Electoral Reform Committee on July 26th, 2016

[Luc Thériault, Bloc, Montcalm, QC]

However, that's not the people's will. Party apparatchiks conduct those negotiations. The people are not involved in platform components and elements that will or won't be integrated by a coalition government.

[Michael Marsh]

One could argue exactly the same in a first past the post system, inasmuch as one party got a majority and implemented its program, but it might only have 30% of the vote, or 35% of the vote, or 38% of the vote. Is that the will of the people if it's implemented what 62% of the people don't want? I think the last Indian government was elected with an overall majority with less than one in three of all votes cast. Is that the will of the majority?

[The Chair, Liberal, Francis Scarpaleggia]

That's an interesting point.

We'll go now to Ms. May.

[Elizabeth May, Green, Saanich—Gulf Islands, BC]

Thank you and good afternoon in Ireland.

I hope some day I'll be seeing you on more than video; this has been very helpful.

Professor Gallagher, I've been reading a chapter from a book you wrote, Ireland: The Discreet Charm of PR-STV. I've always wondered how it is that STV can be proportional, given, as you say, that there's no proportionality that is privileged by the way the seats are organized; there's no separate set of seats to represent the imbalance that's created by voting at the constituency level. I was very taken with the table that was in this chapter and the results of the 2002 general election. The parties had remarkable consistency between the number of seats they had and the proportion of the votes they got even though there's nothing that requires that.

In the work you've done, how do you explain the consistency? The 2002 election results are in your chapter for purposes of example, but I take it that it's fairly typical in results that you get a fair coherence between the percentage of the vote for the party and the percentage of seats that are won. How do you explain that in STV?

[Michael Gallagher]

Within each constituency there's a reasonable degree of proportionality, especially in the larger ones, such as the five-seat constituencies. In three-seat ones, in particular, you might not get such proportional results, but what nearly always happens is that, simply on the law of averages, if a party loses out in one place they'll win out on another occasion. Our third party, for example, the Labour Party, might win 10% of the votes on average in most constituencies and sometimes that's enough for a seat, sometimes it's not enough for a seat. On the law of averages, they end up with something like 10% of the seats, generally. It would be very unlikely, statistically, that they would systematically lose out nearly everywhere. In practice, it does deliver quite proportional results, as you say.

[Elizabeth May, Green, Saanich—Gulf Islands, BC]

I'll go on to another question I have about how this has affected the nature of national parties. It seems that the two largest parties in Ireland have remained the two largest parties over time quite consistently and remain dominant. We had a witness yesterday who was concerned that if we got rid of first past the post, the stability of large parties might be undermined. Not to take too long in asking my question, but I noted that you also mention in this chapter that the Irish Constitution makes no mention of parties—neither does the Canadian Constitution—and electoral law did not recognize them or put them on the ballot until 1963. In Canada it wasn't until about a decade later.

How have you found STV affects the stability of the larger dominant parties?

[Michael Marsh]

I think it's worked in different ways. In the early years what happened is when people voted for a party, they'd vote for, really, all the candidates of that party; they'd vote very much on party lines and wouldn't vote for the other big party.

Nowadays, it's much more mixed, so I think what we see in the results—and the electoral system facilitates that being translated into seats—is that people think rather less of those parties than they used to and even though they might vote for a candidate from one of those parties, they don't, to anything like the same degree, support all the candidates of those parties. They may be a little more candidate-centred than they used to be. If that's what the voter wants, then democracy said that's what the voter should get.

[Elizabeth May, Green, Saanich—Gulf Islands, BC]

Did you want to add to that, Professor Gallagher?

[Michael Gallagher]

I'd just add the caveat I mentioned before, which was that we shouldn't use the electoral system to explain too much. As my colleague says, parties' fortunes wax and wane, and that probably doesn't have very much to do with the electoral system. It would have happened under any electoral system.

[Michael Marsh]

I think we're seeing the waning of large parties right across Europe, and the Irish ones are no different. We've seen it in Britain with first past the post.

[Elizabeth May, Green, Saanich—Gulf Islands, BC]

I want to ask one last question. It's really helpful to have your sample ballot. It's interesting that you have the photograph of the candidate, the party logo next to the candidate, and you list the candidates alphabetically.

Would it be fair to say that single transferable vote systems are very easy for the voter to use, but perhaps more complicated for the electoral officials to count? In terms of what's simple and what's complicated, it's simple for the voter.

[Michael Gallagher]

I think that's absolutely right. Yes, for the voter it really is very straightforward to vote. The counting process, like a lot of counting processes, actually, with a lot of electoral systems, can sound complicated if you go through a stage-by-stage account, but for the voter, it's very straightforward.

