Supplementary Written Memorandum from the Parliament for Wales Campaign
This submission from the Parliament for Wales Campaign complements detailed written evidence provided earlier on the powers of the National Assembly. However, it stands on its own, whatever may be agreed in respect of the powers.
It responds to the March 2003 Consultation Booklet issued by CRC, with clear answers to all the specific questions raised. We also detail our reasoning where the booklet gave only brief or passing mention to a topic or if its wording appears to suggest a possible predisposition against a proposition.
The specific topics covered herein are:

Voting turnout – how FTPT and AMS contribute to public apathy

How the size of the National Assembly influences government

The divisive effects of AMS within the National Assembly

Advantages of STV in respect of all these identified issues

The attached two-page Appendix lists plausible STV constituencies using County Authority boundaries or paired parliamentary constituencies. These would not need amendment to cater for future population movements or changes at Westminster.
1   Voting turnout
Lower turnout is a long-term trend, due in part to our first-past-the-post election system. FPTP encourages parties to focus on ‘marginal’ constituencies and ‘floating’ voters and, effectively, to ignore most of the electorate. In consequence of this situation, ‘safe’ constituencies have become an election-free zone and those under the age of 50 (close on half the electorate) in parts of Wales may never have seen a canvasser at an election or met their local councillor, AM or MP.
Average turnout in the recent National Assembly election was under 40% – ranging from 24.9% (Alyn & Deeside) to 50.4% (Ynys Môn). It is noticeable that the bitterly contested Rhondda was the only Labour constituency in the top ten for turnout – and, moreover, that of the twenty-four constituencies polling below 40%), no less than 22 are ‘safe’ Labour-held seats and Cardiff Central is very safe Liberal Democrat territory. (Wrexham, formerly the safest Labour seat in North Wales, bucks the trend).
This is not just a problem of locality and local politics. It is facile to hope that voters in ‘safe’ Labour seats are really quite content and would vote if they feared ‘their’ party might lose.
Under the surface, something far more sinister has been at work, contributing to a much wider disillusionment – which will not be resolved by voting electronically at supermarkets or by slick advertising. We believe that we are now witnessing a deep-seated disillusionment, due in part to our election system.
Most insidious of all, many socio-economic groups feel isolated as all parties deploy similar policies aimed at middle-income middle-opinion middle class voters. This leads ‘core’ supporters of both major UK parties to feel ‘taken for granted’. Few take their disillusionment so far as to vote for a traditional ‘enemy’ but, increasingly, more and more are staying at home at election time.
What Wales needs is an election system that values every vote equally and so obliges all political parties to campaign with equal intensity in all constituencies and across all socio-economic groups. This is our primary argument for advocating STV.
2   Size of National Assembly
We support increasing the number of Assembly Members to 80-100. In addition to reasons advanced elsewhere, we believe 60 is too low a number for fruitful three-way interaction between the government, its 2½ oppositions and, equally important, the effectiveness of the government’s own backbenchers.
Good government requires well briefed but not over-worked ministers, vigilant and effective opposition and a reflective and independent backbench that keeps the government in touch with its own supporters and trends in public opinion. Not one of these three elements of good government is operating effectively.
2.1 We believe the nine Welsh ministers have a significantly greater workload than their counterparts at Westminster and in Scotland and have fewer or no junior ministers and no PPSs. Moreover, insofar as not having executive powers must require extra liaison, this adds to their relative workload. They each need at least one full-time junior minister and all need adequate support staff (as in Scotland).
2.2 Unlike at Westminster, the opposition is split 2½ ways and even the largest has fewer AMs available for shadowing than there are ministers and junior ministers. So virtually every opposition AM is a front-bench shadow (and necessarily less well briefed and without adequate support). Whereas at Westminster five or six MPs often work as a team shadowing one single ministry, the ministers in the National Assembly only have to respond to 2½ poorly briefed and mutually antagonistic shadows.
2.3 Whereas at Westminster, backbenchers far outnumber the ministerial teams, here in the National Assembly the ratio is roughly 50:50. This reduces the influence and independence of the backbenchers relative to the so-called ‘pay-roll’ vote. With 80 AMs, backbenchers would outnumber the ‘payroll vote’ by 2-to-1 – a healthy ratio for productive government-opposition-backbench interaction.
2.4 This is why the National Assembly should have at least 80 and preferably 100 members. For pro rata parity with Scotland, the number would be 75-80 - for parity with Northern Ireland it would exceed 150! These comparisons with Scotland and Northern Ireland confirm that the number of AMs should be not less than 80.
2.5 The present formulæ (of one constituency AM for every seat at Westminster and 4 regional AMs for every former seat in Europe) may not survive the next report from the Boundary Commission or change in our representation at Westminster. It is prudent (whatever the number of AMs) to decide on a system that will be unaffected by future external changes over which the National Assembly can have no control.
2.6 If the Commission were to recommend (and legislation were enacted to create) multi-member STV constituencies, boundaries could be frozen. Legislation then could specify a total number of AMs and leave the calculation for each constituency to (say) the Electoral Commission, using the latest figures from the electoral register*.
2.7 With (about) 18-20 STV Constituencies, it is feasible to have any number of Assembly Members from 60-100 – including odd numbers!
3    Divisive effects of AMS
The existing system has many shortcomings. Of these, its divisive impact on day-to-day working relationships within the National Assembly is, in our opinion, more of a problem than any transient irritations at election time. These divisive effects concern the responsibilities of and tensions between Assembly Members, the potential misuse of the election system and the exaggerated and destructive hostility between parties.
3.1 The first problem arises from having two types of Assembly Member, with different duties, responsibilities and workload. Whereas every government AM has duties that overlap locally with an MP, all 20 regional AMs are from the opposition and feel free to roam. With the government AMs all in one category and most of the opposition in the other, this exacerbates tensions between Assembly Members.
3.2 The second problem is geographic. Despite being the leading party (with 37% of the votes), the government is unrepresented over most of Wales! By contrast, in the smaller but more populated areas of Wales, where the government won 30 seats, the combined opposition parties obtained more votes but were rewarded with only two constituency seats. AMS has not ended the anomalies of the first-past-the-post election system – it has added a few extra!
3.3 Concerning the election system as such: this is open to abuse and, at the recent election, there was at least one party trying to woo voters into tactical vote switching. Although this strategy failed for this election, with familiarity it is inevitable that Labour supporters will realise the futility of voting Labour on the second ballot and, instead, vote tactically for a next-to-Labour Party.
3.4 Our fourth complaint concerns its effect on how parties view each other and, as a corollary, how politics is conducted and, in response, how politics is perceived.
As noted earlier, with FPTP there has been a long-term trend towards political parties targeting the same people and offering similar sounding policies. This means inter alia that the major threat to any party comes from the party closest to it in policy and targeting the same socio-economic groups. In Wales, this has created an irrational antagonism between the Labour Party and Plaid Cymru, with both parties spending far more time denigrating the other than arguing for (what they won’t admit are) their broadly similar policies on most issues where the National Assembly has any power.
In consequence of this broad agreement on essentials, the parties are reduced to fierce encounters on trivialities (encouraged by the media who prefer sound-bites to serious in-depth debate and are only too willing to trivialise a subject by personalising any dispute). The electorate, whose political experience is conditioned by the media, responds by distrusting politicians and is less and less inclined to participate or vote.
3.5 No electoral system for Wales can by itself reverse a trend of several decades. However, it would be better to have a system that eliminated the divisions of ¶3.1 and ¶3.2, was not open to the abuse feared in ¶3.3 and, rather than encouraging inter-party tribalism as noted in ¶3.4, actually conferred benefits to a party acknowledging any points of agreement. We submit that STV satisfies all these requirements.
4   Advantages of STV
This is not the place to detail how STV works or explain why it is voter-friendly or any of the many other arguments in its favour. This summary is confined to issues that directly relate to the National Assembly or have specific relevance for Wales.
In respect of the issues raised in this submission, these are the salient points:

There are NO ‘safe’ seats with STV: even in Blaenau Gwent and Methyr Tydfil, there would be a genuine contest for at least the last seat. This must oblige all political parties to campaign in all constituencies.

Political parties cannot risk taking their core vote for granted and must have policies that appeal to all their supporters, not only the ‘wobblies’.

All AMs are equal and have the same constituency responsibilities.

All parties are likely to be represented throughout all regions of Wales.
STV encourages political canvassers to highlight points in common - to win second preference votes from supporters of parties with similar policies. This would be a marked departure from current practice and could be the beginning of a process leading to the replacement of "yah-boo" politics by the more constructive politics that the public evidently would prefer.
Three additional points of relevance for Wales are:

The legislation creating STV constituencies and defining the total number of Assembly Members will not need periodic revision for future population movements or future changes to the Westminster constituencies With the boundaries defined (see Appendix), the National Assembly (or perhaps the Electoral Commission) merely has to determine the entitlement of each constituency against the latest electoral register.

The tradition of non-party participation in local government in many parts of Wales is easier to uphold if votes are cast for individuals and candidates of parties competing on a more equal basis.

The 10% of the electorate that currently votes for various minor parties (UKP, SLP, Greens, etc) tend to feel isolated from the political process. If their second preference vote helped elect someone (albeit of another party), this sense of isolation could be lessened.
Two further points are worth considering:

Unless someone wants to argue that the Irish are smarter than the Welsh, ranking preferences on a single ballot paper by STV produces fewer spoilt votes than FPTP and, clearly, must be much simpler than voting firstly for a candidate and a second time for a party.

