Tuesday, November 15, 2016

Importance of understanding District Magnitude within #EngagedInER discussions

I find the material on the Proportional Representation Society of Australia and Australian Malcolm Baalman's On Elections site far more informative than what can be seen on the (self-called) "Fair Vote Canada" site which is seen as Canada's primary promoter of Proportional Representation.

Once people move past thinking that Proportional Representation is a description of what they want, rather than only a description of what they don't want (single-member electoral districts), we can move to discussing some of the more critical questions.


It is this last question I think we need to discuss more than I've seen thus far.  From the PRSA:

DEFINITION: District magnitude is the number of representatives that voters in a particular electoral district are entitled to elect. In systems that use single-member electoral districts, the district magnitude is 1. At the other extreme, with party list systems, in which MPs are not directly elected, as in Israel and the Netherlands - where the whole nation is a single lower house electoral district - the district magnitude can equal the number of MPs in that house.

People have done the math on optimizing parliamentary representation.  As one example The Principles of Parliamentary Representation from 1884 suggested a mathematical optimal DM was 6.  The PSRA suggests that more modern analysis recommends an odd number be used.  (And yes, that is 1884 not long after Canada became a country in 1867, a reminder of just how out-of-date the First Past the Post system some Canadians seem to be nostalgic about really is.)

I do not, however, believe that mathematics alone should be determining the DM, and other regional factors should be included.

I grew up in Sudbury, which is in Northern Ontario.  While I moved to Ottawa when I was 19, I have kept family and other ties and remember the political climate as I was growing up.

Northerners don't trust the political views of southerners, and don't believe that they can represent them.   While I was on the "yes" side of the Ontario electoral reform referendum in 2007, I became quite aware that Notherners were strongly on the "no" side.  They believed that MMP would dilute their influence, both by increasing the size of ridings (they believed that the extra seats granted to Northern Ontario would be removed by the system) as well as their belief that "Southern Ontario political parties" couldn't represent them.  They didn't see the possibility that those 30% party seats could represent Northerners, so saw this as a Southern Ontario power grab.

Whether I agreed or disagreed with those sentiments doesn't matter.  What this did suggest to me was that if the government is to be seen as legitimate for the entire country that these regional differences must be accommodated.

A District Magnitude Proposal

An ideal system might be one where the District Magnitude is determined by the districts themselves.  District boundaries would be decided by Elections Canada as it is now, but whether neighboring districts would bind themselves to create a larger district magnitude would be locally decided.

The BC STV proposal had the Electoral Boundaries Commission decide the boundaries for the province-wide single member system (still First Past the Post) as well as the multi-member proposal (STV, but  with varying rather than fixed district magnitudes).  Under the BC STV proposal, some districts would remain single-member districts but with the use of the ranked ballot being province-wide.

With all the misinformation presented by Fair Vote Canada and others about single-member ranked ballots, fair discussion of a system which includes a mixture of single-member and multi-member districts becomes much harder.  Unfortunately, without allowing for the possibility of districts which for local reasons don't want to be multi-member we will always have disagreements.

I believe leaving the district magnitude up to the districts themselves would avoid this problem.  In larger urban areas, the areas that tend to most favor PR, the district magnitude will be high.  In remote or rural areas where the districts represent a massive geographic region, the district magnitude will be lower.  I do not believe this is something that urban areas should be imposing on rural/remote areas.

With this context we can more reasonably decide if we want to put restrictions on local district organizing.
  • Is a district magnitude of 1 disallowed, or only discouraged?  I believe it should be allowed to accommodate the interests of rural/remote areas, but this appears to be divisive.  I am curious to hear from someone from outside of a urban area that wishes to disallow this option?
  • Should district magnitudes being odd be enforced, or only encouraged?

Critical differences in forms of proportional representation

While not the primary topic of this post, I want to ensure anyone who supports PR is aware that there are wide differences in systems which fall under that umbrella.  PRSA has the following to say about STV, known there as Hare-Clark:

HARE-CLARK: Proportional representation using the single transferable vote is the only form of PR where the person elected is directly elected by the voters. Therefore the voters have  ultimate control over who is elected, as in the Tasmanian and Australian Capital Territory Assemblies, which use the superior Hare-Clark electoral system. In Australia's upper houses, however, political parties' self-serving overlays of stage management and regimented voting arrangements have been imposed, in a fairly effective attempt to reduce that control.
 For those concerned with the divisive nature of the recent US presidential election, they should be aware that it is the addition of indirection between voters that is at the root of much of the ongoing controversy.  In the US system they have an Electoral College for electing their President which in most states is non-proportional, and is stage managed to protect the bipartisan interests of the two large political parties.  Introducing political parties onto a Canadian ballot would bring us closer to the flaws we see within the US system by introducing an indirection between voters and representation in parliament.  While this additional indirection between voters and parliament is a feature of many forms of PR such as MMP, it is not a feature of STV.

We have the additional issue in Canada which is that we don't have a separate executive branch voted indirectly by citizens, but a system where members of parliament vote in the government.  In a well functioning Westminster system it is possible for the Prime Minister and/or the government to change without forcing a general election, offering more stable government.  Giving political parties seats in the House of Commons means there is yet another level of indirection from voters where parties rather than voters have additional influence.

While some organizations involved in electoral reform are only concerned about the fate of political parties in elections, there are other people who consider the influence of political parties on our democracy to be a critical problem that needs to be addressed. (See: On Elections: Elections should serve voters, not parties).

No comments: