Legal Beagle: Council Elections: STV Q&A
125 Responses
First ←Older Page 1 2 3 4 5 Newer→ Last
-
Ben McNicoll, in reply to
The problem with an MMP approach at local govt level, as I see it, is that if the party lines are explicitly the same as national (small n) political parties (and they would almost certainly have to be to be elected), you'd end up with a situation where each local government is either (A) implicitly subservient to, or (B) completely in opposition to the central government depending on who's in power at the time.
In both of those scenarios the ability of local govt to chart their own course which meets the needs and desires of the local populace is hugely undermined.
The already strained relationships and differing visions in both Christchurch and Auckland would disappear, yes, but only because disagreeing would become impossible for the local council.
The semi-fiction that the mayors are not explicitly aligned with either major national political party allows for some face-saving, and helps the two levels of government work together when their philosophies or goals don't align.
ETA: OK, that last para sounds a bit naive, perhaps, but imagine how the Central Rail Link decision would have gone down, were Len Brown explicitly a Labour mayor.
National couldn't have attempted to head it off as an electoral issue by agreeing to it, as it would have been seen as a Labour policy win. So Auckland wouldn't have got it until Labour was in government (which admittedly is still the case, but at least now the argument is over when not if the CRL will go ahead).
-
Sacha, in reply to
wise observation
-
Steve Todd, in reply to
Yes, exactly Paul, although I would drop the words "in some sense". That is one of the many reasons, contrary to Rich's view, why STV is the very best electoral system yet devised by humankind. And we in New Zealand are so lucky to have the very best, up-to-date, computer-compatible version of that very best system.
As asides, I, too, am pleased the DCC has gone with random candidate-ordering. And, a recount in the Central ward last time would not have changed the final result, assuming all the voting documents were input correctly (which they would have been). That is one of the beauties of computers - if you input the same information, repeatedly, you will get the same outcome, repeatedly.
(I have seen the votes being scanned/input into the computers (at a trial), and the process is very rigorous. That lady who was the runner-up in Central last time, was too far adrift of the 10th and 11th winners. For her to have got up on a recount, would have required that it was found that some 40+ votes were originally miss-entered, and that simply would not have been the case.)
-
I am a great supporter of STV. And elections. If we didn’t have Health Board elections they would be comprised entirely of Tony Ryall’s appointees. Health Board elections were stopped under the last National government’s round of health reforms and the boards were appointed – no local democracy at all (in fact Simon Upton sacked the lot in 1991 with about two weeks’ notice). If you remember it was not a happy time for the heatlh system, with lots of cut backs, user charges, poor wages, skills shortages and big computer disasters. Bringing back elections for some board members was an election promise of the incoming 1999 Labour government. So re democracy – use it or lose it.
I also like STV. STV favours a left block which is likely to be more amorphous – the winner is likely to be people’s general first, second or third choice, rather than the first choice of only a few who have staked all one candidate. If Auckland had STV in 2007John Banks would never have been elected.
DHB elections have a lot of candidates because there are only 20 DHB regions – ie electorates – in the country. FPP requires as much work for the informed voter – you need to read the candidates’ bios and decide who you like. You either tick up to that number (in FPP) or rank (STV). STV is a much fairer consensus building voting system so a little bit more effort required for ranking is not much to ask.
-
BenWilson, in reply to
The problem with an MMP approach at local govt level, as I see it, is that if the party lines are explicitly the same as national (small n) political parties (and they would almost certainly have to be to be elected), you’d end up with a situation where each local government is either (A) implicitly subservient to, or (B) completely in opposition to the central government depending on who’s in power at the time.
True, but there is no need for them to be those parties. And MMP does seem to apportion actual power in a way that is more proportional than STV. But the low number of candidates in local body compared to national level elections means that the proportionality is unlikely to translate into anything anyway (a party getting 5% is going to get no representation, most likely), so STV seems to me a better option for local government.
Representative democracy generally seems to work better at that level, there really is a care factor about the level of local interest the candidates have. People are more likely to vote on an issue or a cluster of issues, than on a general ideological alignment, because the issues will be what actually happens, and the general ideology of the elected groups will get no air whatsoever. However, I doubt that actual vote turnouts will rise much under any system.
-
Ben McNicoll, in reply to
True, but there is no need for them to be those parties.
