There are three problems with the current system.
The quota is stated as an absolute majority, which is both unusual and makes electing multiple people (for example, in Executive Committee elections) impossible without creatively interpreting the IRPs. This is because there are only two ways of redistributing votes; (1) redistribute “left-over votes” of threshold-passing candidates; (2) if there are no threshold-passing candidates, eliminate the candidate with the fewest votes and redistribute their votes. If you interpret absolute majority correctly as a majority of all votes cast, then only one candidate at the very maximum could ever be elected under this system. Any EC would then just consist of one (un)lucky, very powerful and burnt-out person. If you ‘creatively’ interpret it, assuming that whoever wrote this section of the IRPs wanted this system to work, an ‘absolute’ majority of left-over votes creates a strange voting system where you just elect the person with the highest number of remaining votes until all slots are filled. We’re not quite sure what such a system is, other than a) objectively bad and b) a strange implementation of multi-member first-past-the-post maybe?
This is why FYEG has never actually used this system*, instead always using a droop quota, the most common quota used in STV systems. We would recommending codifying this practice in the IRPs. All STV systems that we know of use either “Hare” or “Droop” quota. The Hare quota is calculated by dividing the number of valid votes (V) by the number of seats (i.e. V / n), while the Droop quota is the number of valid votes divided by the number of electable positions plus one (i.e. V / (n+1)). The Droop quota is standard, and widely considered to be better than Hare. To illustrate that, we do not know of any political systems that actually use Hare instead of Droop (other than Hong Kong where it was introduced by the Chinese government for evil reasons). If you really want to read more about the comparisons, here is a riveting Wikipedia page on the matter.
The IRPs do not define which STV system to use, but they should. For the sake of consistency and transparency, we suggest naming which STV system to use.
Specifically, we suggest the use of the Scottish STV System. This is because (i) the rules are relatively simple (for STV, that is 👀), (ii) they produce proportional representation, (iii) the rules are comprehensively defined in Scottish Law, and (iii) it is used in the political system of at least one member organisation. For a more comprehensive explanation of Scottish STV, read this blog by Opavote. An alternative system is Meek STV. Although it’s considered “the creme de la creme of STV counting rules”, Meek is both VERY complicated and (because of that) almost never used. The only exceptions we know of are in New Zealand and some Stack Exchange forums. For more on Meek, read this.
The IRPs suggest for the gender balancing mechanism to be used in the middle of the vote redistribution, which has a distorting effect and is impossible with any technology we are aware of. This one is kind of technical, so hang in there people. Our IRPs require at least 50% of people on each body that has more than one person to “self-identify as woman, trans or genderqueer’. The way of following FYEG’s gender quoata (see box) means that the balancing mechanism happens during the redistribution. Basically, once the quote is reached (for example, 4 cis men are elected to a body of 8 people), the remaining candidates who are excluded by this quota (e.g. cis men) are removed and their votes redistributed. This is a nice idea but really complicated to execute.
There are two alternative options:
Use the gender balancing mechanism BEFORE the election. This basically means doing two separate elections - one for people who are not female, trans or genderqueer to reduce them to an admissable quota. We consider this not to match the spirit of the IRPs.
Use the gender balancing mechanism AFTER the election. We believe this is closer to the spirit of the IRPs, while being practical. This would mean we calculate the result of the election on whatever software being used, and check if the result is incompatible with the gender quota. If it is, (e.g. 5 cis men elected to a body of 8 people) we eliminate enough candidates of that gender that is incompatable with the gender quota, starting from the bottom of the list to ensure the quota is respected, before re-running the election software.
There is no tiebreaker in the IRPs. One of the quirks of any STV system, especially when there are only 100 or so votes like in FYEG, is that ties are possible. In order to avoid any possibility of ambiguity, we would recommend adding some listed way of breaking ties, which could also be used to decide between competing counter-amendments, etc. We would recommend, for simplicity’s sake, explicitly stating that a tiebreaker vote between two candidates would be held, and if this was again a tie the GA could choose to either vote again or let it be decided by chance.