THURSDAY, 10^th JULY 2003

EVIDENCE OF DAVID MORRIS, CHAIR; DR RUSSELL DEACON AND DR JOHN COX, FRSA, PARLIAMENT FOR WALES CAMPAIGN

Present

Commission

Eira Davies

Tom Jones

Dr Laura McAllister

Peter Price

Ted Rowlands

Lord Richard

Thank you very much for this opportunity to listen to your views. Could you formally, for the sake of the record, identify yourselves?

David Morris

I am David Morris. I chair the Campaign for a Welsh Parliament. I want to thank you for letting us make a presentation. I would like to emphasise that we are a mixed party, in terms of our politics. And of no one party, so there is no one particular influence within our organisation. It is important we say that at the outset.

We are not new to campaigning for greater powers, and a Welsh Parliament. We would hope, in fact, that by this submission, we will make some steps towards a kind of parliament, certainly towards what Scotland has, in the very near future.

We have two people here to make the presentation. I made one in Swansea. I have had the personal opportunity of making my views already known, so I shall not repeat myself as time is of the essence. We have Dr John Cox, who will speak second, and Russell Deacon who will speak first.

Dr Russell Deacon

The document has a brief introduction, in focusing on what are our aims. Then we put within it a series of elements of which Dr Cox will go into more detail.

Our first element – I will just go over them briefly, over what the main headlines are – our first element is the increase in the number of assembly members, who will be parliamentary members, to 80-100, without a decrease in the number of MPs at Westminster.

The second element, we will see the electorate using a single transferable vote. Dr Cox will go over that in detail in a minute.

Element three is a review of the role of the Secretary of State for Wales, which still exists in some form now. That would be with a full consultation within Wales between new parliamentary members.

Element four would be the introduction of the Welsh Civil Service. It would be similar in nature to that in Northern Ireland, but not encompassing so many parts.

Element five is the granting of primary law-making powers. In the document, we have highlighted a number of areas where we found that would be beneficial. We go on then to detail. The type of legislative process we would like, which is similar to that in the Scottish Parliament. The document then deals with the state of existing United Kingdom legislation’s overlapping powers.

Element six looks at judicial devolution based on the Welsh primary legislation. The rest is how we believe the legal system will have to be amended to encompass a law-making parliament.

Element seven looks at the funding of the system and a review of the Barnett formula so that it is more equitable to Wales.

Element eight deals with the international position of the Assembly, and the representation of Wales abroad.

Element nine shows the increase in the areas of responsibility for the National Assembly by taking on new powers and functions. These would go in areas such as law and order, broadcasting and local government elections.

Element ten is not within the remit of your commission, but is the desire of the parliament campaign. That is the granting of tax-bearing powers. These would be wider than the Scottish Parliament, and over greater areas than that.

Element 11 is the greater degree of all Wales’ involvement by the assembly in support for regional committees.

Those were the main aims of the Parliament Campaign, as we saw applied to the Richard Commission.

Lord Richard

Thank you very much indeed. I think we would like to concentrate more on the election side. Somebody is going to introduce the second part, of what the electoral arrangements are.

Dr John Cox

Yes, I can do that. You have three critical slides from the presentation that have been printed for you (Note: The PowerPoint presentation slides are reproduced at the end of this transcript).

The basic elements of the paper starts off with an argument about voter apathy, of voter turnout being as poor as it is, in part because of the electoral system. The first of the printed slides (Slide 2) illustrates this quite well, in so far as you will see by the strong coloured red at the bottom of the table, that all the 24 seats, bar two exceptions – Wrexham and Cardiff Central - are safe Labour seats, that have got the lowest votes.

The 16 that had comparatively good turnouts, and I emphasise the word comparative, were all ones which were contested. There is, we believe, a 10% potential improvement in turnout if every seat were to be genuinely contested, rather than a waste of time for the voter. However, in the written paper, I also make the point that this is a long-term problem of voter apathy arising from the present electoral system.

I was talking to my paperboy today who comes from a family whose parents and grandparents have now not voted at all over a period of 40 years. He and took part in the February 15^th demonstration against the war in Iraq, he and is a politically conscious 15-year-old - but sees absolutely no relevance of elections to his political consciousness.

