the vote and 55% of Parliamentary seats. Yet it is only the latter that is said to be somehow undemocratic. I just don’t see how that is the case. The complaints of the proponents of electoral reform sound in principle, but they strike me as selective and inconsistent with what they are happy to accept in other situations.

For my part, I am no more bothered by the disproportionality or wasted votes in the elections for Parliament than by the same “problems” that arise within Parliament itself. What concerns me is how to limit the ability of a Parliamentary majority, however constituted, to harm both the opposition in Parliament and the citizenry at large ― including its supporters of course ― while it is in office, and how to prevent that majority from entrenching itself so as to become impossible to remove from office at the next election. Electoral reform is not a solution to these concerns. (Indeed some version of it may ― though this too is a speculative argument ― make the entrenchment problem worse.) It is therefore, in my view, an unnecessary disruption of our politics, and a distraction from the issues on which we should be focusing.

NOTE: Feel free to comment, but I doubt I will respond unless I think your observations respond directly to the point I’m making.

Yes, They Can

Constitutional amendment with provincial consent is not required for electoral reform.

The federal government’s pursuit of electoral reform has raised a number of questions: is reform necessary or desirable at all? If so, what new electoral system to adopt? Should the people be consulted before reform is implemented? And now, thanks to op-eds published by Michael Pal in The Globe and Mail, and Yaakov and Jonathan Roth in The Toronto Star, we also have to ask ourselves whether it would be constitutional. In my view, however, the answer to the latter question is much clearer than to the others. The constitution is no obstacle to Parliament acting alone to implement (most forms of) electoral reform. (By the way, in case you think that my opinion on this is relevant to assessing the argument that follows: I think that electoral reform is a bad idea, and I think that if the government insists on endorsing it, there should be a referendum before reform is implemented.)

Section 44 of the Constitution Act, 1982 ― which prof. Pal describes as an “obscure provision,” and Messrs. Roth do not mention at all ― provides that “exclusively make laws amending the Constitution of Canada in relation to the executive government of Canada or the Senate and House of Commons, subject to sections 41 and 42. As a starting point, it would seem logical to consider electoral reform an amendment to “the Constitution of Canada in relation to … the House of Commons,” and thus within the purview of Parliament, except insofar as sections 41 and 42, to which I will return, provide otherwise.

Those who think unilateral electoral reform would be unconstitutional point to the Supreme Court’s opinion in Reference re Senate Reform, 2014 SCC 32, [2014] 1 S.C.R. 704, which introduced the notion of “constitutional architecture” that limits Parliament’s amending power under section 44. The “architecture,” which seems to consist of “assumptions that underlie the text [of the constitution] and the manner in which the constitutional provisions are intended to interact with one another,” is an entrenched part of the constitution, and cannot be amended by Parliament acting alone.

Prof. Pal suggests that the first-past-the-post electoral system is part of that architecture since, although it “is not mentioned directly in the Constitution, … [n]umerous parts of the Constitution presume that [it] is in place.” Moreover, in his view, electoral reform “would affect provincial interests,” although “[l]esser changes than a move to proportional representation could be interpreted as” doing so “only trivially … and not really changing the constitutional architecture.” For their part, Messrs. Roth insist that the Constitution Act, 1867 “expressly assigned to each province a fixed number of ‘electoral districts,’ each entitled to return ‘one member’ to the House. This ‘constitutional architecture’ plainly presupposes district-based elections.” They also point to the use of first-past-the-post in the United Kingdom, to whose constitution ours was intended, according to its preamble, to be “similar in principle.”

But it is not enough, it seems to me, to say that first-past-the-post is how we always elected representatives to show that it is a part of the “constitutional architecture” as the Supreme Court understood that notion in the Senate Reform Reference. In that opinion, the Court said that “the institutions provided for in the Constitution” ― such as the House of Commons ― “can be … changed to some extent under ss. 44 and 45, provided that their fundamental nature and role remain intact.” [48] That, in my view, is what electoral reform would do. It might change the House of Commons to some extent (though to what extent would depend on the shape the reform takes), but would not affect its “nature” as the representative part of our national legislature or its role of serving as the electoral college for the choice of a Prime Minister, making laws,  and pretending to hold government to account.

