Posts by Steve Todd
Last ←Newer Page 1 2 3 4 5 Older→ First
-
Legal Beagle: Council Elections: STV Q&A, in reply to Rich of Observationz,
I think the latter is more likely, too – concentration of resources.
In the Lambton Ward, for example, the election is wide open. Labour does not know with any certainty that its candidate, Mark Peck, is going to win one of the two vacant seats, despite the fact he is a former MP.
But, to win two seats in a 3-seat ward, requires a party receiving 50% of the votes (quota 25%). Therefore, if Labour had put up two candidates, it would have risked some of its vote “leaking” away from the running-mate (say, Peck), to other candidates, when the other Labour candidate was excluded from the count (as that other candidate surely would be). Although most votes would transfer to Peck, the leakage that would also occur could just be sufficient to deny Peck the third seat (in a tight contest). Too risky. Better to concentrate resources.
Regarding the Greens, it looks like Iona Pannett has a lock on her seat. Her final keep value in 2010 was 70.6% (up from 90.5% when first elected in 2007), so she should comfortably be re-elected again. She has worked hard on local issues in the ward, both before and after her 2007 victory, and now has a high profile.
But, again, there is no way a lesser-known Green candidate, such as David Lee (who I understand lives in Lambton, but is standing in Southern, meaning Sarah Free, who lives in Southern, had to stand in Eastern), would pick up a quota of votes in addition to Iona. Leakage away from Iona upon Lee’s exclusion would not be the issue for her that it would be for Peck, but why waste resources on something that’s not going to happen?
You might think I’m contradicting myself, having said in my previous post that, in Ireland, parties usually put up one candidate more than the number of seats they expect to win. But that is in a national party-based polity where the two main parties (Fianna Fáil and Fine Gael) have traditionally known with near certainty how many seats they will win in each electorate; it just being a question of which candidates will fill them. (The apple-cart was thoroughly upset in 2011, but that’s another story.)
That is not the situation in Wellington City. While Iona is almost certainly safe (in my view), her seat is not necessarily a safe seat for the Greens. Were she not to stand in 2016, there is no guarantee her vacancy would be filled by the Green candidate replacing her.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Brent Jackson,
It’s all to do with the differences between the percentages of votes received and the percentages of seats won in any given electorate, where the elections are party-based, Brent.
To explain, in a 5-seat electorate, if a party receives 40% of the votes, it will win two seats (40% of the seats), meaning the ‘Index of Proportionality’ for that party in that electorate is 100.0. (The quota is 16.67% - 100 / 6 – low enough for the party to win two seats, but not three.)
But, in a 4-seat electorate, 40% of the votes is enough to win two seats (50%), meaning the index of proportionality is 125.0 (50 / 40 = 1.25). (The quota is 20% - 100 / 5 – low enough for the party to win two seats.) That means the party is over-represented by 25%. I’ll call this Example 1.
Example 2. In a 3-seat electorate, 25% of the votes is enough to win one seat (33.3%), meaning the index of proportionality is 133.3 (33.3 / 25 = 1.33). (The quota is 25% - 100 / 4.) That means the party is over-represented by 33.3%.
In a 4-seat electorate, 25% of the votes is still only enough to win one seat (25%), but the index of proportionality is now 100.0. As you can imagine, other examples could be set out showing similar relationships between votes and seats for DMs of 7 >> 6, 5 >> 6, or whatever.Now, in Ireland, in the bad ol’ days, the party in power drew the constituency boundaries in a manner that would enhance its prospects at the next election. For example, if the party knew it had 40% support in a particular part of the country, it would create a 4-seat constituency out of a current 5-seater, to increase its index of proportionality from 100, to 125. (Example 1.)
Also, in those parts of the country where the party in power knew the main opposition party had 25% support in a 3-seat constituency (index 133.3), it would create a 4-seat constituency from that current 3-seater, to decrease the opposition party’s index to 100.0. (Example 2.)
When the opposition party next came to power … well, you guessed it. (Ireland now has an independent Constituency Commission – thank goodness.)
The differences in the indexes (i.e. between over-representation, or under-representation) in odd-numbered electorates is less than in even-numbered electorates. For example, in a 5-seat electorate, if a party wins two seats (40%), the index in respect of 33.33% of the votes received (two quotas) is 120.0 (40 / 33.33 = 1.20); for 37.50% of the votes, the index is 106.7 (40 / 37.50 = 1.07); for 40% of the votes, the index is 100.0; for 42.86% of the votes, the index is 93.3 (40 / 42.86 = 0.93). The difference between the high index and the low index for these vote percentages is 26.7 (120.0 – 93.3 = 26.7).
In a 4-seat electorate, if a party wins two seats (50%), the index in respect of 40% of the votes received (two quotas) is 125.0 (50 / 40 = 1.25); for 42.86% of the votes, the index is 116.7 (50 / 42.86 = 1.17); for 50% of the votes, the index is 100.0; for 57.14% of the votes, the index is 87.5 (50 / 57.14 = 0.88). The difference between the high index and the low index for these vote percentages is 37.5 (125.0 – 87.5 = 37.5).
Therefore, it can be seen that odd-numbered electorates flatten out the proportionality indexes for any given number of seats, over varying percentages of votes received, affording fairer representation for the parties that win seats in such electorates.
This is no doubt one of the reasons why the Royal Commission recommended that, were STV to be adopted in New Zealand, 80% of the electorates should be 5-seaters. The other 20% would not exceed seven seats, or have fewer than 3 seats. It is to be noted that one or two 4-seaters, or one or two 6-seaters, were not precluded from this variation from the 5-seat norm. For example, under STV, the South Island might well have been allocated 26 MPs, either as result of this year’s census, or following the 2018 census. That would likely have meant a 1 x 6-seat electorate and 4 x 5-seat electorates.
However, having said all that, there are situations where even-numbered constituencies are provided for, regardless. They are usually seen in polities where the boundaries are unchangeable, such as for the Australian Senate where the state boundaries are fixed, and in respect of the 108-seat Northern Ireland Assembly, where six members are elected from each of that province’s 18 single-seat FPP Westminster constituencies.
None of this has much to do with STV elections at the local level in New Zealand. The National Party does not stand candidates, and Labour and the Greens appear to do so somewhat half-heartedly. As I’ve pointed out somewhere up-thread, only 18% of local authority candidates are party aligned (presumably including various local groupings that often come and go with each election cycle). Therefore, there is really no reason why STV wards and regional council constituencies cannot have a mixture of odd- and even-numbered seats, as best suits the representation requirements of each council. In a basically non-party polity (by choice of the small-n national parties), even-numbered seats are not an issue.
-
Thanks for this posting, Graeme.
As you imply, to get a more accurate idea of which of the serious candidates is preferred by a majority of likely voters, two-candidate preferred (TCP or 2CP) polling is required.
The questions could be something like these—
Q. If the [name of territorial authority] mayoral election was held today, to which candidate would you probably give your first preference vote? If not sure, which candidate are you currently leaning toward?Q. If your first preference candidate were to be eliminated during the count, to which other candidate (your second preference choice) would you wish your vote to be transferred? (If the respondent can’t answer, no matter.)
Q. If your second preference candidate were to be eliminated during the count, to which other candidate (your third preference choice) would you wish your vote to be transferred? (If the respondent can’t answer, no matter.)
NB. Respondents who answer ‘don’t know’ to the first question are not included in the results.
Given that this type of polling would be a first for New Zealand; that we’re talking local elections here, and that, consequently, most people approached “on the fly” would have trouble giving more than two preferences, little would be served by pushing on and asking a fourth question.Once the pollsters have gathered the voting intention data, they then perform a notional count, through as many iterations as are necessary, to determine the last two candidates standing (even if one candidate has attained an absolute majority of votes before the other most-preferred candidate has been identified). The total of votes given for each of the two remaining candidates is then expressed as a percentage of the total of votes remaining in the count, to produce the 2CP result.
For people who knew what they were doing, a (manual) hand-count would not be too onerous – say, 750 people polled; up to just three preferences given? Easy. However, for the record, there is at least one computer program available (ahem).
The poll result would be reported in two parts: (1) the first preferences given for each of the candidates – say, 750 in total; (2) the 2CP result based on the lesser number of people who expressed a preference for one or other of the two remaining candidates (that number being the total of votes remaining in the count).
There’s no reason why it couldn’t be done.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Steven Peters,
Following on from Graeme’s reply, I played around with some figures a couple of years ago and worked out that if STV was being used to elect Parliament now (2008 and 2011), there would be something like 27 South Island MPs, 80 North Island MPs and 13 Maori electorate MPs.
Based on the Royal Commission’s recommendation that the average district magnitude (DM) of the electorates should be 5, there would be 24 multi-member electorates. The 27 SI MPs could be elected from one 7-seat (Chch) electorate and four 5-seat electorates. The 80 NI MPs would likely be elected from 1 x 7-seat electorate (in Auckland), 14 x 5-seat electorates and 1 x 3-seat electorate. The 13 Maori electorate MPs would likely be elected from 2 x 5-seaters and 1 x 3-seater. The 3-seat electorate would take in the bottom third or so of the North Island and all of the South Island.
The Royal Commission envisaged the separate Maori electoral roll would be discontinued, but that is not yet likely, even though STV would work best (for everyone) if the separate Maori seats were abolished. If they were abolished, there would likely be (today) 92 NI MPs and 28 SI MPs. The 92 NI MPs could be elected from 1 x 7-seat electorate and 17 x 5-seat electorates. The 28 SI MPs could be elected from 5 x 5-seat electorates and 1 x 3-seat electorate.
In Ireland, the parties usually put up at least one candidate more than the number of seats they expect to win in any given constituency. Therefore, National and Labour would likely put up three candidates in a 5-seat electorate (four, if one party knows it will win three seats), and four candidates in a 7-seat electorate.
STV (in Ireland) has been described as “candidate-centred but party-wrapped”. Whereas voters vote for individual candidates, they do so predominantly on a party basis. While near-precise party proportionality is by no means guaranteed, even in Ireland, where the district magnitude is only 3.86 (166 / 43)*, party proportionality is usually remarkably close. It cannot be as near-precise as in party list-based systems, if only because, under STV (in Ireland), proportionality is determined in 3-, 4- and 5-seat constituencies, whereas under MMP, proportionality is determined across a 120-seat nationwide constituency. The trade-off is, of course, that under STV, voters get to construct their own lists (by rank-ordering candidates both within and between the parties, and including independents in the mix), whereas under MMP, they have to accept the ordered lists the various parties present to them, then opt for one of those lists.
* At the next election (expected in 2016), there will be 158 TDs (MPs) elected from 40 constituencies, giving an average DM of 3.95.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Rich of Observationz,
Rich, local elections in NZ are not an example of a 'no party' system. If the political parties that operate at the national level, want to operate at the local level, there is nothing to stop them. If they vigorously joined the fray, STV would enable voters to decide for themselves whether they wanted to be represented on their local councils by overt party people, or by independents, or by a mixture of both. Surely that's better than simply imposing a party system on the people, as you would have it?
By the same token, if the National Party in Wellington considered John Morrison to be unsuitable to represent / promote the centre-right / conservative / business interests in the city, there was nothing to prevent them from putting up a candidate against him.
Under STV, whether in single-seat (mayoral) or multi-seat elections, the vote is not split, so, at the mayoral election next month, the centre-right voters could have decided for themselves which one they preferred, without fear of wasting their votes. That's democracy; that's real choice. In this regard, and assuming Celia and John will prove to be the two leading candidates, you watch the votes for Nicola Young flow at least 2 to 1 to John upon her exclusion from the count.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Steven Peters,
Open-list MMP would be an improvement, but not by much. All you're doing then, is casting your party vote by voting for a particular (i.e. one) candidate of your party. Under such systems, the party votes for the party usually still ensure the leading candidates of the party get elected over the candidates who do well on personal votes.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Steven Peters,
Yes, I do mean that, which is exactly why I advocate what, for the time being, I refer to as true-STV. Under true-STV, the voters choose which candidates a party puts up in any given multi-seat electorate will be elected.
For example, if Labour put up David Cunliffe and Shane Jones in the same, say, three-seat electorate, where it was pretty certain that only one could win a seat, the Labour voters could choose for themselves, collectively, which one they preferred, rather than have that choice imposed on them by the Labour hierarchy, as happens under Australian Senate STV.
In addition, one of those three seats could be filled by a genuine independent candidate (as often happens in Ireland). No need (or desire) to ban political parties (which do serve a very useful purpose in any modern-day democracy); just let the voters decide how they are to be represented, and by whom. Only true-STV gives voters (people) the power to do that, which is why the Establishment / political parties don't like it.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Steven Peters,
Australian Senate STV is not really true STV; it is Party PR, much like List PR, and MMP. For this reason, I have never paid much attention to it.
Just so you know where I’m coming from, in my view, most forms of PR strengthen party power and weaken people power. ‘True’ STV (as used in Ireland, Northern Ireland and some local elections in NZ; for you, your local DHB) weakens party power and strengthens people power. For most forms of PR, “PR” is Party Representation. For PR by ‘true’ STV,
“PR” is People Representation.The reason why Australian Senate STV is really just Party Representation, is because the system essentially forces voters (who are required by law to “turn out”) to vote ‘above the line’, where all they have to do is put a ‘1’ in the party box of the party, or group voting ticket, they support. About 95% of voters do this. The alternative is to rank-order every individual candidate (or nearly every candidate), ‘below the line’, in order of preference. When there can be more than 100 candidates standing for election, most voters opt for above-the-line
This means if a particular candidate of party A, say A4, receives more below-the-line first preference votes than, say, A2 and / or A3, the flow of preferences will ensure that A2 and A3 will end up with more overall votes than A4. Voter-individualism is utterly crushed in favour of the candidate-order determined by the party hierarchies.As I have explained here previously, the leaders / prominent candidates of small parties can indeed be elected under STV. Under Australian Senate STV, when the lesser candidates of the main parties have been progressively eliminated, through lack of support, the flow of preferences will often eventually extend down to the minor / micro parties, with whom the main parties have transfer agreements. According to Wikipedia, at the election last Saturday, transfer agreements (and cross-transfer agreements between the minor / micro parties) appear to have resulted in micro parties Motoring Enthusiast and Sports (plus some others) gaining a Senate seat each, despite receiving record-low primary votes of 0.50% and 0.22%, respectively, in what were 6-seat contests where each of the winning candidates had to attain the quota of 14.29%!!
