I was interested in hearing that the country of Vanuatu was using SNTV so looked up info about the politics of this obscure Polynesian country. There are only a few countries in the world that use Single Non-Transferable Voting, so I find this interesting.
The Republic of Vanuatu, formerly known as the New Hebrides, is a British Commonwealth country with both a prime minister and a president.
Having been independent from Britain and France for 40 years at the time of the 2020 election, It elects its national parliament in part through a district-level PR system. Forty-four of the country's 52 MPs are elected through SNTV; the others through winner-take-all First Past The Post.
Single Non-Transferable Voting (SNTV) is not sophisticated but operates as a sort of plurality election but one that produces Proportional Representation through multi-member districts and each voter casting just one vote.
If anything, Vanuatu politics are criticized as being too fair, producing what is seen as unstable minority governments due to its fairness and to the voluntary splintering of the country's voters to a multitude of different voting blocks.
The Guardian reported that Vanuatu "is often thought of as a politically unstable nation, with shifting political allegiances, seemingly based more on expediency than ideology." (https://www.theguardian.com/world/2020/mar/18/coronavirus-fears-and-controversial-passport-sales-vanuatus-election-explained)
(But as described below, Canada actually has had more elections in recent decades than so-called "politically-unstable" Vanuatu.)
After Vanuatu's 2016 election there were 17 parties in the 52-member chamber with none holding more than six seats. Due to the 2020 election, 19 parties now have at least some representation in the chamber. Coalition governments have been a feature of government in Vanuatu for quite some time.
Vanuatu suffers from only a moderate voter turn-out, 57 percent, about the same as Canada's, currently. Although in the 1990s apparently Vanuatu experienced voter turn-out rates of about 80 percent. (This is mentioned in Derbyshire, and Derbyshire, Political Systems of the World (1996)).
Vanuatu also suffers from low rate of election of women. The 2020 election produced no improvement on that front. It seems that women voters are not voting for women candidates.
But Vanuatu's elections appear very fair and representative, at least when compared to Canada's. and its PR system is simple in the extreme - no convoluted vote math or transfers but mixed representation is elected in each multiple-member district each time.
The district system used in Vanuatu's 2020 election is structured similar to how Alberta was when it used PR-STV in the 1924-1956 period: some single-member districts, some multi-member districts.
in Vanuatu's case, 10 multi-member districts with the number of seats in each ranging from 2 to 7, and 8 single-member districts.
As there are no vote transfers under SNTV, parties do best if they can judge their vote totals and run only as many candidates as can benefit. (But they can do okay even if they misjudge, especially if many other parties make similar errors.)
Parties seem quite conservative in their expectations. Few predict winning even as many as three in a seven-seat district, so seldom did a party even run three candidates in a district.
144,000 votes in total, electing 52 MPs
roughly about 3000 votes elected a seat
Leaders Party had 18,000 votes and elected five
Peoples Progressive party received about 3000 votes and elected one.
But there are outliers for sure, including:
Vanuatu Labour Party got a bit more than 3000 and elected none.
The National United Party won four seats with only 5000 votes.
Wide fluctuations in the power of the vote - equal voting this is not!
In Vanuatu, representation is not ordained through PR-math but instead through relative lead in votes at the district level. There are no vote transfers to average out candidates' popularity. Thus the inter-relationship of party vote tallies and seat counts is not strong.
But it is interesting that the Vanuatu Progressive Development Party with half a percent of the vote overall won 2 percent of the seats. (In Canada the Greens with 7 percent of the vote won less than 1 percent of the seats.)
The Vanuatu Progressive Development Party was the least popular party to take at least a seat. It received only 825 votes. Six other parties receiving more than 800 votes won no seats at all. So that is un-proportional.
But on other hand, none of those six parties got more than 1600 votes; only three parties took less than 1600 votes and got a seat. So a great majority of seats filled this day went to parties with more than 1600 votes and only three went to parties with less than 1600. And each of those three were able to concentrate their party vote becaue they ran no more than one candidate in any district. More on this below.
