Crawl Across the Ocean

Tuesday, December 06, 2005

Green Party Debate - Rationale For Inclusion

As promised, this is a follow-up to this call for action to let your media know what you think of the Green Party being excluded from the debate.

In that post I didn't explain why I thought the Green Party should be included so I'll do that here.

To begin with, we have to ask, what is the point of having a televised debate? Seems obvious enough, the point is for voters to get more information about their options so they can make an informed choice.1

So the goal, or optimal result of a debate would be to expose the greatest number of people to the greatest number of potential voting options, over the long term.

This leads to a tension between two things. On the one hand, the more party leaders present in the debate, the more viewpoints a viewer will be exposed to. On the other hand, as the number of leaders present increases, each leader gets less time, so it is possible that including leaders of parties that few people are considering voting for, at the expense of giving less time to leaders of parties that most voters are going to consider voting for, could be sub-optimal.

Furthermore, a debate which includes many different people, or which includes people that a voter is not considering voting for may reduce their willingness to watch the debate, also producing a sub-optimal outcome.

The media has never come out and clearly stated what their criteria for including a party in the debate is, which seems like a huge oversight. Given how much scrutiny is given to election advertising generally, the allocation of what amounts to by far the most valuable advertising in the entire election is left in the unaccountable hands of a few unelected people who make a selection based on - well, who knows, really.

Assuming that we are unwilling to just say that it should be left up to the subjective judgement of the broadcasters to decide who's in and who's out, without requiring any rational justification for their decision whatsoever (a position so weak, I see no need to spend time refuting it) it seems reasonable to point out that the only remotely logical, consistent position left to the broadcasters is that they only include parties if they hold a seat in parliament (ignoring that this rule has not been consistently applied at the provincial level).

As an aside, it's worth noting that, given how our electoral system works, such a policy would be tantamount to an official policy on the part of our broadcasters to only support the development of regional protest parties, not national ones - and we wonder why national unity is an issue!

Anyway, as much as it would be nice if our broadcasters could reinforce national unity, rather than implicitly supporting the breakup of the country, the point of the debate is not to promote national unity, it is to educate voters. So does a policy of only including parties with a seat in the House of Commons serve this purpose? Well, it is better than no rule at all, since it generally does differentiate between parties with more (have a seat in the house) and less (don't) support. But it is clearly not an optimal rule to use because the goal is to represent parties which are considered by a large number of voters and representation in the House of Commons is not the best measure of this.

To make it painfully clear why having a seat is not the best proxy for being considered by a large number of voters, consider an example. Say the Green party ran candidates in every riding, received 1,000,000 votes in the next election and failed to win a seat, but the Alberta Separation Party were to win a seat in a 2006 byelection. Would we then include the Alberta Separation party in the 2008 election debate and exclude the Greens? Even if they were only running a handful of candidates in rural Alberta? The 'must have a seat in the house to be in the debate' rule will backfire in this case by excluding a party likely to be considered by many, and including a party likely to be considered by few.

A better rule for the networks to use would be to set a threshold based on the number of votes received in the election. The most logical threshold would seem to be the same one used by Elections Canada to determine whether or not a party receives public funding:
"The party must garner either 5% of the vote in the electoral districts in which it has confirmed candidates, or an overall 2% of the national vote."

If a higher threshold is desired, then perhaps 5% of the total vote could be used, since 5% is, to my knowledge, the highest threshold used anywhere in the world for parties to qualify to receive seats under proportional electoral systems. Anything beyond 5% seems likely to produce a sub-optimal debate by excluding a party considered by many voters.

Which is a reminder that while considering the actual vote total (rather than the seat total) is an improvement, we are still just approximating what we are really trying to measure, which is how many people are planning to seriously consider voting for a certain party (or would if exposed to their views in a debate).

It sounds obvious, but one of the most important factors a voter must consider in deciding who to vote for, is which parties are running candidates in their riding. If a party is not running in a voter's riding, we can expect the impact of that party's platform on the voter's decision to be greatly diminished.

By this logic, it might seem like a reasonable criteria would be to include in the debate any party which is running a candidate in every riding since it is those parties which are worth considering by every single voter in the country.