[Michael Marsh]

Can I just add one caveat to that? In Malta, for instance, which uses the same system, the ballot is structured by party. That, I think, probably makes it even easier for the voters, because sometimes it can be quite hard to find all the candidates of your party on a long list.

Second, when it comes to counting, the system that's used in Scotland, for instance, in local elections, is electronic, so it's instant.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you very much.

We'll go to Mr. Aldag for five minutes.

[John Aldag, Liberal, Cloverdale—Langley City, BC]

Thank you, Mr. Chair.

Our committee has been given several principles to work from. This isn't the full list, but we're looking at a few things: How do we increase the engagement of our citizens? How do we design a system that is accessible and inclusive? How do we maintain integrity? Two specific areas we've been asked to look at are online voting and mandatory voting.

I'd like to start with online voting and any thoughts or comments on research you've done or discussions you've had within the Irish context. Perhaps I could get your thoughts on that, just to start.

[Michael Marsh]

In simple terms, we don't have online voting. Postal voting is very difficult here, and I think online voting is a long way away.

[John Aldag, Liberal, Cloverdale—Langley City, BC]

From a research perspective, have either of you dabbled in that, or has it not been something you've spent time on?

[Michael Marsh]

There is research on postal voting. In some jurisdictions, I think Sweden, most people vote long before the election takes place. The hope was that postal voting would make it easier to vote, and therefore would raise turnout. Most of the research with which I am familiar says that what happens is that those people who would vote anyway find it easier to vote, and those people who wouldn't vote anyway don't vote just because they can vote by post. It facilitates the regular voter, not someone who's turned off from the system.

[John Aldag, Liberal, Cloverdale—Langley City, BC]

Okay. Thanks.

Do you have any thoughts on the question of mandatory voting?

[Michael Marsh]

I think mandatory voting probably addresses the symptom rather than the cause. If there's a lack of interest in politics, which manifests itself in people refusing to vote, forcing them to vote doesn't necessarily make them interested in politics. The phenomenon in Australia known as the “donkey” vote, in which people just fill in the ballot alphabetically because they don't know enough or they don't care enough to fill it in any other way, is very much a product of mandatory voting.

[John Aldag, Liberal, Cloverdale—Langley City, BC]

Great. Thank you.

The first slide of your PowerPoint presentation stated that possibly MPs could be more socio-demographically representative of the population as a whole. As we're looking at trying to reflect our ever-diversifying population, looking at such things as more women in politics, do you have any comments on that point? It's not a ringing endorsement saying that this will result in a greater socio-demographically representative group of MPs. Do you have any thoughts on the strengths or weaknesses of your system, and how it could enhance representation of the population?

[Michael Gallagher]

Generally speaking, countries with PR systems do have more women in Parliament than countries that don't. Being confident that this is cause and effect is a different matter. Maybe it's just that the countries are different, that their whole political culture is different.

Ireland, I must say, has always done very poorly on this criterion. At the last election, we introduced candidate gender quotas. That did have the effect of increasing the proportion of women to 22%, which is the highest for Ireland, but Ireland is not a great exemplar on this point. Again, I think it's one of those things that is not really determined centrally by the electoral system.

To go back to something I said before about open-list and closed-list PR systems, with a closed-list system the party in effect picks the MPs. The party can put a lot of women at the top of the list, or people from ethnic minorities, and in that way really engineer the composition of its parliamentary group. It can be used for that purpose, but at the cost of freezing the voter out of the decision-making process.

[John Aldag, Liberal, Cloverdale—Langley City, BC]

Here is a quick technical question. I've been reading about all these various systems, and in a lot of them when a certain threshold is reached, then the surplus votes go into a new pile and get redistributed. My assumption is that the first one gets locked in, and then come the ones that still need to be counted. I'm trying to figure out how that would work in our system, where we have a number of polls. How do you determine which are the first batch of votes, the votes that are locked in, and which ones get redistributed?

[Michael Gallagher]

There are lots of ways of doing that. In Australia it's done differently than it is in Ireland, and even within Ireland, when we elect the upper house, the Seanad, it's done in a different way. Essentially, it's seen as a random process in which the votes transferred are a cross-section of all of them with regard to their next preference. They're randomly selected, however, with regard to lower preferences.

That's the real nitty-gritty of PR-STV, which I probably shouldn't even be trying to explain, because it makes it all sound much more difficult than it really is.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.

We'll go to Mr. Kenney now.