Every party with more than 3000 votes got at least one seat.
The three parties who took a seat with low vote count
There is a reason for how these three parties took their seats with such low vote count - each party receiving less than 3000 votes.
What made them different from all the other parties with low vote tallies who got no seats?
The reason is that the party votes were not spread among multiple candidates. a party running mutiple canditesin a distrcit may spit the party's vote tally amogn the two and thus deny either from getting a seats. But here I mean not only in any district but in the country as a whole.
Those three parties each ran only one candidate ac.
Vanuatu Progressive Development Party ran just one candidate and he won his seat.
The same holds true for another one-seat winner, the Ngwasoanda Custom Movement. Its one candidate won in Maewo (a one-seat district), where he received 1300 votes. This was 64 percent of the district vote. Actually less than half his vote tally, only 474 votes, would have been enough to beat out his nearest competitor. (Waste through excessive leads happens in Canada as well under our FPTP system. In some Alberta ridings in 2019 the Conservative won with more than 80 percent of the vote, more than four times his nearest competitor's tally.)
Vemarana also just ran one candidate and won one seat. This was won in seven-member Santo, where 1141 ws enough to take a seat.
====
Another party that elected one member ran three candidates but in different districts.
Vanuatu Cultural Self-reliance Movement ran three candidates. It received a total of 1600 votes and elected one. One of them took zero votes, the other only 365. so that left the lion's share of the party's votes to just one candidate who ran in a district where as litle as 1100 votes was enough to take a seat.
=====
The number of votes needed to take a seat varied from district to district
Here's the rundown:
seat lowest winning favoured
count candidate's total effective party
District (DM) vote tally votes votes represented
Ambae 3 550 4030
Ambrym 2 971 4261
Banks 1 1091 3356 3356 3356
Efate 5 1160 20,850
Epi 2 781 3975
Luganville 2 871 6381
Malo 1 1334 2743 2743 2743
Maewo 1 1300 2102 2102 2102
Malekula 7 835 16,202 6850 9624
Paama 1 289 853 853 853
Pentecost 4 757 9952
Port Vila 5 1255 20,521
Santo 7 1100 23,782 11,550 13,401
Shepherds 1 344 884 344 344
Southern 1 929 2370 929 929
Tanna 7 1175 19,551
Tongoa 1 314 1193 314 314
Toress 1 1012 2054 1012 1012
=========
(And most voters saw someone belonging to their favoured party elected to the chamber, even if not in the district where the voter lives. More on that below.)
Despite the fairness that must have been applied to make the seat count in each district compare to the population, voter turn-out in each district ranged from 38 to 70 percent, so there was bound to be unequal ratios of votes cast per member in the district. (Only floating district seat counts would address that.)
so we see winning vote tallies
one-seat districts range 289 to 1331 (one-seat district are prone to massive waste, either through the winner getting just a minority of the votes (as little as 18 percent is enough to win under FPTP) or to the winner achieving a landslide due to having a massive lead over his or her nearest competitor.)
two-seat districts range 781 to 971
3-seat district 550
4-seat district 757
5-seat range 1160 to 1255
7-seat range 835 to 1175.
==============================
In Canada by comparison where FPTP is used, vote tallies and so forth ranged just as much if not more:
successful candidate receiving highest percentage of votes cast
76 percent Robert Kitchen Souris Moose Mountain
successful candidate receiving lowest percentage of votes cast
29.49 percent René Villemure Trois Rivieres
successful candidate receiving highest number of votes cast
44,456 John Barlow Foothills (Alberta)
successful candidate receiving lowest number of votes cast (in a province):
Labrador 4,119.
of unsuccessful candidates, this one received highest number of votes cast: Peterborough Liberal Maryam Mounsef 24,664 votes
of unsuccessful candidates, this one received highest proportion of votes cast (in a province):
Markham Conservative Bob Saroya 42 percent of votes cast.
======================================
in Efate, a district with 5 members, winners were elected with the following percentages:
11percent
9
8.5
6.5
6
total of 41 percent of vots cast were used to elect winners.
59 percent of vofes cast did not elect anyone.
In contrast to Vanuatu Cultural Self-reliance Movement, which elected one member with 1600 votes overall (described above), other parties sufferd from large-scale waste of the votes they received.
The Leaders Party, which ran 25 candidates, collected 18,000 votes but took only five seats. It ran multiple candidates in many of the multi-member districts and thus suffered from vote-splitting.
The Leaders Party ran three candidates in Ambae and won none.
It ran three in Efate and won just one, although getting enough votes to win two seats.
It ran three in Malekula and won just one, although getting enough votes to win two seats.
It ran two in Port Vila and won none, although getting enough votes to win a seat if gathered behind one candidate.
The same happened in Santo.
So the general effect of Vanuatu's 2020 election was not mathematically proportional but the system did produce mixed representation in each district, or island and region of the country, which is more than can be said for Canada's FPTP system.
Election result was not proportional overall
Each multi-member district produced mixed representation -
no one party swept all of a region's seats --
but the election result was not exactly proportional overall.
Result was not proportional at district level either. There was some bad vote splitting. This was not addressed by vote transfers, as it would have been under STV. The "NT" in SNTV means non-transferable votes!
Usually the results of Vanuatu elections are announced quickly -- no time-intensive transfers. But the death of an election commission official held up the announcement of the results in 2020. it took two weeks for the result to come clear.
The election result was fair at least as compared to the winner-take-all First Past the Post system used in Canada.
No Vanuatu party achieved a sweep of a region's seats;
no Vanuatu party got the winner's bonus of 20 percent of the seats that often accrues to the leading party under our system.
The most variance among the parties with more than 5 percent of the vote was Land and Justice, which got 4 more seats than its vote share. (Meanwhile in Canada, the Liberals got 40 more seats in the House of Commons than they were due based on their vote share.)
Vanuatu's system of SNTV in multi-member districts meant a mixed crop of reps was elected in all districts (except the single-member districts). Because there was no great bonus to the leading party or parties, the result was a minority government and shared power.
No single party received a majority of the seats, as is produced often under our system, sometimes unfairly produced by giving a majority of seats to the leading party, even if the leading party does not have a majority of votes.
Also due to so many parties electing at least one member, most voters could look at the elected chamber and see someone representing thier favoured party.
if we look at the district votes cast as party votes, we see that about 78 percent of voters see someone of their favoured party in the chamber.
And this result is not due to voters being forced to engage in strategic voting (as far as we can tell)
And it is not due to votes being funneled into just two parties as happens under FPTP to more or less degree.
in the districts a range of candidates was offered to the voter:
7-seat district of Santo -- 20 parties ran candidate(s); 7 elected one member.
7-seat district of Malekula -- 18 parties ran candidate(s); 5 elected 1 or more members.
in the 1-seat districts -- 4 to 7 parties ran candidates in each.
in 4-seat Pentecost -- 14 parties ran candidates, four parties elected a member.
Vanuatu's instability not so unstable
The splintering of the vote and the fairness of the elections together leads to diverse parliament and it is said this produces instability. But Vanuatu actually has held fewer elections than Canada over the last 40 years. (Despite our un-representative FPTP system, Canada has had five minority governments since 1980.)
There may have been government changes in between elections I do not know about that. But I do read that the Vanu'aku Party ruled - although not without intense troubles - from 1979 to 1991?) so that is something that looks like stability for that period anyway.
I also read in one place that that the Union of Moderate Parties was the governing party in Vanuatu from 1991 to 1998, and I read elsewhere that a group called the Unity Front came to power in the 1995 election, so some confusion there.
Notwithstanding a general idyllic impression of the South Pacific, Vanuatu's support for the independence (anti-French) struggle in nearby New Caledonia became an ideological political football (or hot potato) with international repercussions and caused intense Vanuatuan power struggles in 1988/89.
The career of Sato Kilman shows the sometimes frequent changes in the prime minister-ship.
Kilman was prime minister on four different occasions:
- from December 2010 to April 2011;
- from May to June 2011. His premiership was subsequently annulled by a court of law.;
- He was elected Prime Minister again on 26 June 2011, thus beginning his first legally recognized term in the premiership. He served until 23 March 2013.; and
- June 2015 to February 2016.
These changes do not necessarily mean a change of government though. The coalition that gave Kilman the prime ministership may have stayed in power, even if he did not remain as PM. The People's Progress Party of which Kilman was leader may just not have had the clout to keep him in that post as compared to other members of the coalition government.
Vanuatu has held only eight elections since 1980.
Canada has held 12!
So which country truly suffers from instability?
Unlike in Canada, in Vanuatu, the prime minister is elected by the MPs. (In Canada no election is held for the PM but the leader of the largest party (or the one that can summon the support of a majority of the MPs) is named PM.
The effect of the 2020 Vanuatu election was a coalition government.
Gracia Shadrack of the Leaders Party, was elected Speaker.
Bob Loughman of the Vanua'aku Pati was elected Prime Minister, 31 votes to 21.
According to an article in the Asia Pacific Report "Vanua’aku Pati’s Bob Loughman becomes new Vanuatu PM" (April 2020), there were two political groupings when Parliament convened to elect the new PM.
One grouping was the Vanua’aku Pati (VP), the Union of the Moderate Parties (UMP), the National United Party (NUP), the Nagriamel political movement, the Green Confederation Party, and other MPs. They made up a total of 30 MPs on one side of the chamber.
The other political grouping was made up of the Land and Justice Party (with 9 MPs), together with the Reunification of Movement for Change (RMC), the Leaders Party of Vanuatu (LPV), the Vanuatu Liberal Movement (VLM), and other political groups that had elected 22 MPs.
Perhaps there was a softening of these two power blocks because a Reunification MP was named Minister of Education when Loughman named his cabinet the next day.
Loughman's cabinet included member(s) of the Vanua'aku Pati (with 7 MPs overall), Union of Moderate Parties (5), National United (4), Reunification (7), Iauko Group (2), Rural Development (2), People Unity (1), and the Green Confederation (1). These parties held 29 seats, a strong majority in the 52-seat chamber.
Overall massive number of parties
48 different parties ran candidates in the 2020 election
19 of them won at least one seat in the election.
The Land and Justice Party got the most seats but just a minority of the 52 total seats - only nine seats.
The result was dis-proportional in that the vote total of the Land and Justice Party was surpassed by three other parties who received fewer seats than that party.
(As mentioned, the Land and Justice Party took more than its due share of seats. And some of this seems due more to luck than anything else. (and luck can mean a great deal in any election system.)
It won the Banks seat by just one vote.
It won a seat in Pentecost by just 20 votes over the nearest contender in a fight for the last seat.
However the fairness of the Vanuatu election was light years ahead of the Canadian result in 2019
where the party with the most votes got the second most number of seats,
the two leading parties received far more than their due,
two other parties received far less than their due,
and a fifth party, the BQ, received a large sweep of Quebec seats (41 percent of Quebec seats) with less than 35 percent of the Quebec vote.
Some party vote tallies and seat counts
In Vanuatu, the Land and Justice Party received 17 percent of the seats with only 10 percent of the vote. (explained above)
The Leaders Party received more votes thanthe Land and Justive Party (12 percent) but got less than 5 percent of the seats.
This dis-proportion is partially caused by higher efficiency of a party that wins more seats - fewer of its votes are wasted.
This problem is sometimes accentuated in systems that use single-member districts. There only one party can take the seat and any votes cast for others is wasted.
But no one party won a large number of these districts in Vanuatu in 2020. Seven parties won at least one single-member district, and only one party, the Vanu'aku Pati, won more than one seat in these districts.
Banks Land and Justice won although receiving just 33 percent of the vote
Malo-Aore Vanu'aku Pati (49 percent of the vote)
Maewo Ngwasoanda Custom Movement (64 percent of the vote)*
Paama Leaders Party (34 percent of the vote)
Shepherds Green Confederation (39 percent of the vote)
Southern Islands Vanu'aku Pati (40 percent of the vote)
Tongoa Union of Moderate Parties (26 percent of the vote) (314 votes)**
Torress Vanuatu National Development Party (49.98 percent of the vote)
* The Ngwasoanda Custom Movement ran just one candidate and that person won the Maewo seat with a majority of the votes in the district.
** The 314 votes that the Union of Moderate Parties received in Tongoa to win the seat were crazy efficient. The Union of Moderate Parties received about 11,000 other votes and only four other seats.
The disparity of voting power was also caused by district variations.
the total votes cast in the single-member districts varied, as did the number needed to win.
the winner in Paama (Leaders Party), in Shepherds (Green Confederation) and in Tongoa (Union of Moderate Parties) each received only about 300 votes.
While the others needed many more than that to win.
This included the two seats won by Vanu'aku Pati.
In Banks 1000 votes went to the winner.
in Malo-Aore and Maewo the figure was 1300.
The Southern Islands winner took about 900.
In Torress the winner took about 1000 votes.
The size of the districts caused some of this -
the number of eligible voters varied from 5100 in Banks to 1400 in Shepherds;
total votes cast varied from 3356 in Banks to 853 in Paama.
A different thing also caused some variation. The percentage of the vote received by a winner of a single-seat district varied from 64 percent in Maewo to 26 percent in Tongoa. In Banks, where the most votes were cast, it was 33 percent; in Paama, where the least number of votes were cast, it was an almost identical 34 percent.
In the case of the seven-seat districts, there was also a range, although not as wide a one.
Voter turn-out in each was about 54 to 60 percent.
Malekula 16,000 votes cast winners took as few as 800 votes.
Santo 23,000 votes cast winners took as few as 1100 votes.
Tanna 20,000 votes cast winners took as few as 1175 votes.
=======================
Some parties used their votes more efficiently than others.
The Leaders Party received more votes (12 percent) than the Land and Justice Party but got about half its number of seats, winning about 10 percent of the seats. The Leaders Party suffered from over-optimism. It ran too many candidates in many districts and thus suffered bad effects from vote splitting.
A cleaner strategy was that of Union of Moderate Parties which ran 20 candidates and elected five, the same number won by the more-popular Leaders Party.
The Union of Moderate Parties took 8 percent of the vote, two-thirds of the Leaders tally, and took about 10 percent of the seats.
Its policy was to run just one candidate in most of the districts.
It ran candidates in 12 of the districts, and none in the other six.
It had six sitting MPs when the election was called, and elected five. (The overlap between the two groups is unknown at this time.)
1 candidate in Ambae
1 in Ambrym
1 in Efate
1 in Epi
1 in Malo
3 in Malekula (the three combined received enough votes to take one seat)
1 in Paama
4 in Port Vila where it elected two
3 in Santo where it elected none (the three together took enough votes for one of them to easily win a seat)
1 in Southern Islands
2 in Tanna elected both of them
1 in Tongoa, elected one.
The worst mistake of the Union of Moderate Parties appears to be in Malekula where vote splitting seems to have hurt the party.
(Any of these impressions of vote-splitting are based on idea that any individual candidate did not add to the number of votes reaped by the party but instead shared out the party vote, thus in some cases depriving the party of success.
On the other hand it is possible that the votes reaped by candidates (and thus the party vote when collated) was based on personal attractiveness and not the party label. So a reduction in the number of candidates may in fact lessen the party's vote tally, and even with a thinner slate of candidates, a party might still be denied success.)
Malekula is also where the Leaders Party received 1900 votes and just one seat.
These two parties -- Union of Moderate Parties and Leader Party -- might have done better if the Land and Justice Party had been similarly over-optimist but it simply ran two candidates in Malekula and elected both.