So, put simply, that is how micro parties are “able to emerge in the Australian electoral system.” Is this a “symptom of disproportionality, rather than its opposite”? No, it is not. Despite polling poorly as ‘parties’, the individual candidates concerned each attained the required quota of votes in order to be elected. The fact they needed transfers from other parties (with whom they had transfer agreements) is neither here nor there.
For the individual candidates in Australian Senate elections, regardless of which party they are standing for, it is clear that getting a high place on the party list, rather than being liked by the voters, is what matters. That is how party list systems work.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Isabel Hitchings,
I have spent some time thinking about how people could vote “in tiers”, as you would like to do, and have concluded there is only one way of doing it. Tim McKenzie suggested (in response to you in 2010) that a “Condorcet method, …, could cope”, but there would still be the problem of identifying and extracting the seven individual *winners* (in DHB elections at-large) from the various tiers of candidates, bearing in mind we would all have our own idea as to who should be in our first, second, third, etc., tiers. (The tier of candidates whom you consider would be elected only “over [your] dead body” might well be the candidates *I* would include in my “yes” tier.)
Also, Condorcet methods fail the later-no-harm criterion – the property whereby later preferences cannot harm earlier preferences. With STV, a later preference can never harm (or help, for that matter) the electoral prospects of a candidate already listed, because the later preferences on a voting document are not even considered until the fates of all the earlier-preference candidates have been decided.
The only way of doing it, and which would be an ideal solution for you, particularly under NZ STV (which I won’t take up space explaining here), would be to provide the ability for voters to express “equality of preferences”. With this facility, you could create your own tiers of candidates by voting, say, like this—
A 1, B 1, C 1, D 1
E 2, F 2, G 2, H 2, J 2, K 2
L 3, M 3, N 3, P 3, Q 3, R 3
and, if you really dislike a particular candidate,
S 4
Candidates A to D would each be credited with 0.25 of a vote. If A is the first to be elected or excluded, the 0.25 is transferred at reduced or full value to the next preferences – which of course are B, C and D, and not E (… K). As stated above, the vote does not transfer to E … K until the fates of A, B, C *and* D have been decided, as either elected or excluded. Should the vote eventually be transferred to E … K, its full or reduced value would be shared equally between candidates E to K. The same applies should the vote eventually be transferred beyond K, to L … R.
Not only would this solve your dilemma, and greatly reduce the chances of you “doing [your] head in” (as you said three years ago), it would also be of benefit, not so much for voters who are undecided between their top preferences, but for voters who want to put certain candidates as their bottom preferences, below a whole lot of candidates whom they do not know much about, but for whom being able to give equality of preference would be ideal. Of course, you would still have to read the booklet containing the candidates’ statements, to sort them into your desired tiers.
Unfortunately, while I have an NZ STV computer program that includes an equality of preference facility (and that is ideal for private elections), the likelihood of the NZ STV calculator used for public elections being re-programmed to include an EQP facility is somewhat remote, I should think.
With regard to the task of ranking candidates in order of preference, I note the 2010 DIA Local Authority Statistics booklet states the total number of DHB candidates, nationwide, over the last 12 years were 1,085 (2001), 520 (2004), 429 (2007) and 371 (2010). So, at a probable nationwide average of about 18> candidates per DHB this year, ranking some or most of the candidates in order of preference shouldn’t be too hard.
-
Legal Beagle: Council Elections: STV Q&A, in reply to Keir Leslie,
It is pretty clear where most candidates are coming from. They reveal themselves pretty clearly by which wards they stand in, by their candidate profile statements, and by their voting record once on council, etc.
The last thing we want is the national parties dominating local government, as they do in the UK, for instance. Forcing people to join political parties as a prerequisite to serving their local communities would be entirely counter-productive, in my view. Many good people would be lost to local government, to be replaced by party hierarchy-dominated hacks.
Nevertheless, Rich needs to provide evidence that what you say is true - he's quick to cryptically enquire of others what evidence they have for their statements - and he needs to explain how it would all work.