[EM] Representative efficiency

[Richard Lung]

On 19/10/2023 12:50, Richard Lung wrote:
>
>
> Representative efficiency
>
> A nine-member constituency with a transferable vote, like Cambridge. 
> will have at least 90% efficient democratic representation of the 
> population, including its ethnic diversity and gender balance. 
> (Compare simple plurality with thirty odd or forty odd percent 
> efficiency on only a minimum of -- largely undesired -- choice.)
>
> But there are more marginal considerations. For instance the Hare 
> quota for maximum proportional representation is too high a threshold, 
> requiring some deference from voters, in smaller constituencies, to be 
> met. And the Droop quota giving a minimum PR is too low a threshold, 
> making possible chance victories.
>
> A more representative PR than these two extremes avoids either 
> disadvantage. This is provided by their average, which is the harmonic 
> mean. Hence I introduced the Harmonic Mean quota, votes/(seats + ½).
>
> The HM quota is equivalent to the Saint Lague divisor method (or 
> Webster apportionment) reported by Carstairs to be regarded as the 
> most equitable share-out.
>
> The HM quota brings the most equitable share-out to voters lists (STV) 
> as distinct from party lists. Party lists are a denial of human rights 
> by disregarding individual representation. Party lists, which use an 
> illiterate non-transferable vote, can only express division. Whereas 
> the transferable vote can express a democratic choice of coalition or 
> union, which needn’t be merely partisan.
>
> I heard from a partisan of party lists that STV is only used in a tiny 
> number of countries. Evidently genuinely democratic elections with 
> transferable voting will require overcoming a formidable herd instinct 
> that has a hold of humanity.
>
> It is possible that other averages than the HM quota may be used to 
> increase representative efficiency. And indeed I use four, in the 
> advanced version of my Binomial STV system.
>
>  Regards,
>
> Richard Lung.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20231019/a5db5cf6/attachment.htm>