The Reunification Party similarly ran just three candidates in Malekula and elected two.
The successful targeted campaigns of these two parties left only three seats for the others, and the Leaders Party and the Union of Moderate Parties were not lucky.
The Union of Moderate Parties suffered from optimism, but in its case the optimism was more tempered by common sense. It ran many more candidates than it elected. However, the Union of Moderate Parties took only one fewer seat than it had won in 2016, so not bad despite its lack of success in most of the districts where it ran in 2020.
===============
Outside of the single-member districts, Vanuatu's multi-member districts produced mixed representation
No party received more than two seats in any district.
The greatest diversity was produced in Santo where candidates of seven different parties were elected to the district's seven seats.
Regionalism
I do not know the geography of Vanuatu well, but can see from election results that no region gave an overwhelming sweep of seats to any one party. The electoral system used in Vanuatu did not produce that - unlike the system used in Canada.
Each multi-member district in Vanuatu produced a mixed crop of MPs, with no party taking even a majority of seats in a district.
The most diverse district was Santo where seven different parties were elected to the seven seats.
The Island of Efate also elected diverse representation, both in the island as a whole and in each of its two districts, as described below.
Women
No woman has been elected in Vanuatu since 2008. It is one of just three countries in the world with no women in its parliament. Papua New Guinea and Federated States of Micronesia are the other two. (Why the only three countries in the world like that right now are in the Pacific region is unknown to me.)
There was some optimism leading into the election.
The Shepherds Alliance was to run two candidates - both women.
Other women ran in the 2020 election. 17 women were reported in one source.
A Dr. Thomas reportedly was planning to stand as a candidate for the Oceania Transformation Movement alongside her cousin, Manina Packete. But in the election results I don't see Thomas's name.
Dr. Thomas said "It's going to take many years and a lot of educated work to assist men to change and transform their mindsets, that women should also have a place in the decision-making of a nation," she said.
Other women candidates included:
Anne Leimala Pakoa Tau Independent (Port Vila)
Marie Leitousei Kalkoa "In 2014, Kalkoa was elected as a councillor in the country’s capital of Port Vila, one of five female councillors elected through a gender quota system. [I don't know how this gender quota system works.] She says that the women councillors bonded over the stigma they faced." ("2020 will be my year," The Guardian, June 2020)
one running for the Vanua’aku Pati (name unknown) and others.
No women were elected. It is thus clear that women voters either did not vote or did not give their vote to the local women candidates. Not even vote splitting among women candidates can explain their lack of success otherwise.
At least three women ran in the capital city (Post Vila). Anne Leimala Pakoa Tau did poorly in Port Vila. She got only about 150 votes.
Kolkoa got fewer than 50 votes in Port Vila.
Dr. Thomas's cousin, Manina Packete, came in last in that district with only 12 votes.
Ruth Delarue (Kia Koe Party) came in last in Southern Islands.
Katrina Ulanim Keren William (Independent) came in last in Epi.
Alice Annies Athy (Independent) received fewer than 100 votes in Efate.
Litiana Kalsrap Shepherds Alliance received 89 votes in Efate, also a good distance short of a win.
Alice Kaloran, the other Shepherds Alliance candiate, received just one vote in Tongoa. Ouch!
Apparently multi-member districts are not enough to loosen the male domination of politics. This is surprising as almost a hundred years prior to the 2020 Vanuatu election, a woman was elected in Edmonton in part because it was a multi-member district -- and that was done under the un-proportional Block Voting system. The system used in Edmonton in 1921 was not proportional. It actually produced that measure of gender parity only because the whole Liberal slate was elected -- the Liberals took all the Edmonton seats although receiving only a minority of the votes -- and the slate happened to include one female.
Hilda Lini, a former Vanuatu MP, was a leader in the Pacific’s first ever woman-led party, the Leleon Vanua party. She did not run as a candidate herself in 2020 and the party did not run any candidates under its own name.
Lini was reported as being optimistic leading into the election. "Back at Parliament House, Lini believes things will change when men start looking at what women have contributed to Vanuatu society, rather than holding on to the notion of kastom, the idea of traditional culture that is deeply entrenched in Ni-Vanuatu society and sees chiefs play a big role in community governance." ("2020 will be my year," The Guardian, June 2020)
(Each candidate had to put up a 100,000 Vatu deposit non-refundable - this is equivalent to $1000 Canadian, so pretty hefty. This is four times the size of the candidates' deposit required to run for a councillor seat in an Edmonton election.
Another difference with Canada is that in Vanuatu a person has to be 25 years of age to vote. 18 is the age of suffrage in Canada.)
District results
The district results of the 2020 election, apart from the single-member ridings, produced fair results. Mixed representation was elected in each multi-seat district. In no district did a single party even take a majority of the seats. The voters were splintered into many different groups, and the electoral system reflected that reality. (By comparison, political districts in Canada do not elect diverse balanced representation like that. A single winner in each is all we can expect.)
Efate island was divided into two multi-member districts, each electing five MPs. One district was for the capital city, Port Vila.
Port Vila --- elected a Reunification, Land and Justice, 2 Union of Moderate Parties, Vanua'Aku Pati.
Efate --- elected a Reunification, Land and Justice, Leaders, Vanuatu First, Vanua'Aku Pati.
(The map on the Wikipedia page for the 2020 election should have different colours than are shown for Efate. It should be Red, Orange, Green, Grey and dark green.)
Five parties took one seat each in Efate while the Union of Moderate Parties took two seats in Port Vila, while three other parties took a seat there as well.
Thus the island overall elected two MPs of Union of Moderate Parties, two of the Reunification Movement, two of the Vanua'Aku Pati, two Land and Justice, one Leaders Party MP and one Vanuatu First MP.
With parties represented fairly in each district (if not a proportional result, at least no party was able to take even three seats in any district), gerrymandering would seem to be less important. Thus having two districts split the island was likely not seen as a source of unfairness. Forming ten single-member districts on the island would likely produce much rancor and discontent.
Having a district for just Port Vila (the national capital, a city of 51,000 people) means that local representation, at least at the city scale, is achieved. Meanwhile, the 15,000 "rural" voters just outside the capital city have their own district (Efate), one that is not dominated by the relative multitude of city dwellers that share the island. The district structure of Vanuatu'a system thus creates urban-rural disparity, producing a lower rate of representation in a rural to urban comparison. Note that if urban-rural disparity is an issue, SNTV's district structure allows easy mitigation - simply give the capital city more seats. Such a change does not necessitate drawing of new district boundaries as would be the case under FPTP.
Port Vila (apparently) has two or three times the population of the rest of the island but each "half" has the same number of MPs.
Oddly, about the same number of votes are cast in each "half", with the number of eligible votes also about the same in each "half" -- 44,000 in Efate and 51,000 in Port Vila, although total population in the whole island in 2009 was said to be 66,000, rising to an estimated 80,000 by 2020. The population of Port Vila was put at 51,000 in 2016 when it was said to have 66 percent of the island's population.
21,000 votes were cast in each of the two districts. But the voter turn-out rate in Port Vila was only 36 percent, a low figure when compared to Efate's turn-out rate of 57 percent. So apparently, representation was meant to be stronger for Efate (the same number of members but for a smaller number of voters), but the lower rate of voter turn-out in Port Vila compensated, and the ratios of votes cast per member turned out to be similar in the two districts.
Could the low turn-out rate in Port Vila be a sign of a turning against SNTV among voters in the sophisticated capital city (at least I assume voters are sophisticated there, at least relatively so as compared to the rest of the country)? Are they starting to see SNTV as an overly-simple PR system that wastes too many votes?
If so, let's hope a change to FPTP, such as Canadian voters suffer under today, is not the outcome. Instead reform of Vanuatu to a more refined PR system, such as STV, would be the better route. The simple addition of preferential ballots would be all that is needed in the multi-seat districts, where 44 of the country's 52 MPs are elected.