We have a generation who have grown up because, particularly in Wales, not only is there a large number of seats where the outcome seems to be so predictable, but the appeal of major parties has not been to Welsh voters. The appeal has been to middle England, middle class, middle opinion voters, and there is, therefore, an alienation that comes from that.

The major argument I have put forward for an STV and the Parliament for Wales campaign, is that because every seat with STV would be contested, it starts adding more point to participating in elections. We are certainly not going to reverse a trend of 40 years to one election, but we think that in time it would lead that way.

Ted Rowlands

Voter apathy is a worldwide trend.

Dr John Cox

There are downward voting trends in every country in the world, in reality, but our downward trend is a bit more impressive. The Welsh Assembly votes were a bit more impressive again, and the under-25-year-olds were even more impressive, in terms of voter apathy.

The second element of the written paper I will skip over quite quickly because it simply is an argument for there being up to 80 assembly members, and I’m ready to home in on 80 as the figure, which is a more reasonable number of assembly members.

The implications of there being more assembly members is that, on the present electoral arithmetic, it would not be possible for the current leading party in Wales to get an overall majority. The significance of that is that if the commission were to decide that 60 is a valid number, I would find it difficult to really argue that it makes a jot of difference what your electoral system is.

The present electoral system was designed on the assumption that Labour would never get worse than Michael Foot’s Westminster election year of 43% for Labour, and was designed on the basis that there would be an overall majority for Labour even in that circumstance. In fact, Labour’s vote is less than the worst year that was envisaged at that time.

I have looked back at writings prior to the legislation of setting up the assembly. Every single paper was assuming figures such as 46-50% as the vote that Labour would get, and that they would be in overall majority.

If we assume that the present situation, and the one in 1999, is consistent, we are faced with 60 seats, with a possibility of an overall majority for Labour, with the present electoral system. At the moment, if we increase the number of seats by any amount, even with the present AMS system, we are going to have to have a coalition government.

Our argument here would be that if you are going into a coalition government, it becomes particularly important that your proportion system is genuinely proportional, rather than having a big partner, who is big by virtue of a 10% of over-representation in the seats that they have got, and a junior partner who is getting less seats than he deserves from votes. So it is an argument for a more proportional system, once you go to 80 members. You are welcome to interrupt me at any stage.

Sir Michael Wheeler-Booth

What you have just been saying, is that not an argument by the government? The Labour government, the Welsh government would resist the change and would reject such a recommendation. Won’t they resist putting up the numbers because they are losing control?

Dr John Cox

It is obviously a consideration for the commission of what would be an acceptable recommendation. I am in the privileged position of simply arguing for what I think is right. What I think is right is that we have a proportional system. Whether or not that means coalition or majority is not a matter that should concern us. We are concerned about the wishes of the people, as expressed through their votes, being reflected in the government that we then have for Wales.

Lord Richard

On the basis of your model, your STV model, there would need to be a coalition?

Ted Rowlands

What would the result have been on STV with 80 seats?

Dr John Cox

The second of the printed sheets (appended as Slide 11) which I have asked to be supplied to you shows the potential results for all the STV and AMS systems in respect of proportionality for the leading party If you ignore all the mathematics that leads up to it, if we had had AMS with an 80-member assembly, then what is defined as the disproportionality would have been about 5%. So you would need 45% of the votes for an overall majority to be gained by Labour. That is what that means.

If you used any of the STV systems that were talked about until Peter Hain made a speech a few weeks ago, you would be more proportional than that. The disproportionality is below 5% so you are getting closer to a truly proportional result.

If you were to have the system whereby you had the same Westminster constituencies, with two members per constituency – Ted Rowlands raised it, and I looked at the transcript of the questioning of Ken Ritchie on this – I calculated it two ways. One is the standard way that I use for other STV systems. The second one I had a bash at trying to work out where the existing votes would go. That is why there is a range shown on that sheet.

If you take optimistic assumptions about where second preferences went towards Labour, you get up to 9.5% disproportionality. If you take pessimistic assumptions about it you get 5.5%. To give you a much simpler answer to your question, if you had had such a system in operation, the Westminster constituencies, with two members by STV, there still would not be an overall majority for Labour, on these assumptions.

Peter Price

Surely you have to know the second preferences to make such a calculation?