With STV, voters have the opportunity to decide between the candidates of their preferred party. Bearing in mind that less than 1% of the electorate are members of any of the political parties in Wales, this would be a welcome and beneficial extension of democracy. Hopefully also, as has been proved in Ireland, this will improve the quality of the Assembly Members elected.
APPENDIX STV constituencies
3-6 members is generally agreed to be the best size for an STV constituency. If the number of Assembly Members is 60-100, about 20 STV constituencies will be needed. One possibly arrangement is to pair the 40 Westminster constituencies and decide the number of seats by aiming for near parity in voters/member**.
CONSTITUENCY PAIRINGS

VOTERS

(2003)

 Allocation of seats for each size of Assembly
Newport East/West

117801

3

3

4

4

4

4

5

5

5
Monmouth/Torfaen

123715

3

4

4

4

4

5

5

5

6
Caerphilly/Islwyn

119322

3

4

4

4

4

5

5

5

5
Blaenau Gwent & Merthyr & Rhymney

108695

3

3

3

4

4

4

4

5

5
Cardiff Central & South & Penarth

127976

3

4

4

4

5

5

5

5

6
Cardiff North & Cardiff West

125051

3

4

4

4

5

5

5

5

6
Pontypridd & Vale of Glamorgan

132151

4

4

4

4

5

5

5

6

6
Cynon Valley & Rhondda

94936

3

3

3

3

3

4

4

4

4
Bridgend/Ogmore

112105

3

3

4

4

4

4

5

5

5
Aberavon & Neath

106967

3

3

3

4

4

4

4

5

5
Swansea East/West

116001

3

3

4

4

4

4

5

5

5
Gower & Llanelli

117951

3

3

4

4

4

5

5

5

5
Carmarthen West & S Pembs/Preseli

111598

3

3

3

4

4

4

4

5

5
Carmarthen East & Dinefwr/Ceredigion

107050

3

3

3

4

4

4

4

5

5
Brecon/Radnorshire & Montgomeryshire

99337

3

3

3

3

4

4

4

4

4
Clwyd South & Wrexham

103960

3

3

3

4

4

4

4

4

5
Alyn & Deeside & Delyn

114944

3

3

4

4

4

4

5

5

5
Clwyd West & Vale of Clwyd

103782

3

3

3

3

4

4

4

4

5
Meirionydd Nant Conwy & Conwy

88185

2

3

3

3

3

3

4

4

4
Caernarfon & Ynys Môn

97171

3

3

3

3

3

4

4

4

4
Number of seats

60

65

70

75

80

85

90

95

100
Votes/member quota

36600

34000

31911

29700

27767

26200

24853

23524

22407
Alternatively, STV constituencies could be based on Local Authority boundaries. Unfortunately, because these electorates vary enormously, Cardiff would have to be subdivided into two not-too-large constituencies and the ten smallest County Councils would have to be paired. The end result requires 18 constituencies with more variable electorates than those based on pairs of Westminster constituencies.
Its justification would be political – by identifying Assembly Members with Local Authorities, they might relate more to local issues. As a corollary, Local Authorities might relate more closely to the national assembly.
Number of Seats

VOTERS
(1999)

 Allocation of seats for each size of Assembly
Newport

104,240

3

3

3

4

4

4

4

4

5
Monmouthshire & Torfaen

137,429

4

4

4

5

5

5

6

6

6
Caerphilly

132,864

4

4

4

4

5

5

5

6

6
Merthyr Tydfil/ Blaenau Gwent

96,345

3

3

3

3

3

4

4

4

4
Cardiff 1

119,447

3

3

4

4

4

5

5

5

5
Cardiff 2

119,447

3

3

4

4

4

5

5

5

5
Rhondda, Cynon, Taff

162,323

4

5

5

5

6

6

7

7

7
The Vale of Glamorgan

88,994

2

3

3

3

3

3

4

4

4
Bridgend

101,675

3

3

3

3

4

4

4

4

5
Neath Port Talbot

107,566

3

3

3

4

4

4

4

5

5
Swansea

177,105

5

5

6

6

6

7

7

8

8
Carmarthenshire

134,487

4

4

4

5

5

5

5

6

6
Pembrokeshire & Ceredigion

142,252

4

4

4

5

5

5

6

6

6
Powys

100,371

3

3

3

3

4

4

4

4

4
Wrexham

96,835

3

3

3

3

3

4

4

4

4
Flintshire

114,932

3

3

4

4

4

4

5

5

5
Denbighshire & Conwy

154,899

4

5

5

5

6

6

6

7

7
Isle of Anglesey & Gwynedd

141,991

4

4

4

5

5

5

6

6

6
Number of seats

60

65

70

75

80

85

90

95

100
Votes/member quota

37220

34357

31903

29776

27770

26273

24813

23507

22332
Many other options could be suggested for multi-member constituencies: these two tables merely illustrate that the task is not difficult mathematically.
Once the decision on the total number of seats has been made, the calculation of the seats for each constituency is its electorate divided by the national "quota of votes/member". If the total number of seats is not equal to the number legislated, the quota value is adjusted up or down as necessary until the target total seats is achieved.