Ambae
In the probably-fairly-typical 3-seat Ambae district 4,000 votes were cast
the three winners were of three different parties, so mixed crop of reps elected
winners received about 550 votes each, leaving 5/8ths of votes not used to elect anyone.
a wide range of candidates was offered to voters: 22 candidates, including 4 Independent candidates
most parties ran just one candidate
Leaders Party ran three candidates, elected none
Vanuatu First Party ran 2 candidates but got a meagre total, even together.
Vaanu'aku Pati ran two candidates, altogether received only 200 votes.
People Unity, National United, Rural Development won one seat each.
Compare the election in Ambae to the 7-seat Malekula district.
Malekula district
The 7-seat Malekula district: 16,000 votes cast
Winners received 1200 to 800 votes each, altogether about 7,000,
leaving majority of votes cast to not be used to elect anyone.
41 candidates, including nine Independents and candidates of 17 different parties
voters often had choice of more than one candidate of same party. in those cases, they could choose between parties and also between individuals when voting for a party.
mixed rep. elected - candidates of four parties elected
the single-voting allowed fairness; the voters' splintered sentiment meant no one party took all the seats in the district.
Land and Justice ran two candidates, elected two
Reunification ran three candidates, elected two
Leaders ran two candidates, electing one
Vaanu'aku Pati ran four candidates, electing one
(two of the three unsuccessful candidates received almost identical 6000 votes each (classic vote splitting!), enough together to take a seat.)
National United ran one candidate and elected that person (careful husbanding of its resources!)
People's Progressive ran two candidates but elected none although altogether tis candidates received enough to take a seat. Its most popular candidate missed the last seat by only 70 votes; the other candidate received 250 votes - vote-splitting sucks!
but a People's Progressive candidate did win elsewhere so the party did have representation in the chamber.
Waste of votes
Many votes were wasted by the SNTV system. (But many are wasted in Canada's FPTP system as well. Under STV by comparison, 80 percent or more of the valid votes are used to elect someone preferred by the voter.)
In Vanuatu's two-seat districts, a majority of the votes were wasted.
two-seat districts:
Luganville 16 candidates 13 different parties 3 Independents
winners took 2000 of 17,000 votes cast
Epi 12 candidates 10 different parties 2 Independents
winners took 1600 votes out of 5700 votes cast.
Ambryn 10 candidates 9 different parties 1 Independent.
winners received 2000 votes out of 7000 votes cast.
Note that no party felt confident that it would take two seats so each ran only one candidate in these district contests.
This ensured that a wide range of candidates were offered to the voters.
==================
Vanuatu's SNTV system, if clunky, seems to work!
================================================
SNTV wastes votes and the usual Droop quota does not apply as far as filling of seats
Under SNTV it is possible to win a seat with far less than the notional threshold of Droop quota, or win two seats with far less than two Droop quota.
if you look at recent Vanuatu 2020 election (wiki article on it seems accurate),
you see that:
23 percent was enough to win a seat in a 2-seat district when 33 percent is the notional quota
14 percent was enough to win a seat in a 3-seat district when 25 percent is the notional quota
and so on to
6 percent was enough to win a seat in a 5-seat district when 17 is the notional quota, 5 percent was enough to win a seat in a 7-seat district when 13 percent is the notional quota. just like in FPTP where majority is the notional quota, the waste of votes, the non-transferable votes means no such organized quota.
but we do see much more fairness than in Canada's FPTP (although vote transfers would have made more fairness and less waste of votes).
Vanuatu
10 multi-member districts with the number of seats in each ranging from 2 to 7, and 8 single-member districts. the Vanuatu Progressive Development Party with half a percent of the vote overall won 2 percent of the seats. (In Canada the Greens with 7 percent of the vote won less than 1 percent of the seats.)
...The greatest diversity was produced in Santo where candidates of seven different parties were elected to the district's seven seats
==========================
コメント