I did two media pieces this week, the first an article opposing electoral reform for Policy Options, the second a panel discussion on Donald Trump on TVO’s The Agenda. There is actually a connection between the two, but I didn’t have enough time on the TV show to explain it. And so let me do so here.
First, electoral reform. The point I’m always trying to make in this debate, which is very difficult to get across without being misinterpreted, is that there is a family of recognizably “democratic” voting systems, all of which have advantages and disadvantages, but none of which is more inherently democratic, or fair, than the others. This is why most academic experts, when they debate the merits of the various systems, tend to talk about extremely pragmatic and consequential considerations, not fundamental democratic principles. In other words, the type of discourse engaged in by Fairvote Canada is, in my view, rank demagoguery. There is nothing inherently more democratic about proportional representation than first-past-the-post. The only interesting question, in my view, is what the likely consequences would be of shifting from FPTP to PR.
My suspicion, of course, is that proponents of PR like the system, not so much because it is more democratic or fair, but because they think that it would produce good consequences, for them, and for their own preferred political party. (As a thought experiment, suppose that PR would likely produce Conservative governments, now and for the foreseeable future. How many current PR supporters do you think would continue to support it, as a “matter of democratic principle,” even if it had this untoward consequence?)
Consider the following example. It is clearly much easier for women to get elected under PR than under FPTP. That’s not because PR is inherently more gender-equal. Both PR and FPTP are gender-neutral. Thus a principled commitment to gender equality, just like a principled commitment to democracy and fairness, does not provide any basis for choosing between the two voting systems. However, as an empirical generalization, PR systems tend to produce the consequence that more women are elected than under FPTP. As a result, I think one could perfectly coherently support PR on the grounds that one thought it was important to elect more women to parliament. But then one should not going around claiming that PR is better than FPTP because the latter is “discriminatory” or “sexist,” one should simply say, “I prefer PR, because I think it would have better consequences.”
What I find frustrating about the current debate is that proponents of electoral reform are, for the most part, refusing to put their cards on the table. Instead of just saying what the good consequences are, that they expect to see from electoral reform, they are hiding behind the claim that democratic principles somehow compel us to adopt their favoured system. I’d like to hear what they think the consequences are – or what they imagine the scenario will be, 5 years after having switched to their favoured electoral system. My suspicion is that it is deeply unrealistic. In fact, my suspicion is that proponents of PR are largely NDP and Green party voters, who imagine that with a switch to PR, the Conservatives will be permanently blocked from power, and that the Liberals will not longer be able to get a majority, and so will only be able to rule by forming a coalition with the NDP and/or the Greens.
Think about this for a moment – what are the chances that this scenario will hold for more than one election? Do they really imagine that the country will be ruled by a Liberal-NDP coalition forever?
Here’s a better prediction. One feature of PR is that it tends to produce a proliferation of political parties. So if Canada switched to PR, one could predict with confidence that it would give rise to at least one new political party. And where would that political party be? On the left? Obviously not. Anyone who thinks about it for a moment can see that the first consequence of PR in Canada would be the appearance of a far-right party. You can see, with the old Reform-PC split, the Alberta Wildrose-PC split, that there is huge tension on the right in Canada, and the only thing keeping it together is the electoral calculus imposed by FPTP. Change that electoral calculus, and the first thing you are going to get is a redneck anti-immigration party, which will get around 15% of the vote, and which will hold the balance of power in any parliament where the Conservative party has the most seats.
So here’s the type of conversation that I think we should be having, when we talk about electoral reform in Canada. Not “which system more truly reflects the will of the people?” but rather “how do we feel about the appearance of a far-right anti-immigration party on the national scene?” and “do we think that would improve the quality of our national discourse?” In other words, I want to talk about the nitty-gritty, the scenarios, not some bogus democratic ‘principles’.
Which brings me to Donald Trump. Last month we had a day-long session with Pippa Norris at the School of Public Policy and Governance – she’s one of the leading experts on electoral systems and electoral reform. At one point during an in-house panel discussion she asked the group of assembled Canadians, “What exactly is the problem in the Canadian political system that you’re trying to fix with electoral reform?” The proponents of PR in the room (i.e. everyone but me) flailed around a bit, and conspicuously failed to come up with an answer. Listening to the unpersuasive responses, you could see her silently making up her mind, that electoral reform is not going to happen here. Why? Because it’s a solution looking for a problem.
