Electoral Reform Committee on Oct. 19th, 2016

[Matt DeCourcey, Liberal, Fredericton, NB]

Thank you.

Ms. Ghose, I listened, and despite an admission that a well-run referendum can be a valuable educational experience for a population, I sensed an overall aversion to going down the road of referendum in validating a new electoral system. If I'm wrong, please correct me. I wonder if you can explain your view, or your experience, with the multitude of different electoral reforms and modernizations that have gone on within the U.K., and Scotland, in particular, over the last number of years, and the importance of properly educating and validating those reforms with citizens as they were being implemented, or before they were being implemented, or post-implementation.

[Katie Ghose]

Yes, sure.

As I suggested, I think we've ended up, in quite a British fashion, with sort of incremental and piecemeal changes, which has given us a kind of laboratory of different electoral systems. They have different origins, as well, so we've had brand new institutions. We've had a lot of devolution of power from Westminister, and we've had these brand new institutions in Scotland and Wales, and with those came new non-first-past-the-post electoral systems.

I suppose in looking at those examples, it wasn't necessarily that the citizens perhaps had some sort of a role in validating the idea of a new devolved institution. There were referendums around those, but not always particularly on the electoral system. So the origins have really been quite diverse.

I think I've spoken a little about the 2011 AV referendum. There have been quite a lot of examples where a new system has been kind of imposed, or political leaders have decided that this would be part of an institution and that's what they were going to do. That's been quite a common pattern as well.

I should say on the referendums, and we set it out thoughtfully in our report, that we didn't take a position as to whether in and of themselves they're good or bad for us. It's all about the context and the timing and how they are conducted, and how much emphasis is put on public information, public education, and the public role. We made nine recommendations about the good conduct of referendums, with a lot of emphasis on the public role, starting even when the legislation for a referendum had been put through, having a strong citizen role there.

There has been quite a variety of impetus and motivation behind that, and therefore, the public role is varied as well. The thing that really stuck out for me about the electoral reform referendum was how very low the prior knowledge of the public was, and how there were no real opportunities to become educated about the status quo, about first past the post. If people don't understand the status quo, it's quite hard to have a rich conversation about what to replace it with.

[The Chair, Liberal, Francis Scarpaleggia]

We'll go to Mr. Reid now.

Thank you, Mr. DeCourcey.

[Scott Reid, Conservative, Lanark—Frontenac—Kingston, ON]

I want to start by saying I had a delightful meeting about two or three weeks ago—maybe it was a month now—with Darren Hughes in my office. We meant to meet for half an hour and stayed for two hours. It was really informative. So, Darren, don't take this the wrong way if I address all my questions to Professor Becker instead.

Professor, I want to ask you a series of questions.

For one thing, you run a number of models, and I want to explore a bit of them. I want to start by dealing with something you raised in your written submission to this committee but which you did not raise in your presentation this evening, and that is the metric that you used to judge whether or not a system is proportional, the Gallagher index.

Could you chat with us a bit about the Gallagher index and how you apply it to your models?

[Byron Weber Becker]

Sure.

In the graphs we looked at earlier, there's always a gap between the percentage of votes and the percentage of MPs. The Gallagher index takes all of those gaps and it expresses them as a single number. Rather than saying, well, the Liberals are overrepresented by 15% and the Conservatives are under-represented by 7% or whatever, it takes those numbers and puts them into a single number. We can talk about how it does that if you want, but the best intent....

One of the downfalls of the Gallagher index is that overrepresentation, say, in Alberta, can be offset by overrepresentation in, say, the Maritimes. That does not show up in the classic Gallagher index, so I adapted it to what I call the composite Gallagher index. I calculate the Gallagher index for each region of the country and then average them together, so that disproportionalities in the Maritimes and disproportionalities in Alberta both contribute to an overall picture of how proportional the system is.

[Scott Reid, Conservative, Lanark—Frontenac—Kingston, ON]

First, to be clear, if we go to your written material, the material that was submitted to us, in that we would find the composite Gallagher index being used and not the original Gallagher index.

Is that correct?

[Byron Weber Becker]

I present both of them.

[Scott Reid, Conservative, Lanark—Frontenac—Kingston, ON]

Okay, all right.