This is oversimplifying, however. While it is true that if the Communist Party (or the Marxist-Leninst party) were to run a candidate in every riding then every Canadian could potentially vote for them, that doesn't make it true that every voter actually *will* consider voting for them. For the major parties, we can assume that a large majority of the population is willing to consider what they have to say and would be interested in hearing their platform explained/defended. For smaller parties, this is more difficult to assess. For example, given that roughly 4% of the population voted for the Green Party in the last election with next to no media coverage whatsoever, what percentage of the population is likely to want to hear the Green Party leader speak in a debate so they can hear what he has to say because they are at least willing to consider voting green in the next election? Perhaps this could be estimated by polling, but it seems reasonable to think that it has to be at least 10% of the population, and probably closer to 20 or 30%.

One other thing we know is that, given that there will be two debates in English and two in French, it is unlikely that a person would watch (or listen to) a debate which is not in their preferred language. So let us consider the proposed English language debate which will include 4 parties: The Liberals, The Conservatives, The NDP and the Bloc Quebecois. Given that the goal of the debate is to expose voters to parties they may want to vote for, does it make sense to include the Bloc Quebecois and exclude the Green Party?

# of Seats Held

Bloc Quebecois 54
Green Party 0

# of Votes Received Last Election

Bloc Quebecois 1,680,109
Green Party 582,247

# of Voters Who Have a Party Candidate in Their Riding

Bloc Quebecois 3,507,100
Green Party 13,683,570

# of English Speaking Voters Who Have a Candidate in Their Riding1

Bloc Quebecois 526,605
Green Party 9,667,647

# of English Speaking Voters Willing to Consider Voting for the Candidate in Their Riding

Bloc Quebecois = 263,032 (50% of English speakers in Quebec)
Green Party = 966,764 (10% of English speakers across the country)

Given that the purpose of the debate is to inform voters who might consider voting for a particular party, it seems clear that the choice of including the Bloc Quebecois and excluding the Green Party is sub-optimal, or more bluntly, undemocratic. A more optimal plan would be to include the Green party in both debates, or to include the Green party in one of the two English debates or to include the Green Party in one debate and the Bloc Quebecois in the other.

So to sum up a few pages of carefully constructed argument, why, why on earth do I get an English debate taking place in Vancouver with no Green Party representation but including the Bloc Quebecois when their nearest candidate is over 1,000 miles away!

Update: I thought I would clarify that I do think the Bloc should be included in the debate, I'm just pointing out that there is little or no logical reason to include them if you are going to exclude the Green Party.

Setting aside everything I have said in this post, an alternate methodology would be to start from the assumption that the debate is a service provided to viewers and simply poll the viewers on which parties they feel should be represented in the debates. This might raise some questions of the proper polling methodology to employ to get an unbiased result, but I feel these issues could be dealt with by a competent independent polling company hired to do the survey work - it would certainly be an improvement over the backroom dealing we have now. Of course, based on what polls have been taken, it seems likely this method would also lead to the Green Party being included in the debate, so maybe this is a moot point, from a theoretical perspective.

Ainge has more on this topic.

1 There are other possible goals for the debate (to provide ratings for the networks, to reinforce the hegemony of the existing parties, etc.), but I can't think of any which have any legitimacy. I mean, if the purpose is just to boost ratings, why not have a hockey game involving the leaders, or have them face off in some sort of fear-factor style series of competitions.

2 Estimated assuming 15% of Quebecers speak English as a first language, and 95% of the rest of Canada does.

Take action


  • "So the goal, or optimal result of a debate would be to expose the greatest number of people to the greatest number of potential voting options, over the long term."

    Well geez, let's haul in the leaders of all the other parties without seats too! I, for one, would like to hear (on national TV!) what the Canadian Action Party thinks about abortion, and it would be nifty to hear how the Christian Heritage Party responds to that. As a subplot, imagine the facial expression of the Communist leader while those two are duking it out.


    The only proper rationale for inclusion in a televised debate is if the party holds a seat.

    By Blogger Andrew, at 4:36 AM  

  • With all due respect, you may want to go back and read the two paragraphs following the quote you picked out (and the rest of the post).

    I know, it is long and boring, and time is precious, but if you want to understand (and refute) my point, you really need to read the whole thing (or at least most of it).

    By Blogger Declan, at 8:56 AM  

  • I like the argument that you've instructed here. It's a good one.