To see what this means (i.e. “solution looking for a problem”) one need only compare the situation in Canada to that in the United States. If I was an American right now, I would be an enthusiastic supporter of electoral reform, with PR for Congress, and some kind of ranked ballot for President. Why? The number one pathology of FPTP is that is leads to convergence of the political system around two monolithic centrist parties (Duverger’s law). Right now, that is a huge problem in the United States, but not at all a problem in Canada. (That was the subtext of Norris’s question – given that Canada is not suffering from any of the major defects of FPTP, what’s the motivation to change it?)
If the past several months have taught us anything, it is that the United States should have 4 political parties right now, not 2. There is an obvious sense in which Sanders does not “belong” in the Democratic party, nor does Trump “belong” in the Republican party. In a flexible and responsive political system, both would have formed their own parties, so that there would now be both a “populist” left-wing and right-wing party in America. The majoritarian electoral system, however, penalizes this so heavily, that the representatives of these two populist insurgencies instead tried to “capture” the two more mainstream parties. The Democrats managed to beat this back, while the Republicans failed.
Now here’s the thing about the United States that I find so depressing. No one there is talking about electoral reform. Everyone is talking Trump, Trump, Trump, but no one is talking about making the sort of institutional changes that could easily prevent such an event from recurring. So here we are in Canada, having a serious (in the sense that we really might change things) but also frivolous (in the sense that there’s no real reason to change anything) debate about electoral reform, while in the United States, where they desperately, desperately need democratic reform, no one is even talking about it.
Any why are Americans not talking about it? (And I don’t just mean the media – even the university professors and political theorists, no one is talking about it.) Because they’ve already given up. They simply do not believe that their own institutions are reformable. If you ask political theorists about institutional reform – some of the smartest people in the country, who are paid to think about this for a living – the best they can come up with is some weaksauce response about campaign finance reform and gerrymandering. And yet, if even you got the right alignment of the stars to achieve reform on these two points, at best you would be rolling things back to the way there were maybe a decade or two ago. Nothing fundamental would change.
What America needs is a more fundamental change. The current discussion should not be about people, but about institutions. What about PR for Congress? Or how about a run-off election for President, like in France? Bold reforms like this would instantly change the political dynamic, and solve a huge number of problems.
But, of course, all of that is impossible. Everyone knows it.
What does it mean, however, to say, of the world’s leading democracy, that it is incapable of reforming its own institutions, in order to solve pressing problems? Think for a moment about how unique that is. What other democratic society is incapable of reform? Democracies reform their institutions all the time… except the United States.
And now, just as an exercise, consider how frivolous the current debate over electoral reform in Canada must seem, to an American. (If I were an American, I would be thinking to myself, “Jesus Christ, if I had a democracy that functioned as well as Canada’s, I wouldn’t change a thing…”)
Easy way to defeat Trump – run a centre-right reasonable independent. Will take votes from Trump.
The problem with this, as we have seen, is that this middle-of-the-road “right wing” candidate would have to run with the sole purpose of stopping Trump, rather than running to win. That’s a pretty thankless task, so no one wants to do it.
It’d also really, really alienate a lot of voters that more mainstream Republicans still need to win.
Everyone on the left forgets that from a center right perspective HRC is far worse than Trump. Worse case scenario with trump he pulls something stupid that costs a lot of money, embarrasses the nation and alienates a lot of minorities.
Worse case scenario with Clinton, SCOTUS is lost for a generation, second amendment dies as a consequence, fourth amendment follows close behind, executive branch regulators completely take over the show, sharing economy dies and the entirety of the United States is administered in a way indistinguishable from the state of New York.
“Imagine a torrent of paperwork pouring down on the face of humanity. Forever.”
> The point I’m always trying to make in this debate, which is very difficult to get across without being misinterpreted, is that there is a family of recognizably “democratic” voting systems, all of which have advantages and disadvantages, but none of which is more inherently democratic, or fair, than the others.
I believe you overstate your case here. While you are obviously correct that the Condorcet Paradox exists, a voting system that returns the Condorcet winner if one exists is manifestly more fair than one that does not. If the Condorcet set is larger than one, then it’s fair to say that the will of the electorate is nontransitive, and any method of choosing the winner from that set would be justifiable.
This is the closest electoral systems have to a “no-brainer” decision. The issue you explore in this and your linked articles has more to do with the choice between single-winner and multi-winner systems, and that is indeed a more nuanced choice – not for the least of reasons being that a Westminster Parliament itself is majoritarian in nature.
> Change that electoral calculus, and the first thing you are going to get is a redneck anti-immigration party, which will get around 15% of the vote, and which will hold the balance of power in any parliament where the Conservative party has the most seats.