Yes, it is true that there is no need for them to be those parties. But I would suggest that any new parties forming at local level would be at a serious disadvantage in name recognition and tribal supporters/organisers/activists, and would thus be far less likely to be successful against a "name brand".
-
BenWilson, in reply to
If that were true, then why are they not called Labour and National now?
-
Sacha, in reply to
a party getting 5% is going to get no representation, most likely
That's one seat on a 20-person governing body
-
Graeme Edgeler, in reply to
Given the esteem with which the Electoral Commission is regarded both nationally and internationally, I think we’re probably OK to have that dependence. I’d be more concerned if it was the appointed overlords of ECan who where running the regional elections.
The Electoral Commission plays no role in counting the votes for local body elections. It's role is limited to providing electoral rolls.
-
Rich of Observationz, in reply to
There *are* parties in local government.
In Auckland, STV results in ad-hoc alliances. City Vision is an alliance of the local Labour, Green and Alliance Parties with community independents. Similarly, NACT form "Communities and Residents" (the former CitRats) but aren't as upfront about this.
In Wellington, we have STV and Labour and the Green party endorsing candidates, and NACT trying to pretend that their people (Kerry Prenderghastly, for instance) are *independents*. We also have a mayor who played on being a Green Party member, but wasn't endorsed by the party and has been notably ambivalent on ungreen polices like motorways and runway extensions.
I'd prefer to just be able to vote Green, get (if we succeed) a Green/Labour council that implements a policy programme associated with those parties.
-
BenWilson, in reply to
I’d prefer to just be able to vote Green, get (if we succeed) a Green/Labour council that implements a policy programme associated with those parties.
You can do that under STV. If Greenness is your bag, you rank the candidates on that criterion. Indeed, you can rank within the Green candidates themselves, ordering on your particular brand of Green. Maybe you're about public transport far more than you are worried about the rights of trees.
-
Graeme Edgeler, in reply to
As asides, I, too, am pleased the DCC has gone with random candidate-ordering.
At the last (or maybe last but one?) local body election, the Wellington councils each used different methods. One went for alphabetical, one for a single random ordering, and one for multiple random orderings (i.e different on each? ballot)
-
Steve Todd, in reply to
Yes, I think the Hutt Valley DHB used the incorrectly named pseudo-random ordering of candidates (same random order on all voting documents), and the regional council used alphabetical ordering. It's hard to find these things out without going to a lot of trouble.
Wellington City and Capital & Coast DHB use random candidate-ordering (different random order on each voting document), and hopefully the regional council will, too, this year, now that it has adopted STV.
-
Steve Todd, in reply to
>>> DHB elections have a lot of candidates because there are only 20 DHB regions – ie electorates – in the country. FPP requires as much work for the informed voter – you need to read the candidates’ bios and decide who you like. You either tick up to that number (in FPP) or rank (STV). STV is a much fairer consensus building voting system so a little bit more effort required for ranking is not much to ask. <<<
Yes, in 2001 (under FPP), there were 1,047 DHB candidates nationwide. In 2004 (STV) there was half that number and, in 2007, half that number again. I don't recall the 2010 figures, but I don't expect Isabel Hitchings' task to be too onerous this year.
-
With a mayoral STV, why are the lowliest second votes counted first? What if the top two candidates were very close, but the centre left had just edged out the centre right option, and the more extreme lefty was third and extreme righty was fourth. By taking fourth places' 2nd votes it could tip the second choice into the mayoralty despite the majority of voters choosing left wing politicians.
Unlikely I know.
-
Graeme Edgeler, in reply to
By taking fourth places’ 2nd votes it could tip the second choice into the mayoralty despite the majority of voters choosing left wing politicians.
Unlikely I know.
Not possible. Someone has to get over half the votes. If left supporters all rank the centre-left candidate above the centre-right candidate, and left supporters are more than half the votes, then the centre-right candidate cannot get more than half the votes, so cannot win.
In reality, some left supporters will vote for the centre-right candidate, but if it's the example you give, the right can't win. There can be vote counting anomalies, but this isn't one.
-
To expand a little on Graeme's response:
With a mayoral STV, why are the lowliest second votes counted first? What if the top two candidates were very close, but the centre left had just edged out the centre right option, and the more extreme lefty was third and extreme righty was fourth. By taking fourth places' 2nd votes it could tip the second choice into the mayoralty despite the majority of voters choosing left wing politicians.