Dr John Cox

I had not done it that way. I have defined disproportionality as the percentage seats minus the percentage votes for the leading party. At the moment there is a 10% disproportionality because Labour is winning 50% of the seats with 40% of the votes. On that definition of disproportionality, you need 10% disproportionality in order to get an overall majority. If Labour were to increase their votes to the Michael Foot low of 43%, you would need a 7% disproportionality, if the objective of the voting system is to get an overall majority for the leading party on a minority of the votes.

Vivienne Sugar

Can I ask about the principles behind the pairings you have shown, of constituencies? Just take us through the logic of the pairings that you have suggested.

Dr John Cox

The pairings of Westminster constituencies were basically that they should be as close to each other geographically as possible. It turned up that way (Slide 16). There are other options. Again, I have given you a printout of that, it is the one based on the local authorities (Slide 14). I do not think there is an overall logic.

The geographical thing is simply that you always look for another parliamentary constituency that is right next to it, and you end up with the map that there is. However, I was prejudiced by saying, let us start with the local authorities, then, as far as was possible using the Westminster constituencies, to try to arrive as close as possible to the same result. If people are already in the same local authority, then it makes sense to pair constituencies that are in the same local authority, rather than have them divided, even for the Welsh Assembly, where they would then be divided three ways.

Vivienne Sugar

On your table of the authorities, Swansea would include Swansea, Swansea West and Gower, whereas on the constituency pairings, you have Gower with Llanelli.

Dr John Cox

Exactly. Perhaps that is leading me onto the question of the distribution of seats. It would be technically possible to simply operate STV with the distribution you see on that sheet (Slide 13). But all the pundits feel that STV does not really work well if you only have two members per constituency. Also, the ballot paper gets very long if there are potentially eight people to be elected. If each party put up six candidates, and we have four parties in Wales for sure, probably more, so it is getting a bit ridiculous.

Dr Laura McAllister

Is not the other issue that if you have that kind of disparity between numbers elected for those seats, that you have quite a disparity in proportionality across Wales, and that undermines some of the arguments that you were putting forward as a strong case for STV?

Dr John Cox

That would be so if we were doing it the same way as has happened in Northern Ireland where, irrespective of electorate, they said it was six-member constituencies. I have throughout assumed that we would allocate the seats in proportion to population to try to get equity in terms of population. As an aside, I have dodged the issue of sparsely populated regions getting a little bit more, though it is easy enough to build into the model. But if you are making a presentation, we are not here to present you with a bill for parliament, we are just introducing the argument.

My assumption has been that you would say, we need X number of members in the assembly, we have Y constituencies, and we have now going to allocate the seats for each constituency in proportion to population with equity. Then, later on, if someone wants to put in an extra percentage for sparsely populated areas, they can do so and the mathematics is quite straightforward. In response to your question, the answer is quite straightforward: having different numbers of members per constituency doesn’t affect the question of proportionality in that sense.

Where it does affect it is that in the two-member constituency, obviously there is a greater potential for disproportionality in that constituency which is not going to be offset by the smaller disproportionality in the eight-member seat.

Looking again at that map (Slide 13), it is immediately obvious that you would amalgamate all the areas with an apparent allocation of two with a nearby three, and you would split Cardiff and give it two STV constituencies. Then you arrive at the one that appeared in the appendix of the original written submission.

Dr Laura McAllister

Is not the other point to make about the constituencies that it is less important to have the geographic integrity, other than proximity, because with multi-member constituencies you are allowing the electorate to choose representatives who, by the nature of it, in the Irish case, tend to reflect the different parts of the constituency? I have done that a little clumsily, but what I am saying is that geographic elements are not as important in STVs as they are in some other systems.

Dr John Cox

It is certainly true that if you had a Rhondda Cynon Taff constituency there would be very little danger of Rhondda getting all the people elected if it was felt there was a parochial element to the election. Also, the parties concerned would almost certainly nominate candidates who would appeal to every sector of the electorate.

One of the advantages of STV is that it is not just the parochial issue of where they live, but it is also the type of people elected. You are not going to have all white middle-aged men nominated as often happens today because you are unlikely to be appealing to as many people as you want. STV is better from the point of view of having a genuine overall balance in the people elected, because you have to have a balanced team, when you put up candidates for STV. Whereas if you are putting up for a multi-member constituency like in councils using first-past-the-post, the candidates can turn out to be all men who are over 70, who have an honourable career in local politics, but you couldn’t put a piece of paper between them, in terms of what their appeal to the electorate.

