[EM] Preferential Party-List Proportional Representation (PPLPR)

[Juho Laatu]

On 07 Nov 2014, at 03:19, Kevin Venzke <stepjak at yahoo.fr> wrote:

> Hi Vidar,
> 
> 
>> ----- Mail original -----
>> De : Vidar Wahlberg <canidae at exent.net>
>> À : election-methods at lists.electorama.com
>> Cc : 
>> Envoyé le : Jeudi 6 novembre 2014 3h26
>> Objet : Re: [EM] Preferential Party-List Proportional Representation (PPLPR)
>> 
>> On Fri, Oct 31, 2014 at 10:52:06PM +0100, Vidar Wahlberg wrote:
>> 
>> I'm curious, there seems to be very little interest for party-list
>> systems on this mailing list.
> 
> 
> Here are two personal reasons, purely my own feelings:
> 
> 1. I'm interested to try to identify the stance of the median voter. A single-winner
> election seems to be the most direct way to try to get this. A party list system 
> seems to delay the question for later, and may not preserve the original information.

Yes, a two-party system can measure some key questions in a quite direct manner. If party A wants more of X, and party B wants less of X, then at least within each single-winner district the message of the voters becomes very clear. If they give majority and the single seat to party A, then they want more of X.

At national level there will be some rounding errors in the sense that the total outcome (allocation of seats) may not follow the majority opinion of the whole country. It may also be a problem to a voter that if party A is in favour of X and Y, but the voter likes X and doesn't like Y, then neither party is a perfect choice for this voter, and the end result may not reflect the majority opinion.

There can be also topics where the voter wants Z, but neither party promotes that topic. That can however be fair in the sense that if majority of the voters want Z, then we can expect that sooner or later one of the two parties can start promoting Z in the hope of beating the other party. Minority opinions (with clearly less than 50% support) would not be represented at all in the representative body since doing so could take votes away from the party.

Party-list systems are used in multi-party countries. If some of the parties support X and some don't, the voters will get exact results in the form of pro-X parties getting majority or minority of the votes and seats. The result is thus exact. The more indirect nature of the multi-party systems comes into play here so that it may be that the government is not formed based on this dividing line (X) but based on some other reasons. It may be that although X has majority support among the voters, it will have minority support within the government that has majority support among the parties and voters. It is thus up to the "government negotiators" to decide which combination of topics (with wide support) will be driven by the government after the election.

I note that in Sweden there is a new tradition of parties declaring already before the election what kind of coalition they plan to build. (This approach has both good and bad points.) This is possible in Sweden mainly because the social democratic party is tarditionally very strong, and therefore automatically forms one of the alternatives (together with nearby smaller parties). In countries with numerous equal size parties this approach is less useful.

> For instance, it could be that there are no parties targeting the median voter, due
> to it being sufficient or more effective to target other voters.

I think it is characteristic to multi-party countries that to some extent all parties try to reach the median voters. The key reason behind this strategy is that the most extreme/devoted supporters of that party will vote for it in any case, and therefore the competition on who gets more votes will happen at the border lines between the parties, which means, close to the centre and close to the median voters.

A typical party structure might be such that the largest parties are quite close to the median voters, but then there are extremist outliers closer to the edges of the party space. The large parties need to fight also against those small parties that are next to them but at the edges of the space. That causes some problems in formulating their marketing message right so that they will not lose their median supporters. On the other hand the more extreme parties will anyway gladly join the coalition governments and they wll provide additional support to the more moderate major party next to them. That side is thus relatively safe to the large parties anyway.

In order not to oversimplify I note that in this typical party strcture there could be also parties that have some specific topic that they support. Also those parties tend to be close to the median (and wide too) in order to serve the needs of all different kind of voters that support their key target / party specific topic.

Juho



> 
> 2. If it's decided to use a proportional method, then there are plenty of methods
> that I'm fine with. I understand that different priorities could lead to proposing
> different proportional methods, but I just can't get excited about it.
> 
> Kevin
> 
> ----
> Election-Methods mailing list - see http://electorama.com/em for list info