In that case, votes for the left are higher than votes for the right (1st > 2nd and 3rd > 4th, so 1st + 3rd > 2nd+4th) thus, even if all those that first-voted for the 4th candidate 2nd voted for the 2nd candidate, the total votes for the 2nd candidate would not increase to more than half the votes. Thus, the 2nd choice of voters that voted for the 3rd candidate would also be considered, meaning the 1st candidate in your scenario (center-left) would be elected.
-
One thing to think about here is that the system is STV "single transferable vote" - you only have one vote, you don't have a second vote if the first vote has already been stuck to a viable candidate.
In a mayoral (or any race with one winner, much simpler than a multi-councillor ward) what happens is:
1) we count all the first votes
2) assign them to candidates
3) remove the candidate with the lowest number of votes
and redistribute their votes to the next candidate on each of their voter's
lists (if the voter's list is empty discard the vote)
4) repeat step 3 until only one candidate is leftIn the end all the votes lie with two candidates , one of whom holds more than 50% of the still viable votes - in a close 3-way tie it's quite possible that the 3rd place winner on the first round still wins the election. STV always results in a mayor elected with votes by more than 50% of the voters (who voted the full list)) - though he or she may not be the first choice of all those voters.
Myles: as Graeme points out if at any point during the counting any candidate ends up with 50%+1 of the viable votes they are going to win anyway as no redistribution of the rest of the votes can possibly beat them.
-
Functionally we do already have political parties at local body level, most of whom are directly tied to central government parties. The main difference is how open they are: unlike the communists, the right would appear quite happy to conceal their views and aims, whereas the left (broadly) appear to be perfectly happy to so do.
Arguably, Ben McNicoll, the pretense that parties are not involved at a local level makes it far easier for central government to stymie local government, as they can pretend they are acting for “objective” “non-political” reasons. ECan, for instance, would have been much much harder to pull off it was explicitly an attempt to oust a Labour council, say.
-
Rich, from what I understand, the Auckland supercity enabling bill expressly did not allow for STV in Auckland Council elections, so the current state of that council is attributable to FPP in local council elections. Part of the problem is that the electoral reform movement gave up after achieving and then defending MMP at central government elections, given that New Zealand is a centralised, unitary state with a negligible history of devolution.
-
Steven Peters, in reply to
I agree Leslie - very good points. the same ideologies that drive national level politicians and politics, operate at the local level - but a bit more 'down home and folksy'. It would be better if candidates and local body 'groupings' were more transparent in their allegiances. However, the convoluted STV voting system would not enable voters to divide power in ways that reflect the diversity in local communities, particular in providing representation to marginalized groups. But maybe that is its real
purpose. -
Brent Jackson, in reply to
But maybe that is its real purpose.
Its real purpose is to provide a good voting system, where "good" is defined as meeting the fairness criteria of voting systems.
Note that Arrow's_impossibility_theorem proves that no voting system can meet all the criteria, so every voting system will have advantages and disadvantages depending on which voting criteria are deemed more important.
Interestingly, the criteria used by the Royal Commission on the Electoral System were quite far removed from the theoretical "fairness" criteria. Also, none of the 5 recommendations of the Commission were actually implemented.
-
I have been reading through your October 2010 “STV Q&A” and the responses it generated. In an exchange with Paul Campbell regarding the outcome of the 2007 mayoral election in Wellington City, you stated that the incumbent, Kerry Prendergast, “was declared elected without having an absolute majority of votes cast.” This is not so. I have not seen any posts since then, in which you inform us that you now realise your error, so here is my explanation of the correct position.
Section 5B of the Local Electoral Act 2001 (Reprint as at 26 July 2004), referred to by Paul, only gives a “*General* Description of Single Transferable Voting”. The definition of “absolute majority of votes” is set out in Part 2 of Schedule 1A of the Local Electoral Regulations 2001 (Reprint as at 26 July 2004), clause 26, being, voting documents less non-transferable votes, divided by 2. Where the resulting number “is not a whole number, it is rounded up to the next whole number.”
Clause 31 states, “A candidate is elected if his or her votes equal or exceed the absolute majority of votes.” But, clause 32 states, “If no candidate is elected, repeat the operations in clauses 26 to 31.” In other words, the “absolute majority of votes” is re-calculated, taking account of the (new) total of non-transferable votes at each subsequent iteration of the count. This has the effect of reducing the absolute majority of votes as the count progresses.