Vivienne Sugar

We have heard from witnesses that a sense of identity with the local community is one of the problems of the current additional members. However, your paired constituencies bring together some interesting combinations, where people would not necessarily identify with the same issues. I wondered the extent to which these pairings could actually change the way that people vote. The obvious one is putting Gower and Llanelli together.

The other comment is that this still uses Westminster constituencies that are due to be reviewed in the future. The other model uses unitary authorities that may also be subject to review in the future.

Dr John Cox

I do not want to be shot down for what has been done for the pairing of the Westminster constituency because that is a bad basis on which to have the STV debate. If, for no other reason, that parliament may well have boundary changes, then the Welsh Assembly then has to follow. Even if it is logical today, it may not be logical as a result of whatever the boundary condition comes to. I am not advocating that at all, but I have put it in as an illustration as so many other people have talked about it as being a possibility.

Sir Michael Wheeler-Booth

You are in favour of it being based on local authorities?

Dr John Cox

Based on local authorities, but not controlling in the sense that one local authority is one constituency. There is a case for pairing some local authorities, because some of them are quite small as they are.

Dr Laura McAllister

Basically you would have MPs working in one geographically area, and AMs working within quite a different one. One can imagine that functioning better if the Assembly had a different configuration of powers. Even so, there would be issues about overlap, boundaries and divides.

Dr Russell Deacon

That already occurs, under the existing system. You already have the list members, and the boundaries.

Dr Laura McAllister

But you are not defending this system.

Dr Russell Deacon

No, but we are saying that it already exists.

Ted Rowlands

Can we just reaffirm that on the Welsh campaign it takes the view that the AMS system is bad with or without open or closed lists?

Dr John Cox

Yes. We have not entered into the question of open lists and closed lists for that reason. We think that the AMS system has been a proven failure and let us move on to what the alternative is.

Lord Richard

What you are saying is that it does not provide people with decent representation.

Dr John Cox

The antagonism is evident. It is not sufficiently proportional, which is as much due to the fact that we only have 60 members. The AMS could be proportional between parties if you were to increase the numbers to 80. If you had the same ratio of list members to constituents, in Scotland, it would be relatively proportional. Proportionality as such is not the key argument against the AMS.

What is much more serious is the different types of members – they do not like being called different classes of members - it creates an additional antagonism to the antagonism which was there anyway. Particularly in the context of Wales, where you have all the government’s AMs are constituency AMs. Most of the opposition are not.

If one of my local regional list AMs tries to do anything at a constituency level, it is being treated as muscling in on a patch, in a way which would not exist if they were multi-members, it would then be legitimate. Then you have this silly, trivial argumentation about so-and-so who was defeated ignominiously in an election, but who has come back by the back door. It generally debases politics.

Sir Michael Wheeler-Booth

A fortnight ago in one of our public meetings there was a lot of opposition to the present system. Most of those opposed said, let us go back to first past the post. That is what they wanted, and that is in a Labour stronghold!

Dr Russell Deacon

One of the things about first past the post is that you can get 65 of the electorate not voting for that particular member, and that member gets in. Therefore, the vast majority of the constituency actually does not vote for their MP, yet they still get in. Under the STV system, that would not be possible.

Dr John Cox

The whole terminology of politics is turned upside down. We hear about minority and majority governments. With majority governments, in first past the post, there is usually a minority of votes. With the so-called minorities having too much say, with STV, it means that you have a majority of people who will have voted for the parties who are in government. We have reversed what are majorities and minorities against the voters in this way. What we would like to see is majority governments, meaning that the majority of the voters in the country voted for the government.

Sir Michael Wheeler-Booth

The trouble with what you are suggesting is that some of the parties really do not want it. They do not want it for reasons that are fairly comprehensible. For example, the Conservatives do not like the idea of having one pro-Euro candidate and one anti-Euro, who fight each other. They really dislike it very much, and I can understand that.

