Analysis ... South Wales Central

Following my analysis of the constituency and regional seats in South Wales West a few weeks ago, here is my long overdue analysis for South Wales Central.

South Wales Central

1   Cardiff Central
2   Cardiff North
3   Cardiff South & Penarth
4   Cardiff West
5   Cynon Valley
6   Pontypridd
7   Rhondda
8   Vale of Glamorgan
Regional seats
1   Plaid Cymru
2   Conservative
3   Plaid Cymru
4   Conservative

Six out of the eight constituency seats are currently held by Labour, but the LibDems hold Cardiff Central and the Tories hold Cardiff North. Plaid Cymru and the Tories each have two regional seats.

The Constituency Vote

It is very hard to see Labour losing any of the seats they hold. But strange things have happened in politics before, and some strange things are happening at the moment in Pontypridd. In the 2007 election Labour got 41.8% of the vote with the LibDems in second place with 27.6%.

Given the general fall in the popularity of the LibDems as a result of them being in coalition with the Tories in Westminster, you'd expect Labour to feel safe; but they are clearly rattled by the tactics of the LibDem candidate, who they are accusing of harassment. One thing we should know from previous elections is that the LibDems are not above, what shall we say, "robust" local techniques. But having complained six times before and had each one dismissed, to do it again does appear to be an over-reaction. It's very unlikely, but Labour might just succeed in throwing this seat away.

Plaid Cymru did relatively well in Rhondda and Cynon Valley last time with 30.1% and 27.9% of the vote. But in both seats Labour were comfortably in the upper 50s and should therefore be safe.


The two interesting seats are Cardiff Central and Cardiff North. In 2007 Cardiff Central was won by the LibDem's Jenny Randerson with 51.8% of the vote, way ahead of Labour with 21.9%. She is not standing this time, and the LibDem candidate is Nigel Howells. The two factors against him are that he is less well known (though he is a Councillor) and the general drop in support for the LibDems. The two factors in his favour are the previous margin of victory and the fact that the LibDems can fight very intense and effective local campaigns even when their national showing is poor.

In reality the result actually matters much more to Labour than it does to the LibDems. For Labour, their only chance of winning a majority in the Assembly is to win more constituency seats. They cannot win any list seats in any of the three south Wales regions. But if the LibDems lose this seat, they do stand a chance—though no more than a chance—of picking up a regional seat instead.

This means that those who don't want to see Labour win a majority are going to have to vote tactically for the LibDem candidate, even if it means holding your nose as you do it. This is especially true for those who would normally vote for Plaid, the Greens or the Tories. If the LibDems win this constituency, it means that they will win one fewer regional seat than they could otherwise win ... and that seat will then go to either Plaid, the Tories or the Greens. It is a no-brainer. If you live in Cardiff Central and would normally vote Plaid, Green or Tory, you would be mad not to vote LibDem for the constituency, and then for your preferred party in the regional vote.


The scenario is almost exactly the same in Cardiff North, though not quite. Jonathan Morgan won this seat for the Tories in 2007 with 45.3% of the vote, with Labour second on 30.9%. That margin would be safe enough in most circumstances. Factors in Labour's favour are that it is a smaller margin to overcome than in Cardiff Central and the relatively high profile of Julie Morgan, who used to be the MP for this seat until she lost it to the Tories this time last year.

Factors against Labour are that Julie Morgan will be 67 years old this year, and with this term set to last for five years rather than the normal four, some might well ask why she has not taken the opportunity to join husband Rhodri in a happy retirement. The answer is probably that only someone with her profile could hope to beat Jonathan Morgan. Yet it does beg the question of how short people's memories are. Can people in Cardiff North really have forgotten why Gordon Brown's party was voted out of office in Westminster? And do people really think that the cuts in our block grant now would have been at all different if Labour were still in power? Labour are blaming the Tories for what they would have had to do anyway.

The consequences for the regional vote are slightly different in this seat. The Tories currently hold two regional seats as well as this constituency seat. If they lose Cardiff North it is all but impossible for them to win a third regional seat to make up for it. If you're a Tory, this makes it all the more important that you hold onto this seat, and the Tories are and will be putting a lot of their energy and resources into keeping it.

