October 4, 2016
Second only to deciding which electoral system Canada should use, the question of whether a referendum should be required to ratify whatever decision it reaches has emerged as perhaps the most important choice confronting Parliament’s Special Committee on Electoral Reform.
Unfortunately, the somewhat different question of whether referenda work well as public policy decision-making tools has been all but lost in the jockeying for political advantage that has characterized these discussions. Standing in the shadows of Brexit and the recent Colombian vote against the proposed peace accord with the FARC, we’ve been concerned by this absence and decided to try and fill this gap by thinking through the question from an evidence-based perspective.
“Democracy is the theory that the common people know what they want and deserve to get it good and hard.”
– H. L. Mencken1
A historical perspective
Theoretically, referenda sit uncomfortably within Canada’s Westminster parliamentary tradition. Edmund Burke, the father of British conservatism famously opined: “your representative owes you, not his industry only, but his judgment; and he betrays, instead of serving you, if he sacrifices it to your opinion.” Clement Attlee, a founder of the British welfare state, was more succinct in labelling referenda a “device of dictators and demagogues”.
The referendum that is likely the most familiar to Canadians is the 1995 Québec referendum on sovereignty-association. In addition to almost splitting the country, that vote is also indirectly responsible for one of Canada’s greatest contributions to international law, namely the Supreme Court of Canada’s secession reference, which has been widely lauded for providing clarity on the legal rules that govern the break-ups of countries.
While the secession reference embodies the thorough discussion Canadians have had on how to hold a referendum, rigorous consideration of the question of whether holding one is even a good idea has been lacking – a problem increasingly on display in the pages of Canada’s major newspapers. Tackling this question can be politically daunting, but the importance of the issue clearly calls for an evidence-based consideration of the pros and cons of referenda as a policy-making tool.
Most concerning is the charge that referenda are illiberal tools that reduce constitutional democracy to mob rule by obliterating the constitutional half of this equation in favour of the democratic one.
First, many proponents argue that referenda can improve the public’s understanding of public policy. This would be positive as low “political knowledge” is one of the factors depressing voter turnout in Canada,2) an alarming trend that can produce poorer governance. By giving the populace more control over public policy decisions, the argument holds that we would also see a commensurate rise in individuals educating themselves and taking responsibility in the making of these decisions.
Similarly, referenda also appear to strengthen what political scientists call “political efficacy” which refers to the sense that individual political action has an impact on the political process.3 Falling political efficacy is an important driver of declining voter turnout and reversing this would be a positive outcome. Unfortunately, the educative and efficacy effects of infrequent one-off referenda are likely to be limited. To truly unlock these benefits, referenda would need to occur more frequently – a change that would require a significant refashioning of our politics along say, Swiss lines.
More importantly, proponents argue that referenda are the gold standard in democratic legitimacy. By providing a definitive answer to a question, even non-binding plebiscites provide a rare and clear statement of the public will that law-makers ignore at their own peril. Indeed, some laud referenda precisely because they create a valuable means of checking increasingly powerful executive branches of government, an ability that may otherwise be lacking in parliamentary systems.4
On the other hand…
Alternatively, referenda can also harm societies’ cohesiveness. By boiling complex issues down to a simplistic yes or no question, referenda remove much of the scope for compromise in politics. Given that parliamentary politics is in large part designed to be a non-violent compromise-generating machine, this is quite dangerous – especially when some seek to stoke such divisions for personal gain. This is especially problematic when, as probably occurred with Brexit, the majority of voters may actually prefer a third middling option over either of the two extreme ones on offer.
Worryingly, even when referenda are designed to be nuanced, they can be ineffective because they produce too much uncertainty. The 1995 Québec referendum question is a good example:
“Do you agree that Québec 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 Québec and of the agreement signed on June 12, 1995?”
The “bill respecting the future of Québec and… the agreement signed on June 12, 1995” that are referenced run to just over 4,800 words. What voting ‘yes’ on this question actually meant so confused the campaign that much of the aforementioned Supreme Court reference actually dealt with the necessity of avoiding such a situation by offering voters a clear question. Taken together, these twin problems suggest that some questions are simply not amenable to clear yes/no answers and arguably make a strong case against referenda for all but the most binary questions.
Additionally, referenda campaigns can become particularly rancorous because many of those campaigning will never be held accountable for the vote’s outcomes and are thus encouraged to exaggerate their arguments and promises. This lack of accountability can dangerously inflame popular passions and lead to factionalism (that great bogeyman of the US founders) and even violence.
Another classic concern about referenda is that they are prone to manipulation, especially by the governments who call them. Indeed, governments decide when referenda are called, what the question is, whether they are binding or not, how the ‘For’ and ‘Against’ campaigns will be funded and what rules they must abide by – all factors which can influence the results. Indeed, even solidly democratic regimes have manipulated referenda by requiring arbitrarily high thresholds for success and by mounting poor informational/educational campaigns.
Most concerning is the charge that referenda are illiberal tools that reduce constitutional democracy to mob rule by obliterating the constitutional half of this equation in favour of the democratic one. The long history of referenda in the United States used to exclude subsidized housing (read: racial minorities) from wealthy neighbourhoods are stark illustrations of this concern. Indeed, the constitutional in constitutional democracy is there precisely because of the need to sometimes check the tyranny of the majority. The sad history of Proposition 8 in California provides a recent example of this point.
So, are referenda good or bad? Well, it depends…
Of course, the practical question is not really whether referenda are good or bad public policy decision-making tools but rather whether they are better or worse than other options in specific situations.
A consensus does seem to have emerged around the idea that referenda represent a necessary tool for resolving questions of great importance, such as a major constitutional change or secession. Since these are such weighty questions, the ability of referenda to robustly bestow a mantle of democratic legitimacy on a decision is generally seen as valuable and as overwhelming other concerns.
Outside of such straightened circumstances, however, the case for referenda seems less solid – especially given the availability of better alternatives. Most obviously, given that voters already elect and pay legislators to do the hard work of making tough decisions so they don’t have to, it becomes difficult to justify holding a referendum without an extraordinary reason to do so. Even if there is a desire for direct input from ordinary citizens, the innovation of other tools (e.g. citizens’ reference panels) offers cheaper and likely superior mechanisms for developing good public policy.5
Ultimately, referenda are usually the result of political dynamics beyond the scope of this discussion. But from the perspective of effective public policy, it seems pretty clear that, absent some additional revolutionary changes to our political system, referenda tend to yield little in the way of positive outcomes in normal circumstances. Moreover, given that these minimal positive outcomes risk being swamped by more serious negative ones, the case for referenda as a public policy tool seems pretty weak.
More related to this topic
Michael Crawford Urban
Oct 4, 2016
- H. L. Mencken “A Mencken Chrestomathy: His Own Selection of his Choicest Writings (New York: Vintage Books, 1982) p 622. [↩]
- Howe, P. 2010. Citizens Adrift: The democratic disengagement of young Canadians (Vancouver: UBC Press [↩]
- Kim, T. 2016. “The Effect of Direct Democracy on Political Efficacy: The Evidence from Panel Data Analysis”. Japanese Journal of Political Science 16 (1) 52–67. pg. 53. [↩]
- Daly, E. 2015. “A republican defence of the constitutional referendum”. Legal Studies 35(1) 30–54. pg. 32. [↩]
- Pal, M. 2012. “The Promise and Limits of Citizens’ Assemblies: Deliberation, Institutions and the Law of Democracy”. Queen’s Law Journal 38(1): 259-294. [↩]