As you point out later, this is not necessarily a bad thing, despite my probably-similar-to-your feelings about “redneck anti-immigration” policies. Without such an electoral outlet, this faction might be tempted to co-opt the “mainstream” conservative party on an irregular basis, and then under a plurality-majority system (single-winner seats or an Italy-style “winner bonus”) the faction would have outsized impact.
> What America needs is a more fundamental change. The current discussion should not be about people, but about institutions. What about PR for Congress? Or how about a run-off election for President, like in France? Bold reforms like this would instantly change the political dynamic, and solve a huge number of problems.
If you’re opening the door to structural reforms, it would also be interesting to consider the proliferation of veto points. One advantage of a parliamentary system over the US Presidential system is that in the former, there’s a coherent group of people responsible for every part of the agenda. In the latter, both the President and Congress can throw up their hands and declare that what passed wasn’t what they actually wanted, and in so doing avoid at least some accountability for it.
> And now, just as an exercise, consider how frivolous the current debate over electoral reform in Canada must seem, to an American.
Being in that situation, I wouldn’t call it frivolous. Instead, it seems more a breath of fresh air, with reasonable points being discussed more or less reasonably. I’ve repeatedly called Canadian politics “sane” compared to its American equivalent, probably to the utter boredom of those around me.
What Heath seems to be advocating is a FPTP system, except when far right or left factions threaten to take over one (or more) of the two major parties which are characteristic of FPTP. In which case, PR might be the least bad option, in that it prevents the far right or left from gaining enough power to enact their policies.
Which seems to place an extreme value on immediate political expediency.
The other thing we lack down here is robust political parties, i.e. organizations that treat their own platforms as something other than peripheral marketing tactics. That stems from the division of our legislature into two basically equal halves, i.e. into our system of planned incoherence.
Given that the Liberals are in power, perhaps the real important question is: which voting system would benefit them? Given that AV/instant-runoff votes are thought to benefit centrist parties, I would have thought that is the system they would go for.
I would have thought it was obvious that the problem many people want to solve with electoral reform is the problem of lesser evilism – that is, a system that forces you to vote for something you hate in order to combat something you hate even more. If Ross Perot wants to run on right wing ideas and people want to support him, or Ralph Nader wants to run on left wing ideas and people want to support him, why should the system translate those actions into election of a more left wing (right wing) government? Isn’t a system that doesn’t have this perverse behaviour superior to one that does?
I’ve been around a lot of Green/NDP supporters of electoral reform and it often seems to come back to wanting to be able to just have a party that represents what they believe in, without being forced to always choose between voting Liberal to keep out the Conservatives or voting on principle and risking splitting the vote and electing what you most hate by voting for what you most like.
“There is nothing inherently more democratic about proportional representation than first-past-the-post.”
It seems you are either denying that a correspondence between number of votes and representation is a key element of democracy or denying that political parties are a valid part of the political landscape – neither of these points of view seems tenable.
Perhaps you meant that first past the post has other advantages that offset its lack of proportionality so there is no ‘real’ difference, but this is necessarily a matter of opinion, so of course you are welcome to your own opinion, but to assume that others with different opinions must be lying because there is objectively no difference between systems is to privilege your opinion as fact.
Finally, you seem to be saying that the U.S. system is broken because it forces the right wing vote to coalesce into a single party when it would naturally be two, but the Canadian system works well while it forces the right wing vote to coalesce into a single party when it would naturally be two. I don’t follow this line of reasoning (and I suspect Albertans in particular might find it irritating at the moment).
Are you sure you aren’t just being contrarian because you are surrounded by pro-PR people who you think are idiots? Because nothing you’ve said here really makes any sense, and that is not like you.
On that last point — I’m quite serious when I say that I think the argument hinges upon consequential/empirical considerations, not first principles. (Hence the appearance of saying two different things.) There is no question that FPTP pushes in the direction that Duverger’s law suggests. The question is whether this has become oppressive or not, in a particular country, with a particular political culture, at a particular time. I think that in the U.S. it has become oppressive, and the political system needs to be opened up (that’s a bit of a simplification, since the two parties in the U.S. have also managed to legally entrench themselves, making it almost impossible for new parties to be formed and to compete effectively — so it’s not just the voting system). In Canada, however, I think there are centrifugal forces that prevent the push toward the median voter from becoming oppressive. (As a result of which you do not have the same alienated fringes who feel constantly denied representation.) So the same tendency that is oppressive in U.S. political culture is, I think, salutary in ours.
And no, it’s not that I’m just being contrarian. Fundamentally, it’s because I like majority governments. The first two years of a majority mandate are, for me, the only time when politics is exciting, where stuff can really happen. I prefer ideologically pure governments doing what they actually want to do — even when I disagree with it — than complicated, messy compromises, that slowly turn the state into a kludgeocracy.