Prof. Pal never says, unfortunately, what parts of the constitution rely on first-past-the-post, and for my part, I am unable to figure out what they are. It is noteworthy, for instance, that section 3 of the Canadian Charter of Rights and Freedoms provides that “[e]very citizen of Canada has the right to vote in an election of members of the House of Commons” (emphasis mine), as opposed to, say, “the election of a member.” The latter formulation would presuppose election in single-member districts. The former does not, since an election under a system of proportional representation is still “an elections of members of the House of Commons.” As for the provisions assigning a specified number of districts to each province, which Messrs. Roth invoke, these were obviously intended to be, and have been, amended by Parliament acting alone since 1867, as the number of districts and members of Parliament was increased.

Note, by the way, that contrary to what Messrs. Roth say, one or two of the districts created by section 40 of the Constitution Act, 1867 were actually entitled to two representatives in the House of Commons. Indeed, multi-member districts were common in the United Kingdom in 1867 ― most English Members of Parliament represented counties or boroughs that returned two members each, and some counties had three representatives. There were even exceptions to the principle of geographical representation (as well as the one man, one vote principle), in the shape of university constituencies that allowed the holders of some degrees from some universities to elect additional representatives for their almae matres. Quite apart from the fact that the legal effect of the preamble to the Constitution Act, 1867 is matter of doubt on which the Supreme Court has wavered over the years, the claim that geographical constituencies electing single members of Parliament using a first-past-the-post system was a matter of long-standing fundamental constitutional principle in the United Kingdom in 1867 is historically inaccurate.

Thus I am not at all persuaded that the constitution’s entrenched text in any way depends on or implies the first-past-the-post voting system. If anything, I suspect ― though I have not done the historical research to prove it ― that section 3 of the Charter might have been written specifically to avoid entrenching this arrangement. Nor do I think that electoral reform would impermissibly affect the constitution’s architecture. While changes in the relationship between Parliament and the executive are conceivable in the wake of a move to some form of proportional representation, they are unlikely to be fundamental in nature. As for changes to the relationship between Parliament and the provinces, I do not understand how any could result ― except in the one case which is also ruled out by the constitutional text.

The exception to Parliament’s general ability to enact electoral reform under section 44 of the Constitution Act, 1982 concerns reform plans that would sever the relationship between members of Parliament and provinces from which they are elected ― in other words, those versions of proportional representation that would distribute seats on the basis of national, rather than provincial vote totals. The trouble for such systems lies in the entrenchment, in section 41(e) of the Constitution Act, 1982 of “the right of a province to a number of members in the House of Commons not less than the number of Senators by which the province [was] entitled to be represented” in 1982, and in section 42(a), of “the principle of proportionate representation of the provinces in the House of Commons prescribed by the Constitution of Canada” (emphasis mine in both cases). It is thus the provinces ― though not any territorial subdivisions within the provinces ― that form the basis of representation in the House of Commons, and that principle is indeed part not only of the constitution’s architecture, but of its very text.

Provided that it respects this principle, however, Parliament is constitutionally free to change voting arrangements by ordinary law enacted under section 44. Such changes would, no doubt, be of great political significance. But while that may be (I think it is) an important argument in favour of giving the people a say over electoral reform, it is not, in itself, a reason to consider that constitutional amendment with provincial consent is necessary to effect such changes. The Senate Reform Reference does not hold the contrary. The question of electoral reform’s constitutionality is, I believe, a distraction from those about its desirability and the process by which its desirability ought to be determined.

A Civic Choice

This is the last substantive post in my duty-to-vote series. I have already addressed a number of instrumental arguments in favour of such a duty: claims that it allows better aggregation of information about the voters’ preferences, that it enhances the legitimacy of our political system, and that it improves the quality of election campaigns. In this post, I address a different type of argument: that we must vote not because universal voting serves some other purpose, but just because it is a “civic duty.” One cannot, it is said, be a good citizen if one does not vote.

