STV inspired one Dunedin mayoralty candidate to film a video featuring alpacas.
Another shared a graphic showing what happened to votes in 2019.
One candidate said, succinctly: "Under the Dunedin STV computer voting system, please only mark candidates you really want and leave the rest blank."
Advice flowed — both good and bad.
In the local body elections this month, the Otago Regional Council was using STV, or single transferable vote, for the first time.
It meant voters across the region were writing numbers on their ballot papers, ranking candidates they wanted to represent them, in order — as few or as many as they liked — 1, 2, 3 ... first preference, second preference, third preference.
The Dunedin City Council has used the system since 2004.
This year, the results included a mayoralty win for Sophie Barker and six new councillors being elected, joining eight who were re-elected.
There was, of course, commentary about the system before the elections and discussion continued on afterwards.
The Otago Daily Times asked two experts for further input.
Prof Andrew Geddis, from the university’s law faculty, is a leading commentator on New Zealand electoral law.
Regarding the Dunedin mayoralty result, what is the best way to describe the winner under STV?
Prof Geddis: The winning candidate will always be the one supported by a majority of remaining votes, after discounting all the other candidates that cannot achieve this status. Or, alternatively, they are the least-disliked candidate in the field.
Prof Hayward: The winner was the candidate who was the first to reach the required 50% support from the votes that remain in play.
Does STV do a better job than first-past-the-post (FPP) of electing a council with broad representation? If so, what does this mean and how is it achieved? Is the system helpful for minority interests?
Prof Hayward: Yes, STV does a better job than FPP of electing a council that reflects how the community voted. Under FPP, you cast multiple votes for multiple candidates. So one group of voters, who could be the largest minority, could function as a bloc of voters and elect all the candidates to represent them, leaving the majority of the voters with no candidates at all to represent them. STV was designed to fix that problem. STV gives you one vote, so under an STV election, that large minority bloc of voters would elect one or more candidates to represent them (depending on the size of the bloc), and the rest of the community would also get their chance to elect candidates to represent them. In that way, STV ensures that most of the community who vote in the election get a say in electing someone to represent them.
The Dunedin City Council has all its councillors elected at large, or from one pool across the whole area. The Otago Regional Council has several constituencies, or wards. Does STV function more effectively in either situation?
Prof Hayward: The larger the group of candidates, the more effectively STV can reflect the wishes of the whole community. Currently, when electing 14 candidates in one ward, only 6.7% of votes will be wasted (will not count to elect anyone in the final iteration). If Dunedin moves to two wards of seven vacancies in each, for example, that might make it easier for voters by reducing the number of names on each ballot, but it will mean that 12.5% of votes will be wasted in the final iteration, which reduces the effectiveness of STV as a proportional system.
Prof Geddis: The benefit of an "at large" election is that every voter has the opportunity to register their preference in relation to every candidate. The down side is, as we saw at this last election, that you can get very large fields of candidates that are difficult for voters to work through. And that may lead voters to default to ranking names and faces that they have heard of, rather than considering each candidate on their merits.
Some people suggested ranking one mayoralty candidate only. For most voters, is this bad advice?
Prof Hayward: The best advice to voters is to vote for the candidates you would like to see elected. If you only like one mayoral candidate, then vote for that one candidate. If you like more than that, rank more than that. If your vote is not needed to elect your No 1 candidate (because they are not elected), your vote will transfer to your No 2 candidate. STV ranking is an insurance policy to keep your voting power alive until one candidate receives more than 50% support and gets elected.
When ranking candidates for the city council, is it better to think of this as voting for a person, rather than people?
Prof Hayward: The STV vote is one vote, but it may be split to support more than one candidate. For example, when Mr Simms was elected to council on the first iteration, he had almost three times the number of votes he needed to reach the quota and get elected, so about two-thirds of each of his votes then transferred to his voters’ second preferences (if they had ranked a second preference). In that case, voters supported more than one candidate with their one vote.
Prof Geddis: When you give someone a ranking, you are saying "I want this person on council ahead of everyone I rank below them [and it will kick in] only if everyone I’ve ranked higher is either already elected or eliminated from the vote count".
Voters should not worry too much about where they might allocate their 11th and 12th preferences, should they?
Prof Hayward: Voters shouldn’t worry about anything when they are voting using STV. There is no secret strategy required — just rank the candidates you like in the order you like them.
What is the No 1 misconception about STV that you would like cleared up?
Prof Geddis: That ranking a candidate lower than your more preferred candidates somehow might help that less-preferred candidate get elected ahead of your more preferred ones. This can never happen.
Prof Hayward: That there is a complicated strategy required to cast an STV vote. The electoral system was designed to remove the need for strategy that was [so necessary or prevalent] in FPP elections. Vote honestly — rank as many or as few candidates as you like, in the order you like them. When the results are announced, you have a very high chance of seeing someone you ranked get elected.