Dr John Cox

I think that we have got used to authoritarianism. I have listened to many people saying that we must have strong government. A good friend of mine, Llew Smith, makes this one of his main arguments for first past the post and he is still in favour of it to undo all the bad things the Conservatives did. I say, "Are you prepared to put up with the price of another 18 years of such rule if the clock just switches a few percent the other way"? He thinks in terms of, "we have to be in power and do what we want".

The idea that you are actually trying to work together with people, and that there is consensus politics, is almost alien to people. I have attended conferences of political parties as a lobbyist. If you did not tell me beforehand that I am just walking into the Liberal Democrat conference, I would not be able to tell half the time what the party was. The sentiments are the same, the things that motivate people are the same, they are all, at grassroots level, to do good for the community, but they have different hates. That is the difference between us, different labels.

Lord Richard

A label is quite important, actually. You can’t dismiss it quite as easily!

Dr Laura McAllister

Can you say, for the benefit of the Commission, about how STV encourages consensus in the way parties operate, because of the second preference and so on?

Dr John Cox

This is a very strong point. I have done my share of canvassing for these various hate machines as well. If you are on the doorstep, canvassing for party X, and someone says, I am inclined to vote for party Y, your judgement is, well they are so far away from us that it is not worth bothering with, or, if they are quite close to us, your judgement is let us slag them off. This is the fundamental reason why Plaid Cymru and Labour hate each other so much; it is because they are a bigger threat to each other. They are so close, that they will not admit it.

There are probably people around this table who would say that I am quite wrong. With first-past-the-post the closer a party is to your own party the more you have to find distinctions and slag them off. If you are in an STV situation, and you learn that someone may be about to vote Plaid Cymru, you can go for their second preference vote and say, we do agree on this, this and this, so I hope you will be giving us your second preference vote. If that happens at the canvassers level, at local level, then it also affects the way the party projects itself.

With STV the parties have to be conscious of the fact that they are only going to get those extra seats, which are going to win them the support they want, if they can be appealing to the parties which are closest to them. They can slag off their real enemies as much as they like, but they have to appeal to the ones that are closest to them. I think it would change the way in which politics are conducted in this country, to the benefit of us all.

Lord Richard

I see that, but if you have this system, how do you actually produce more consensus? I can see you will elevate the way in which elections will take place. But the actual results mechanism: how does that work?

Dr John Cox

We have a political culture that has been ingrained in us now for centuries. It grew from a two-party system, and is finding itself more and more strained now that we have got into three-party, and, in Wales, a four-party system. We have an electorate system that reflected a straightforward Whig versus Tory thing. Frankly, for Whig versus Tory in a two-party system, first past the post is just as proportional as any in that situation, so there is no problem. Disproportionality comes when you try to use something that only works in a two-party situation.

Our political culture is developed from there. It would not happen that, at the first assembly election using STV, that everyone would start co-operating, because we still have the whole of our last 50 years of experience. But it would at least be pointing in the direction of consensus politics and we have to take a long view on this.

Peter Price

Can I ask about constituency size? What you said about the four-party history in Wales, in your paper you refer to three to six members as generally being the best size for an STV constituency. If you have a three-party system, the lower end of that, over the whole area is proportionate. By the time you get to a four party system, the lower end of that range becomes far more under strain in order to achieve proportionality. Can you comment on the size of that generally accepted range, in terms of Wales?

Dr John Cox

First of all, when I say generally accepted, it is based upon experience in Ireland and the research that people at the Electoral Reform Society have done. That may not be totally applicable to Wales. If you looked at it purely mathematically, it is true that for a four-party system you really do need to have a minimum of four seats. On the other hand, recalling what I have just said about first past the post being fine in a two-party system, you do not necessarily get the full disproportionality that is implied mathematically. Where one party loses out in one constituency, it might be offset by the votes in another.

Disproportionality is never as bad as the mathematics suggests for individual constituencies. I think that there is basically a compromise. Even with a figure of six members elected, the ballot paper is going to be fairly large. Although you get more proportionality with six members than you do with five or four, you do have to make a pragmatic compromise in this case. If you went for the minimum size being four, then you are going to end up with fairly large constituencies in some cases geographically, and longer ballot papers than are strictly necessary. I think it is a pragmatic choice, and I am going along with the Electoral Reform Society, and what the general experience in Ireland is, that they have said that three to six is about right in their situation.

