Monday 11 February 2013

Jersey’s Electoral System and the Venice Commission


At the end of last week, Deputy Southern lodged an amendment to the referendum law to change the format of the referendum so that it removed Option B, to leave it as a simple "yes" or "no" vote on Option A.

I argued very strongly for this format of referendum after the publication of the interim report, but had reservations over whether to seek to carry on arguing after the final report had been written and the draft law lodged. But since Deputy Southern has had the initiative to do so, I have written this report to back it up.

I think it is important to have the facts and statistics laid out bluntly as I have tried to do here. I have also changed some of the charts I had previously used to include the most up-to-date figures available.


Jersey’s Electoral System and the Venice Commission

The Venice Commission says - “Equality in voting power, where the elections are not being held in one single constituency, requires constituency boundaries to be drawn in such a way that seats in the lower chambers representing the people are distributed equally among the constituencies, in accordance with a specific apportionment criterion, e.g. the number of residents in the constituency, the number of resident nationals (including minors), the number of registered electors, or possibly the number of people actually voting.”[1]

It goes on to say – “the maximum admissible departure from the distribution criterion adopted depends on the individual situation, it should seldom exceed ten per cent and never 15%, except in really exceptional circumstances (a demographically weak administrative unit of the same importance as others with at least one lower-chamber representative, or concentration of a specific national minority).[2]

The Chief Minister, in answers to a question from Deputy Pitman on the 29th January 2013, said “the Venice Commission is a code of good practice in electoral matters and therefore it is not written in law.[3]

Whilst this is technically true, the European Convention on Human rights is written in law, and Protocol 1, Article 3 states “The High Contracting Parties undertake to hold free elections at reasonable intervals by secret ballot, under conditions which will ensure the free expression of the opinion of the people in the choice of the legislature[4]. However, the Venice Commission is a body of the European Council, the same body that governs the ECHR and provides the courts to uphold the convention. The purpose of the Venice Commission is to define exactly what constitutes a “fair” election, and how constituencies must be composed to ensure that the principle of “equal suffrage” (which is fundamental to the fairness of an election) is adhered to. So to be in contravention of the Venice Commission, is by extension, arguably to be in contravention of Protocol 1, Article 3 of the ECHR, which therefore opens Jersey’s electoral system up to embarrassing legal challenge.

The Chief Minister also went on to say “it does recommend the principle of equal voting powers but it also recognises that there might be exceptions for geographical, administrative or historical boundaries[5]. As the previous quotations from the Venice Commission show, the exceptions for breaching the 10% deviation limit are to protect minorities and for weak administration units. It is highly unlikely that the Parishes would fall under this category, especially when the functions of the Parish administrations and the States are not fundamentally linked, and when it is possible to combine Parishes to create electoral districts for the national parliament. In fact, one could argue that St Helier would deserve to be overrepresented on the basis that it has the highest proportion of foreign nationals and minorities residing there, and that they currently have no representation at all in the States of Jersey.


Under Jersey’s current electoral system –

Table 1


Parish
Eligible voters[6]
D + C
D+C/E
% Deviation
St Mary
1,340
2
670
-60.7
St John
2,280
2
1,140
-33.18
Trinity
2,370
2
1,185
-30.54
St Martin
2,970
2
1,485
-12.95
St Ouen
3,200
2
1,600
-6.21
Grouville
3,870
2
1,935
13.42
St Peter
4,010
2
2,005
17.53
St Lawrence
4,280
3
1,427
-16.35
St Clement
7,170
3
2,390
40.09
St Brelade
8,590
4
2,147
25.85
St Saviour
10,590
6
1,765
3.46
St Helier
26,890
11
2,717
59.26
Total
77,560
41
1,706
119.96


As can be seen from table 1, of the 12 Parishes, 10 of them fall outside of the 10% deviation limit prescribed by the Venice Commission. The difference between the two furthest deviations is 119.96 percentage points.

When the population figures from the 2001 census are used, the total deviation was 95.5 percentage points. So as the time goes on and different Parishes expand at different rates, the problem of malapportionment is getting worse, not better.

Under the Electoral Commission's proposed Option A –

Table 2

District
Parish
Eligible voters[7]
% deviation from average
1
St. Helier 1
13,960
7.99
2
St. Helier 2
12,900
-0.20
3
St. Clement, Grouvile, St. Martin
14,010
8.38
4
St. Saviour, Trinity
12,960
0.26
5
St. Lawrence, St. John, St. Mary, St. Ouen
11,100
-14.12
6
St. Brelade, St. Peter
12,600
-2.52


Average – 12,962



Table 2 shows that in the proposed new and simplified system, only one constituency falls out of the 10% deviation limit. However, it does still fall within the ultimate 15% limit. Given that it is impossible to provide better voter equity without abandoning the existing Parish boundaries, it is unlikely that District 5’s deviation would be considered a big problem.

The total deviation from highest to lowest is 22.5 percentage points. This is a huge improvement on the 119.96 points in the current system.

Under the Electoral Commission's proposed Option B –

Option B is complicated by the fact that two concurrent sets of constituencies would exist.

Looking first at the Parishes separately as single member constituencies –

Table 3

Parish
Eligible voters[8]
% Deviation from average
St. Mary
1,340
-79.27
St. John
2,280
-64.72
Trinity
2,370
-63.33
St. Martin
2,970
-54.05
St. Ouen
3,200
-50.49
Grouville
3,870
-40.12
St. Peter
4,010
-37.95
St. Lawrence
4,280
-33.78
St. Clement
7,170
10.94
St. Brelade
8,590
32.91
St. Saviour
10,590
63.86
St. Helier
26,890
416.06
Total
77,560

Average
6,463



As you can see, not a single Parish falls into the 10% limit. The overall deviation is 495.33 percentage points. Using the previous census figures, the total deviation was 425.18 percentage points. So the problem is getting worse, not better.

Now, combining the two sets of constituencies, these are the figures –

Table 4 –

District
Parishes
Eligible voters[9]
Number of States  Members
Voters per D + C
% deviation from average
1
St Helier No. 1
13,960
5.5
2,538
32.19
2
St Helier No. 2
12,900
5.5
2,345
22.14
3
St Clement, Grouville, St Martin
14,010
8
1,751
-8.8
4
St Saviour, Trinity
12,960
7
1,851
-3.56
5
St Lawrence, St John, St Mary, St Ouen
11,100
9
1,233
-35.78
6
St Brelade, St Peter
12,600
7
1,800
-6.25
Total/ average

77,530

1,920



As can be seen in table 4, the total deviation is 67.97 percentage points. You can also see that district 5, despite having a lower population, has twice as many representatives than districts 1 and 2.

Note that half of the constituencies are outside of the 10% limit, two of which are over 3 times the limit.


Conclusion

It is clear from these statistics that the status quo and Option B are both totally incompatible with the limits set out in the Venice Commission. A move to Option B would not fix any of the problems with Jersey’s democratic deficit, and will over time exacerbate the problems and lead to potential legal challenges and years more of debate, which will further disenchant the people of Jersey with politics. Not to mention it will cost more taxpayers money.

However, Option A provides Jersey with an electoral system that will provide all islanders an equal say in the composition of the States of Jersey.

The Electoral Commission itself received an expert report from Dr Alan Renwick that confirmed that Option B actually increases malapportionment from the current system and is in clear violation of Venice Commission.[10]



Option B and the Electoral Commission’s principles and research 


In the Electoral Commission’s report, it outlined what overriding democratic principles must be adhered to for a democratic reform. They were as follows –

  1. All electors should have the same number of votes.
  2. Constituencies should as far as possible be of equal size.
  3. A candidate should generally require a significant number of votes in order to be elected to the Assembly.
  4. The electoral system should be simple, fair, and easy to understand. 



Further to the statistics provided, simple logic can be used to further show the problems with Option B.

Option B is incompatible with each of the commission’s principles.

  1. Whilst each voter may in theory have the same number of votes, in practice, most Constable elections are uncontested. So those in Parishes with an uncontested Constable election will have one less vote than those in other Parishes. Under Option A, each election will be contested and each person will have the same number of votes.
  2. Whilst each super-constituency will have a similar number of voters, the Parish constituencies will have hugely varying populations.
  3. Whilst those in the super-constituencies will require a substantial number of votes, that will not be the case in the Parish constituencies. Many will be elected unopposed, and so receive no votes (this may become more common owing to the fact that Constables work load will increase substantially in a reduced chamber). Even in contested elections in the smaller Parishes, the winner will not have obtained anywhere near as many votes as the Deputies.
  4. How simple something is, is ultimately subjective. However there is no doubt that Option B is less simple than Option A. There will be two categories of States Member, one of which will be elected primarily to run the Parish (though this contravenes the commission’s report that says they should see themselves as primarily States Members), and many will find it incomprehensible (as they already do) as to why the Constables are necessary in the States to the detriment of equal suffrage. 

Further to this, the commission’s report states that all members (including the Constables) would have to see themselves as primarily States Members, rather than anything else. This would fundamentally change the nature of honorary role of Constable, and ultimately detract from the Parish system which is so important to the island. In a reduced sized chamber, each States Member would be required to take on a greater work load, with more Constables having to take on roles, potentially even ministers.

This would ultimately reduce democracy in the Parishes, as the father/ mother of the Parish would be forced to dedicate themselves to the States before the Parishes.

Option B is illogical and inappropriate. It is evidently a product of having States Members on the Electoral Commission, rather than it being an independent panel. It should therefore be removed from the referendum to save Jersey time, money and potential embarrassment. 





[1] European Commission for Democracy through Law (Venice Commission) Code of Good Practice in Electoral Matters, guidelines and explanatory report, paragraph 15
[2] Ibid
[3] States of Jersey hansard, 29/01/2013
[4] European Convention on Human Rights
[5] States of Jersey hansard, 29/01/2013
[6] Estimate using 2011 census data. Same figures as used in the Electoral Commission Final Report page 14.
[7] Electoral Commission Final Report page 27
[8] Estimate using 2011 census data. Same figures as used in the Electoral Commission Final Report page 14
[9] Electoral Commission Final Report page 27

11 comments:

  1. Ever since I read the wording of the Venice Commission and knowing how Mr Bailhache is very concerned with Jersey's international reputation (re: his response to J Hemmings), I could never see how the commission could include option 'B', so odd and as you highlight, it could prove to be very embarrassing to Jersey.

    Hopefully someone will see sense!!

    ReplyDelete
  2. Unfortunately, if Sen Bailhache & Co can convince enough voters to support option B it will be so much more difficult to challenge the retention of the Constables via Strasbourg.
    The "margin of appreciation" defence will be much strengthened and the traditionalists such as Bailhache will make feudalism seem almost attractive to his fellow judges.
    In fact a human rights challenge against the existing set up would probably stand a batter chance now at Strasbourg. Making it even worse - but with a will of the people endorsement - could settle the matter for another century!!

    ReplyDelete
  3. Well done Sam another great analysis. For me, the most important point in Geoff Southerns proposition is the two votes thing. He explains it quite clearly and I hadn't noticed it until now.

    If you want the constables to stay you can use your two votes on options B & C. If you want the constables to go you can only use one vote on option A. It's an absolute disgrace that they have actually proposed what they have.

    ReplyDelete
    Replies
    1. To be honest I actually think that that is a misunderstanding of how the voting system works.

      Nobody in this referendum will have two votes. In the final count, only one vote per person is counted.

      If you vote for the option that comes last, your vote is discounted. It is only then that your second preference becomes your single valid vote.

      Having two options that keep the Constables will split the pro-Constable vote. If B or C comes last, the vote will then be reunited.

      This means that the Constables will only win if their total support exceeds 51%. And if that is the case, well then we lost fair and square.

      There isn't an inherent bias in the question format. The UK Electoral Reform Society would have never cleared the question if it was biased.

      Delete
  4. Excellent, robust and concise analysis. One very minor thing- your use of the word 'most' in counter argument 1. I'm not sure if this is a fact although I accept it is entirely possible- e.g. St. Ouens - 1 contested Constable election in 100 years. If you can proove that 'most' is correct- excellent. If not, I'd replace 'most' with 'many' so, 'in practice, many Constable elections are uncontested' which is correct.
    Southern's excellent ammendment to the EC proposition also needs applause and your analysis only strengthens it. Well done. Suddenly the EC and reforming Jersey's (feudal) political system has got a load more interesting. Congratulations.

    ReplyDelete
    Replies
    1. Good point, I suppose it depends what timeframe is being used. Certainly at the last election it's own it was "most" (8/12 I seem to recall). Maybe it is different going back further.

      Delete
    2. There is no need to be coy or understated about this point - the evidence is clear.

      In 2011, 8 out of 12 Constables were returned unopposed.

      From 1999 to 2010 inclusive 35 out of 49 Constable elctions were uncontested.
      Which is 71%, = more than 2/3

      So since 1999 until now, more than 2/3 of Constables have been returned unopposed. Put another way, of the 12 Constables sitting in the Statwes at any one time, 8 or more are there without having fought an election.

      NOTE. Iin the first line above I do not use the word "elected" because they are not "elected" to the States, they are "elected" to their parish role only. An important distinction.

      The 1947 Privy Council Commission, having proposed the removal of the Jurats and the Rectors from the States ( ! ) stated this about the Constables, “No one should be elected to the States by virtue of election to any other island or parochial office”

      As clear a nod as can be imagined, that in due course, the Constables will have to leave the States too.

      Delete
  5. Have you sent this to all States Members? If not, then you should. This is an excellent posting Sam.

    ReplyDelete
    Replies
    1. Thanks anonymous. I started writing this as just a note to Deputy Southern but got carried away and wrote all that I have. So I'm going to share it with States Members in the next few days, and also decide whether I take it to the media or not.

      Perhaps you could also contact your States Members to ask them to support the amendment too?

      Delete
  6. I amconcerned that the proposed method of counting the referendum results has never been used before in Jerseypolitics. It has always been first past the post.

    ReplyDelete
    Replies
    1. It should be fairly simple, though the counters will need to be aware of the process.

      What will happen is each Parish hall will have their ballots to count. There will be three piles, A, B and C. They will go through all of the ballots and put them in the piles for which they gave their 1st preference vote to.

      Then the whole island has to wait until all of those 1s are counted, and if one option has gained over 50% of the vote, it wins. No need to do anything further.

      But if there is no majority, all of the ballots in the option that came last, are then retaken and put in one of the remaining two piles according to which ever option they gave their 2nd preference to.

      So after that, one pile will be bigger than the other, and then we have the winner!

      Should be okay I think.

      Delete