At the WCC mayoral election referred to above, 51,328 valid votes were cast, meaning the initial “absolute majority of votes” was 25,664. No candidate’s votes equalled or exceeded that figure, so the lowest polling candidates were successively excluded from the count (and their votes transferred to the next preferences indicated by the relevant voters) at each of the next seven iterations. At the ninth (final) iteration, the total of non-transferable votes was 8,436. Therefore, the final “absolute majority of votes” was (51,328 less 8,436) = 42,892 / 2 = 21,446.
At that point in the count, Prendergast had 21,868 votes and was duly (and properly) declared elected. The two other candidates still in the race, following the exclusion of Helene Ritchie at iteration 8, Ray Ahipene-Mercer and Bryan Pepperell, had 10,899 and 10,125 votes, respectively. There was no need to transfer Pepperell’s 10,125 votes, because (1) Prendergast already exceeded the absolute majority of the votes *remaining in the election* (21,446), and (2) because, even in the extremely unlikely event that all of Pepperell’s votes had transferred to Ahipene-Mercer, his total of votes would still only have been 21,014 – well short of Kerry’s 21,868 votes (and of the 21,446 votes needed).
So, to be clear, Kerry was properly elected, because she attained a number of votes that represented 50.98% of the 42,892 voting documents on which a preference for her over Ahipene-Mercer and / or Pepperell was indicated. The 8,436 voters who had not indicated a preference for any of the last three remaining candidates, quite properly played no part in the final outcome. By *partially* abstaining from the election in that manner, they had, in effect, said that they had no interest in those candidates, and were therefore quite happy for those voters who remained in the election to determine which one of them would win. That being the case, the absolute majority of votes needed to win the election was reduced, accordingly.
I notice that Paul appears to continue to believe that the “absolute majority of votes” is the absolute majority as calculated at the first iteration (see his reference to “voters (who voted the full list)” 3 days ago). As I hope I have shown, that is incorrect. The absolute majority of votes could better be seen as a figure that is “50%+ of the votes remaining in the election at any given iteration.”
In conclusion, under NZ STV, no candidate is ever elected without attaining the final absolute majority of votes in single-seat elections, or the final quota in multi-seat elections.
-
Graeme Edgeler, in reply to
So, to be clear, Kerry was properly elected
No dispute from me.
-
Steve Todd, in reply to
>>> However, the convoluted STV voting system would not enable voters to divide power in ways that reflect the diversity in local communities, particular in providing representation to marginalized groups. But maybe that is its real purpose. <<<
Oh dear, Steven, how wrong you are. I think you need to do some reading, my friend. For starters, allow me to refer you to the papers mentioned in my first post, at page 1, above; the second paper, in particular. The bits I would particularly like to make you aware of are reproduced, below--
>>> (c) The sorting of voting papers according to first preferences in effect arranges the electors who voted into generally unequal groups, each group supporting a single candidate. The transfers of surpluses and exclusions reduce the groups in number according to the number of places to be filled and make the initially unequal groups each approximately equal to a quota. The electorate is thus
arranged into the desired number of nearly equal opinion groups, each group with its own representative.>>> (d) Nearly every vote is effective in helping to secure the election of a chosen
candidate. The percentage of effective votes in an election is a measure of voter
satisfaction, and thus of the validity of an election method. In an election for seven places in any given district, the proportion of effective votes is about seven-eighths, or 87%. Under multiple-FPP, effective votes are routinely less than 50%. Nearly every elector who votes has an equal effect on the result and is directly represented by someone whom he or she has helped to elect.>>> In voting, different electors may attach different weight to several criteria simultaneously.
>>> The single transferable vote gives proportional representation of this opinion structure of the electorate with an accuracy dependent only on the number of representatives simultaneously elected. The single transferable vote gives freedom of choice to electors and ensures, as far as possible, that that choice is satisfied and not distorted or frustrated. <<<
In other words, the more vacancies being filled (within reason, but that's another issue), the more "desired number of ... opinion groups" get to elect their own representative, thereby reflecting, as you say, "the diversity in local communities", particularly with regard to "marginalised groups."
Now, even if the part of your comment I have reproduced above, merely reflected a level of cynicism on your part, regarding the effect of STV in local elections, do you really think the STV electoral system would have finished second behind MMP, in the opinion of the 1986 Royal Commission on the Electoral System, if it did not "enable voters to divide power in ways that reflect the diversity in local communities"? I don't think so.
Post your response…
This topic is closed.