Vivienne Sugar

Can we go back to your assertion that consensus is a good thing. My experience is in local government. I can point to a number of local authorities where there is no overall control, and they are dysfunctional. They are incapable of making decisions: they are inward looking. They spend all their time vying with each other for positions and so on, rather than a having common vision for the area. Plus, if consensus also means coalition, which it may or may not do, it is about the basis on which you vote. Do you vote for the manifesto of the political party of your choice, or do you vote imagining what sort of coalition programme may come out of whatever the results might be? I would like to hear your comments on that.

Dr John Cox

First of all, I agree that I am very prejudiced in favour of consensus. I am sure you do not want me to hear me rationalise my prejudice in favour of consensus. I just think it is better than the type of conflict that we have with the present system.

Lord Richard

Do you use coalition and consensus to mean the same thing?

Dr John Cox

No, I do not. I am not terribly happy with the coalition, because that has within it the implications that there are people who actually are opposed. It is a fair point. Things such as that famous Lib-Lab pact, there were definitely people who did not agree, who were coming together for what I would call political opportunism. Important as it might be.

Sir Michael Wheeler-Booth

In the 1970’s?

Dr John Cox

Yes. The experience of Ireland is that people who came together because of the necessity of the arithmetic, have actually come closer together. There have been mergers taking place in consequence of people having worked together in this way.

From the point of view of the voter, the voter is doing no more than the obvious thing of voting one, two, three, four, in the order of preference that they have for the candidates who are on the list. It is then up to the parties, when the people have been elected, to reflect this. I am saying that it is placing a constraint upon people that they should try and work together because it could help them. But it is not actually pushing people who are in complete opposites into coalition with each other. That would be in contradiction to the way in which they would have had to have conducted their election campaigns.

Ted Rowlands

From my own experience, people are more likely not to vote because they do not feel that they have any choice, because there are no clear differences between the candidates. You want to create a consensus whereby it would be very difficult to find out what was different between the parties and what they represented.

In many ways, political engagement is caused by commitment to issues and policies that are different. Adversarial politics are just as likely to create a better turnout than consensus politics. On top of that, you have the cynicism that arises from a coalition government, that you have to do some kind of training afterwards to try to create a viable government. The combination of that cynicism and the lack of choice by this consensus politics is much more likely to turn people away from voting than the electoral system itself.

Dr Russell Deacon

It may be true to say that political parties are coalitions within themselves. You can see at the moment, for example, the Labour government is negotiating with its own side over various issues. It means that with more open coalitions, you can actually see where they come from. At the moment, within the parties you cannot tell. This makes it more overt, as far as the electorate is concerned, so you know where the opinions are.

Dr John Cox

It was all much simpler for me when I was a youngster. Today you have a situation where we have a Conservative Party that is in favour of not having fees for students, and a Labour government that is. There are very few people in the electorate who I know who would have imagined this sort of thing to arise.

I believe it is a natural consequence of trends, in that all parties are trying to appeal to the same very small social group, and opinion group. When one party goes into that, they now have to find something that they disagree on. It so happens that the Conservative Party seems to have flipped and are taking the position which all my upbringing would have told me the Labour Party would be taking.

That is the sort of thing that definitely creates a cynicism among the electorate because they no longer know where the principles are, because they are all battling to try and appeal to the same group, and that group is not in Wales. If you were to put the same question to an electorate in Wales, you would have a completely different answer than you would in middle England.

Tom Jones

Your system in a way goes back to the local representation. How would people relate to Wales as a whole?

Dr John Cox

I am not sure I am totally with this question. Is it your fear that because everyone is going to be elected for a larger unit than the present constituencies, though a smaller unit than the present regions, that they will be concentrating their attentions just on that new locality, rather than the interest of Wales as a whole?

Tom Jones

Does the system encourage people to think strategically?

Dr John Cox

There is that possibility, if you were basing things upon the local authority that they will think like local authority representatives. The alternatives seem to be that, if you try to base your unit on Wales as a whole, you are into things like the full lists system, you are not going to get local representatives, which is something which I believe is very important, the constituency link. One has to simply hope that the people you elect for a limited part of Wales do actually have a view about Wales as a whole.

Lord Richard

We have heard evidence, and we expect more this afternoon, to the effect that the list members should not be allowed to pose as constituency members. If you have three-member constituencies by STV, then who does what?

