Our good friends at Fair Vote Canada have decided to adopt the new buzz-word of “evidence-based” as part of their new campaign for electoral reform. The “evidence” in this case is the summary of study by Arend Lijphart that compares thirty-six democracies. Immediately, this should be a red flag right there – comparing democracies that use different models of governance is pretty much like comparing apples, grapefruits, bananas and goats – which this study pretty much confirms. The study also glosses over actual substantive issues like responsibility and accountability in favour of focusing on the act of voting itself. This is not a surprise given the particular ideological bent of Fair Vote Canada, for whom it’s the act of voting that is important, rather than what the vote means, and how that act works in the broader context of the system as a whole.

The summary focuses on seven particular points of comparison, along a particularly specious axis of “majoritarian” and “consensus” democracies.” Those points are the closeness of government and citizen views, the overall satisfaction with “the way democracy works,” voter turnout, political and economic equality, the percentage of women serving in parliaments, the proliferation of political parties, effective economic policy management, and effective environmental policy management.

Immediately you should be able to see some of the problems with these particular points of comparison, in that many are subjective, superficial, or at worst, specious. For example, the measure of “closeness of government and citizen views” is based on a plot of government policy with those of the median voter on a left-right scale, while “satisfaction with democracy” is based on survey data where people may not understand the way in which their democratic system works – simply their perceptions of it, which is hardly a fair comparison to the efficacy of said system. The measure of voter turnout is a particularly fraught question, because there has been no definitive correlation between voter turnout levels and electoral systems. Voter turnout tends to be based more on a more complex series of behaviours, including the age and experiences of the population – people tend to vote more once they start raising families and paying taxes, for example. (Colby Cosh has some particularly effective takedowns of the turnout nerd arguments here, here, and here).

Even the Fair Vote Canada summary acknowledges that it’s highly debateable to equate political and economic equality – but it doesn’t stop them from saying that consensus-based democracies have lower economic disparities (probably because their prime examples, Scandinavian countries, are more distributive economies). The section on women in parliaments doesn’t talk about the differences between women being elected in ridings versus those appointed based on proportional lists, which becomes an issue of tokenism rather than actual representation or indication of the rate of women participating in the political process. It treats more parties as a good thing in a system, because proportional systems allow more of them to win – but neglects the kinds of brokerage efforts within “big-tent” parties in majoritarian systems that contrast with the post-electoral brokering that happens in “consensus” systems. The conclusions of economic and environmental performance based on electoral systems are also inconclusive, and yet the Fair Vote Canada summary says that they also don’t disprove that consensus systems are worse for either, which is a poor argument if there ever was one.

But with these points all relatively debunked, what is missing from any of this consideration are what the votes actually mean. How strong are systems of accountability in majoritarian versus consensus democracies? After all, in majoritarian systems, it is easier to “vote the bums out” than it is in a consensus system where a party can simply shuffle around its coalition partners to remain in power for decades. Similarly, in systems where party lists appoint representatives based on voter share dominate, there is no accountability mechanism for those representatives, except to the party itself. How do opposition parties hold governments to account in consensus systems? Are governments insulated by fixed election dates, or can they be defeated and either a new government formed or an election called, such as with a system of Responsible Government (which largely occurs in a majoritarian context)?

This kind of questioning goes unanswered, because in the rush to proclaim counting systems are so utterly complicated as to make “every vote count” in an unequal and distorted way to satisfy the emotional demands of “fairness,” the actual questions about the substance of governing and accountability are ignored. It would seem to me that if you’re going to compare the robustness of democracies, that the efficacy of those votes actually have meaning. Instead, we are treated to a survey of surface concerns that don’t address any actual substantive questions.

But for a group that campaigns for voter reform based on feelings and the subjective qualities of “fairness” that don’t actually mean anything in the pursuit of governance, finding the thinnest veneer of “evidence” to support their claim – apparently in the name of science – is not a surprise. If anything, it further condemns the term “evidence-based” to something as faddish as super-foods or going gluten-free. A superficial comparison between apples, grapefruits, bananas and goats – hey, they’re all carbon-based! – is not “evidence,” nor does it make a substantive point in the comparison of governance structures. It only demonstrates the just how intellectually bankrupt the arguments that Fair Vote Canada champions really are.