    However, a small number of questions pop to my mind during reading this and the whole debate regarding revolving around whether or not the Green Party should be included in the debates:

    1) When was the Reform Party first invited to the national debates? They ran 72 candidates in 1988 and received over 2.0% of the vote but didn't win a seat but came 2nd a few times. They did however win one in a byelection in 1989. Was the Reform Party included in the 1988 debates? Unless I missed something, no, they weren't. It's not a perfect comparison as they weren't a full national party yet but at the same time they were closer to winning a seat than the Greens were or currently are.

    2) Was the Social Credit Party leader invited into debates during the 1970s when they had a handful to a dozen of MPs and were getting less than 10% of the vote? I don't know if they were and I'm not sure where you can look that up. But if they weren't in the debates, perhaps the Green Party shouldn't be included. And if the Social Credits were in the debates, when they did they stop being invited? Including one party leads to so many questions I think. Should the National Party of Canada have been included in the 1993 debates given they ended up receiving 1.8% of the vote, which was higher than the Green Party in 2000.

    3) How are the debates done in European countries when there are quite a few parties running candidates. For example in the recent German election. In their nationally televized debates, how many parties were included? I am sure this could be information could be found online. But do they include all parties even ones who have no current members? Or is it only the major parties, particularly when the alliances between those are somewhat known beforehand? I think it's important to know how the process occurs in other countries that have a lot of parties.

    At the end of the day, there should be some guidelines set in advance.

    By Blogger Bailey, at 12:16 PM  

  • Some interesting questions Bailey, although I think we should do what is right rather than simply rely on what has been done before, especially considering the flawed nature of the process.

    This Green Party document answers most of your questions w.r.t to the Canadian situation.

    The Reform party was not included in the 1988 election - this was hardly comparable to the Green party since Reform had never even competed in an election prior to 1988 (and they ran fewer candidates than the Greens this time around, and they got less support).

    In 1993, Reform was included in the debate despite never having elected anyone in a general election and despite having much smaller popular support in the previous election than the Greens do this time around and despite only running candidates in sone parts of the country.

    The Bloc was also included in the 1993 debate despite never having elected anyone (Duceppe was elected as an independent in a byelection in 1990).

    Social Credit was excluded, despite holding 11 seats in 1979, a decision which clearly wouldn't stand today.

    The document also lists some provincial precedents for parties being included without holding seats.

    The German system is not quite directly comparable since usually only the major parties declare a candidate for Chancellor (the Chancellor is elected by the legislature after the election), but it does make an interesting case since one of the smaller parties declared a Chancellor candidate and tried to get him included in the leader's debates this part time around (unsuccessfully).

    So only the two front-running candidates were included (although apaprently other parties also participated in a separate round table program).

    As I said, this decision was challenged in court with some interesting analysis here.

    As it turns out, the courts rejected my opinion that the purpose of a televised debate is to inform the voters: "The Chamber again explained that the editorial concept of the program was to have two moderators question the Chancellor candidates whom the networks identified as having the greatest chance of attaining the Chancellorship. The Chamber accepted the argument that the FDP's participation would not have furthered the program's objective."

    I guess if we were to use that definition here in Canada only Harper and Martin would be at the debate (which is what Harper is on the record as preferring).

    I find it surprising that a court would accept 'marginalizing any parties which the networks don't think can win' as a legitimate use of public airwaves during an election campaign but then I'm not a German judge.

    By Blogger Declan, at 1:55 PM  

  • Here in the primitive democracy to your immediate south, we have a 15% threshold in polls leading up to the presidential debates to determine inclusion or exclusion in those debates. That is obviously too high. It shut out Ralph Nader (and Pat Buchanan) in 2000, notwithstanding that Nader and his ideas (and, to a lesser degree, Buchanan) were part of the larger debate in the electoratte, and were in a position to potentially affect the outcome even with small vote shares.

    And this leads me to why, yes, the BQ must be in your debates. Even voters outside Quebec need to know what this party wants, because it could wind up holding the balance of power, and they need to realize that a vote for the Conservatives could also be a vote for the BQ being in a pivotal position.

    What about a "viability" threshold based on a party's having won 15% of the vote in the last election in at least 5% of ridings OR polling at 5% in the current campaign?

    (I propose this having no idea whether this would put the Greens in or out.)

    By Anonymous Anonymous, at 7:58 AM  

Post a Comment

<< Home