Dr John Cox

I have had the experience of being elected to a three-member constituency as a councillor. As I was not a member of the Labour group, I found that everyone who was not a Labour voter seemed to think that I was their representative.

One aspect of this is that if you have a variety of people elected to a constituency, more of the electorate would feel that one of the members relate to them. I had people from completely different wards come to me because they felt they would get the service.

Equally, I am quite sure that there are people who would not have dreamt of coming to me, but went to the Labour members because they felt they were the most appropriate. That is an advantage, that you have a variety of different representatives. The electorate would then find it easier to relate to them.

The most classic case I had was when one of my fellow councillors delivered a very strong speech against unmarried mothers, and such like. There were a fair number of constituents who could not face ever going to this person for help with their problems. It is important that as many constituents as possible feel that there are representatives who will be sympathetic to their case.

Dr Laura McAllister

Following on from that, one of the strengths of the additional member system is that because it gives an enormous amount of power to the parties, in terms of candidate selection, they have been able to promote particular kinds of candidates, with women being the obvious one.

Under STV it is harder to promote currently under-represented groups, other than allowing the electorate to make logical choices. It is all the debate about descriptive representation. Would women, for example, naturally vote for a woman candidate from the Labour Party ahead of a male?

There is no evidence to suggest that that is the case at the moment. How do you address some of those issues, having got to this point of gender parity, particularly in Wales?

Dr John Cox

First of all, the gender parity has come more from the decisions of the three political parties to promote women. It does not have anything to do with AMS as such.

Dr Laura McAllister

But the system allows the parties to do that.

Dr John Cox

Yes. But basically it is the party’s selection that has led to a situation where we have the parity with women at present. The reason why they came to the decision was basically because of the overall pressure on the parties, in the light of changing public opinion that women have to be better represented. STV in this context would be neutral in so far as a party deciding on a list of four candidates, it is highly improbable that they are not going to have two men and two women. Any party will be thinking about this.

Equally, if you are putting up a list in an area which included a large ethnic minority, any party would be crazy not to consider having at least one on their list from an ethnic minority. Even if they were not going to get them all elected, they would be less likely to get votes for any of their candidates, if they were the one party who did not have such a candidate.

The last point I would make is on this question about whether we could go backwards. There was a situation many years ago when I was chairing CND when I made a ruling that everyone had to vote for one man and one woman equally. We then got a majority of women elected as officers, because there were fewer who had been nominated. They have never looked back.

The moment you have the situation, as we have in Wales, where there are people who are incumbents, they are very unlikely to be not elected in any election system that follows. We are there in terms of parity on gender whatever election system we have from now on.

Ted Rowlands

Can I check that on your comparative disproportionality, it is easy to see how you would have calculated the AMS. On STV you need to make certain assumptions about certain preference points. What assumptions do you need?

Dr John Cox

I did not make any assumptions; I did something completely different. I calculated the theoretical disproportionality that comes about purely from the fact that a proportion of the electorate will not have voted for the winning candidate, which is basically the quota. 20% of the electorate will have failed to be electing a successful candidate, if you had had four candidates. That is your potential disproportionality.

Your actual disproportionality does depend upon second preference votes and the accident of whether two or three votes went one way or another in a number of constituencies. I then used that same calculation to work out the potential disproportionality in the Irish elections, for the last eight Irish elections. I worked out what the actual disproportionality was in Ireland, and found myself a nice empirical factor which says, this is a proportion of the potential disproportionality, which if multiplied by the theoretical disproportionality gives actual disproportionality, and that is what I have got.

Peter Price

I followed it!

Dr Russell Deacon

Could we have that in writing?

Dr John Cox

I have a maths first degree, you may have realised. In my justification, the point that you see in the one case, for the two-member for each Westminster constituency STV option, where I tried it with the other technique, turns out to be right in the middle of the range that I did when I tried out the empirical factor version. I was heartened by that result.

Lord Richard

I thank you very much for that.

David Morris

Part of this consensus politics leaves me rather cold. Thank you very much indeed.

Dr Russell Deacon

Can I ask that our presentation, which was not seen, be submitted as additional evidence?

Lord Richard

Certainly.

(The presentation just referred to appears next)