Joseph, would you argue that the Conservative majority of 2011 was ideologically pure, and that it did not make changes to the state (the tax code immediately comes to mind) that were needlessly complicated and inefficient?
I agree with Andrew Coyne on this one: our democracy is lessened when it permits rule by a minority of the electorate. The fact that FPTP results in false majorities that can, notwithstanding the actual percentage of electors who voted for them, relatively easily impose their policies is a bad thing. You can say that’s not a result of FPTP but rather of too much party discipline, but, using your own argument, I see the institutional solution of electoral reform as more easily achieved than breaking the grip of the political parties.
If you were making a comparison between AV (single-member district ranked ballot) and a system of PR (including multi-member district ranked ballot STV, mixed member proportional with 2 ballot questions, etc) then I would agree with what you said.
But to say that FPTP is not inherently less representative and thus less democratic is IMO wrong.
I believe you have fallen prey to Fair Vote Canada’s rhetoric that is so partisan towards PR as to suggest that any system that only considers a single geographic district are all equivalent and all undemocratic. What FVC is saying is dishonest, wrong, and harmful to the educational process required for Canadians to think about any type of modernisation of our electoral system.
There are US citizens and groups talking about electoral reform in the USA, just as there have been in Canada for decades. More Canadians are talking about this because those who think the “unite the X” movements will make Canada into the same flawed 2(ish) party system like the USA are looking for ways to avoid that scenario. And that is what FPTP will lead to, and no other system (AV or the wide variety of PR systems) will.
It is sad that the talk in the USA is about Trump (for or against). If I was a US citizen I would be talking far more about how dangerous I think Clinton would be. I have far less of an opinion about Trump. But that is what FPTP does — reduces complex conversations down to a yes/no question as it only works if there are only two answers to choose from.
IMO Joseph is correct in saying that “FPTP in not inherently less representative & thus less democratic”.
Consider the primary rational presented yesterday by Minister Monsef:
“Look at 1997 as an example. The Reform Party garnered 18.7 per cent of the vote and received 60 seats, whereas the Progressive Conservatives garnered 18.8 per cent of the vote, virtually the same, but received only 20 seats. This system tends to favour parties with regional, rather than national appeal.
And in the election previous to that one, in 1993, the Progressive Conservatives won 16 per cent of the popular vote but only two seats. Meanwhile, the Bloc Québécois received 13.5 per cent of the popular vote and won 54 seats.”
At a national level these results indeed make FPTP seem unrepresentative. But consider the results at a provincial level:
Lib PQ PC NDP Reform Other
1993
Quebec % Vote 33.0 49.3 13.5 1.5 0.0 1.1
Seats 19 54 1 1 0 1
1997
Alberta % Vote 23.0 0.0 15.4 5.7 54.7 1.2
Seats 2 0 0 0 24 0
BC % Vote 28.8 0.0 6.2 18.2 43.0 3.7
Seats 6 0 0 3 25 0
SASK % Vote 24.7 0.0 7.8 30.9 36.0 0.6
Seats 1 0 0 5 8 0
In 1997 Reform won seats in only 3 provinces &, of course, the BQ only applied to Quebec. The provincial results are very much in line with a PR system. So the distortion is only at a national level. Are we to apply a PR formula at a nation level? I would hope not. As long as we elect members at the district level (heaven forbid we should change this!) there is little that is unfair with FPTP.
There’s an unconsidered option: preferential voting, which exists in Australia.
In the US situation, you could have had the four main candidates run for office, Clinton, Sanders, Cruz and Trump.
Then a voter could write down their preferences, i.e. 1. Sanders, 2. Trump, 3. Cruz, 4. Clinton. Or 1. Clinton, 2. Cruz, 3. Sanders, 4. Trump.
First you count up the first preferences. If no candidate reaches 50% of the vote on first preferences, then you redistribute the second preferences of the last running candidate.
So on this case:
31% Clinton
30% Trump
29% Sanders
10% Cruz
You’d grab the second preferences of the Cruz voters and redistribute them. In which case, you’d probably end up with something like this:
31% Clinton
37% Trump
32% Sanders
Now, you’d grab the second preferences of the Clinton voters and redistribute them:
38% Trump
62% Sanders
And now you have Sanders as the winner of the preferential voting system, despite Sanders coming 3rd in the count of first preferences.
To me, it’s an elegant solution to increasing the number of candidates for an election without some of the problems associated with proportional representation.