Andrew Coyne, although he also makes a number of instrumental arguments in favour of a (legally-enforced) duty to vote, invokes this idea of civic duty when he asks, presumably rhetorically: “[w]hy should voting, the fundamental act of democracy, be an option, and not, like jury duty or paying your taxes, a basic obligation of citizenship?” When I first announced this series of posts, Craig Forcese responded (on Tiwtter) that “Civics, like reading, is [a] muscle that atrophies [without] regular use,” and further that “[v]oting [is a] collaborative civics ritual in an atomized society [with] very few” of those. It is, he said, “[a]s much about membership as governance.”

I have to admit that I am somewhat perplexed by the idea of a “civic duty” that exists for no particular reason, just as an incidence of membership in society. Mr. Coyne’s examples of jury duty and taxes can be justified (if indeed they can be), on instrumental grounds. Jury trial is (so we think) a bulwark of liberty, while taxes are needed to keep government running and to help the poor or those otherwise in need of their fellow-citizens’ assistance. But instrumental justifications for a duty to vote, I have argued, do not succeed.

But let’s put that doubt to one side, and let’s stipulate that we can have some duties as a result of our membership in society, regardless of whether fulfilling these duties actually serves any useful purpose. And let’s stipulate that one such duty is to take public affairs seriously, to concern yourself with the way your society is governed, and to share this concern with your fellow-citizens. I’m actually very skeptical that we have such a duty. It seems to be, at best, an instantiation of a broader, and more plausible, duty to contribute to society ― but as Jason Brennan argues in his book on The Ethics of Voting and in a post over at Bleeding Heart Libertarians, we have any number of ways to contribute to society, not all of them having anything to do with politics or public affairs writ large:

For any given citizen, given what other citizens are doing and are good at doing, there will be an optimal mix of political and non-political ways for her to pay her debt [to society, assuming that there is one]. For some citizens, this will mean heavy political engagement at the expense of other pursuits. For other citizens, it will mean complete disengagement so as to free the citizen to pursue non-political activities. For most citizens, the optimal mix will be some combination of political and non-political engagement.

(I’m not sure, by the way, that prof. Brennan is even correct to say that “for most citizens” political engagement will be part of the “optimal mix.” The average levels of political ignorance are so high that “most citizens” may be doing more harm than good by becoming involved in politics in any way. But let’s put that to one side too.)

Anyway, let’s stipulate that we have a duty to become politically engaged together with our fellow citizens. Does it follow that we have a duty to vote? I do not think so. To be sure, voting is a way to fulfill this hypothetical duty, but it is not the only one. Surely, debating public affairs, whether just with your friends on Facebook or in some more public forum, is a form of political engagement. Surely, working for some organization that contributes to the public good, as it sees it, is a way of taking part in the polity’s affairs. I would, indeed, go further, and say that such ways of becoming engaged are actually much more significant that voting. I am pretty sure that I have contributed a good deal more to the res publica, over the last three and a half years, with this blog than I would by casting ― or, a fortiori, by spoiling ― a ballot next Monday. Perhaps we have, as Sean Hunt put it to me on Twitter, a “duty to consider” our options. But then what? If, having considered the choices on offer, you find none of them palatable, I do not think that you do anything wrong by staying home.

For those who, like prof. Forcese, worry that the civic instincts of the abstainers will atrophy without a quadrennial exercise in walking to the polling station, I think that a “duty to consider” or a duty to be engaged in public affairs should be enough. If you follow politics and think about it, you will surely not fail to vote if or when you finally see a party that actually deserves your support, or perhaps even one that is so much worse than the others that you vote strategically against it. It is interesting, I think, that a recent poll found that people who think that voting is a choice are not much less likely to vote as those who think that it is a duty. Among the former, only 5% said they would not vote, while 11% are undecided as to whom they will support. Among the latter it was 0 and 5%, respectively. (26) The absolute numbers are probably lower than they are in reality: overall, 72% of those who were eligible to vote in 2011 said they voted, while the true turnout rate was closer to 60% ― people lie to pollsters (and I wonder whether the purportedly duty-bound do not lie more than those who allow themselves the choice). But in any event, it’s not those who think that voting is a choice who fail to vote in large numbers: it’s those who “don’t know” whether it’s a choice or a duty. Among them, 19% say they will not vote, and 31% are undecided.