To make sure I'm crystal clear on this, when the first prime minister Trudeau was in office, you would have situations where he had zero seats in Alberta and all but two of the seats in Quebec. Those would have cancelled each other out and given the impression that the system was actually less equitable, but it appeared more equitable than it would have been if he'd merely had the disproportion, either up or down, in only one of those regions.

Is that correct?

[Byron Weber Becker]

That's correct. The classic Gallagher index does not account for those regional disparities, but the composite one does.

[Scott Reid, Conservative, Lanark—Frontenac—Kingston, ON]

That's very helpful. Thank you.

I mention this for the benefit of colleagues. I mention this in part because I think that if the committee is trying to design proportional alternatives to the status quo, the composite Gallagher index would be a good way of testing the various models that might be looked at.

I assume it's the case, Professor, to return to you, that when you're faced with multiple versions of multi-member proportional, or multiple versions of single transferable vote, one could actually set as one's goal accomplishing the lowest Gallagher index number that is compatible with the hard barriers we face, such as the fact that we have to have all our seats within single provinces. Would that be correct?

[Byron Weber Becker]

Yes. The composite Gallagher is an excellent measure of the disproportionality of a system, and the lower that number, the better. I think it's also fair to say that once we get into the range of 2% to 4% , at that point I have a hard time making the claim that decreasing it even further needs to be our primary goal. I think that once we have it down in the 2% to 4% range, our goal can shift to asking which one is simpler, or which one preserves the connection between the MP and the local voter better.

[Scott Reid, Conservative, Lanark—Frontenac—Kingston, ON]

When you get down to those very low numbers, I gather that pushing them down further may require heroic measures such as doubling the size of the House of Commons, or increasing riding sizes to make them vastly larger than they are now in order to get another couple of percentage points in improvements. Am I understanding that correctly?

[Byron Weber Becker]

Yes, and those heroic measures in my estimation, once we're already down to 2% or 3%, just aren't worth it. It's the law of diminishing returns. There's more important stuff to pay attention to at that point.

[Scott Reid, Conservative, Lanark—Frontenac—Kingston, ON]

That's very helpful.

I now want to turn to the models you've worked on. I had this conversation with you earlier. You provide numerous models that involve adding seats to the 338 we already have. For example, the most recent model you circulated to us, the AV-plus model, involves 32 list seats being added on. If time were not an issue, this wouldn't be a problem, but time is an issue as the 2019 deadline is approaching. I'm worried that if we were to do that we would run into the issue of having to confirm with the courts the constitutionality of doing this because of the fact that section 52 of the Constitution Act, 1867, says:

The Number of Members of the House of Commons may be from Time to Time increased by the Parliament of Canada, provided the proportionate Representation of the Provinces prescribed by this Act is not thereby disturbed.

You have to disturb it to some degree, and obviously, the courts would say that some degree is permissible, but not too much. Once we go into trying to sort that out before the courts, the 2019 deadline may be lost.

Would it be possible for you to create some additional models that assume we start with the 338 MP cap we have now, but that try to be as close to the lowest possible number on the composite Gallagher index, looking at STV, MMP, and also rural-urban? Would it be possible to do that and come back to us?

[The Chair, Liberal, Francis Scarpaleggia]

Answer very briefly, please, if you can.

[Byron Weber Becker]

Sure. I do have a day job, so I want to talk more with you about exactly what you're asking for, but I would be willing to make myself available to this committee for more modelling work.

[The Chair, Liberal, Francis Scarpaleggia]

Thank you.

Mr. Cullen.

[Nathan Cullen, NDP, Skeena—Bulkley Valley, BC]

Thank you to all our panellists here tonight.

This is one very rich field that I want to get into, so time is of the essence. Thank you for that most recent offer, Mr. Becker.

As I understand it in looking through your graphs, the steeper and more dramatic the curve is, the more misbehaving and the more distortion is going on within the system.

[Byron Weber Becker]

Not quite.

[Nathan Cullen, NDP, Skeena—Bulkley Valley, BC]

Not quite? What am I wrong about?

[Byron Weber Becker]

It's hard to tell on these monitors, but those lines come in coloured pairs. There are two red ones, two green ones, two blue ones—

[Nathan Cullen, NDP, Skeena—Bulkley Valley, BC]

Yes, the more they separate—

[Byron Weber Becker]

—and the more they separate, the more misbehaving the system is.

[Nathan Cullen, NDP, Skeena—Bulkley Valley, BC]

First past the post is bad.

[Byron Weber Becker]

Yes.