In short, voting is neither necessary to promote some ulterior good, nor in itself a duty. It is a right which, as I pointed out in this earlier post discussing the claim that we ought to vote out of gratitude to those who helped secure and defend our right to do so, like all other rights, we can choose to exercise or not. This choice should not be made lightly, but it can, and should, be made freely.

It Won’t Help

This is yet another post on the duty to vote. Here, I address arguments according to we have such a duty because if everyone votes, the quality of election campaigns and, possibly, of governance, will be better than under the current state of affairs, where some people vote, and others do not. This argument, like the information- and legitimacy-based ones that I addressed previously, is instrumental, in that it sees universal voting not as an end in itself, but as a means to achieve another desirable result. And indeed this result ― campaigns and governance (is there even a difference between the two anymore? was there ever?) that are oriented toward the public good rather than to pandering to specific segments of the population ― is a most desirable one. But can it justify a duty to cast a ballot?

Susan Delacourt has argued that if voting were mandatory, people tempted to engage in electoral chicanery destined to depress turnout among their opponents would refrain from doing so. Presumably, the same thing would be true even in the absence of a legal mandate, if they knew that everyone believed voting is obligatory. I have also seen suggestions that employers would more readily allow employees time to vote on election day. Yet it is already illegal to prevent people from voting, and employers already have a legal obligation to allow their workers three consecutive hours in which to cast their ballot. I do not understand how the existence of a duty to vote, whether legal or moral, would change the calculus of individuals already prepared to break the law.

Ms. Delacourt also worries that “[s]ome of the dumbing-down of discourse” that plagues our elections “has taken place because political campaigns have become preoccupied with simply getting out the vote (often with shiny baubles) rather than a debate of ideas.” Andrew Coyne has expressed the same concern, writing that “[a]ll of the parties would be … happy if voters outside their base got lost on the way to the polling station. Elections today are not about reaching out to uncommitted voters, so much as motivating yours — typically out of fear of theirs.” (Emphasis in the original)

Yet, tellingly, neither Mr. Coyne nor Ms. Delacourt has offered any evidence that the politics of nations with mandatory voting are more high-minded, or less prone to “the dumbing-down of discourse” or resort to wedge issues than ours. It is from Australia, the country that inspires the defenders of a duty to vote, that the Conservatives have imported Lynton Crosby, the purported guru of wedge issue campaigns ― though, in fairness, the use of wedge politics responds to much broader trends, which I briefly described here, and which affect all democracies, regardless of whether they make voting mandatory. The Australian habit of changing Prime Ministers every year, or perhaps after every poll, also does not strike me as a sign of good governance. There is, in short, little reason to believe that a duty to vote is a panacea, or even a moderately useful remedy, for what ails our politics.

On the contrary, Ilya Somin points out, a duty to vote might make things worse. The people who tend to abstain when voting is voluntary are on average more ignorant than those who vote and, as Jason Brennan notes, empirical studies suggest that even forcing people to vote “doesn’t cause uninformed voters to become any better informed.” An electorate enlarged by the existence of a duty to vote is thus a more ignorant one. The trouble is, as prof. Somin explains, that “[f]or fairly obvious reasons, relatively ignorant voters are more likely to be influenced by simplistic 30 second ads than relatively well-informed ones (who, among other things, tend to have stronger preexisting views).” After all, the problem for the parties does not end with getting voters into a polling booth. Even if the voters get there on their own, they must be made to pick one party rather than another. And if simplistic ads, wedge issues, and fear are the most effective means to do that, these are the means the parties will resort to. With voters who do not care very much about politics and only vote out of a sense of duty, this is even more likely to be the case than with others. A duty to vote, then, might mean more rather than less “dumbing down the discourse” and reliance on wedge issues.

A duty to vote will not improve the state of our electoral politics any more than it will make politicians take the interests or opinions of the voters into account, or make our democracies more legitimate. It will neither prevent those who are inclined to break the law to interfere with other people’s votes from doing so, nor discourage political parties from engaging in the sort of campaigning that debases our public affairs and prevents thoughtful discussion of policy issues. In short, in my view, a duty to vote simply cannot achieve any of the purposes that its proponents invoke to justify it. Having dealt with the instrumental arguments in favour of this duty, however, I still must address the deontological one according to which voting is simply a matter of civic duty. I will do that in the next post in this series, probably early next week.