Yet the result will also matter for Plaid, the LibDems and the Greens; for although the Tories will not gain a regional seat if they lose this constituency, there is a very real chance that they will still lose a regional seat even if they hold it. But if they lose Cardiff North, then they will almost certainly hold on to their two regional seats. This means that it is absolutely vital for supporters of these parties who live in Cardiff North to vote Tory; for doing so actually increases Plaid's chance of holding on to our two seats, and increases the possibility of the LibDems or the Greens winning one.

I certainly don't underestimate how hard it will be for people who hate the Tories (or at least strongly dislike their policies) to vote Tory. But it's going to be a tight contest, with the bookies currently showing Labour slightly ahead; so we need to make it clear that voting tactically for the Tories will help the chances of your first choice party. Doing it will feel like a betrayal of all your principles, but it really is the best way of both helping your own party and of preventing Labour from getting an overall majority.


The Regional Vote

Now let's turn to look at the way the regional vote works in more detail. Seats are determined by the percentage of the vote on the second ballot paper, but seats already won at constituency level are taken into account. After a seat has been won, the percentage necessary to win another seat is the original percentage divided by two, then three, then four, etc. This is how things worked out in 2007:

Labour ... 36.7%
Conservative ... 23.4%
Plaid Cymru ... 16.7%
LibDem ... 15.7%

Round 1 ... Labour 36.7% ... already won
Round 2 ... Conservative 23.4% ... already won
Round 3 ... Labour 18.3% ... already won
Round 4 ... Plaid Cymru 16.7% ... regional seat 1
Round 5 ... LibDem ... 15.7% ... already won
Round 6 ... Labour 12.2% ... already won
Round 7 ... Conservative 11.7% ... regional seat 2
Round 8 ... Labour 9.2% ... already won
Round 9 ... Plaid Cymru 8.3% ... regional seat 3
Round 10 ... Conservative 7.8% ... regional seat 4

The most important point to note is that Labour, because of the seats already won at constituency level, have absolutely no chance of winning a regional seat. Even if Labour were to lose Pontypridd this time round and fail to win both Cardiff Central and Cardiff North, they would still have no chance whatsoever of winning a regional seat. This means that a regional vote for Labour is a completely wasted vote. Labour supporters should instead vote for the party they rate second after Labour.

For Plaid supporters in South Wales Central, it is the regional vote and only the regional vote that matters. This may sound obvious, but people would be surprised at how many voters vote for their first choice party at constituency level, but think that they should give their regional vote to a second choice party. Labour supporters should do this in all three of the south Wales regions, but not Plaid supporters.


Predicting the results this time round will depend on the percentages each party gets and whether they have won a constituency seat. Much as I hate to say this, the order is probably going to be Labour (because most of their supporters won't realize that they are wasting their vote, and Labour certainly won't tell them that they are) followed by Plaid and the Tories neck-and-neck, the LibDems, and then the Greens and UKIP neck-and-neck.

With that order, and excluding the Labour vote because it is irrelevant, Plaid will hold onto their two regional seats if any of these things happen:

•  They get more of the vote than the Tories
•  They come second to the Tories, but the Tories hold Cardiff North
•  They come second to the Tories and the Tories lose Cardiff North; but Plaid get more than twice the votes of any other party
•  They come second to the Tories and the Tories lose Cardiff North; but the LibDems hold on to Cardiff Central

I'm pretty confident that one of these four things will happen, so our two seats will be safe.

After that, things get more complicated, and perhaps too complicated to list. The Tories stand a chance of holding on to their two seats, even if they hold Cardiff North; and if the LibDems lose Cardiff Central they might not win a regional seat to make up for it.


As for the Greens, they reckon South Wales Central is their best chance of winning a seat. I've got a lot of respect for most Green policies and would like to see them pull it off, but it's a big ask.

The only way they can do it is if the LibDems hold Cardiff Central and the Tories hold Cardiff North, as this effectively moves them up two steps. But they would also need to get all three of these things as well:

•  more votes than UKIP
•  more than half the vote of the LibDems
•  more than a third of the vote of the Tories, or more than half the vote of Plaid

The latest YouGov poll shows Plaid on 18%, the Tories on 14%, the LibDems on 7%, and both the Greens and UKIP on 4%. I think that probably puts the Tories too low, yet if the Greens can get 5% then they will win the fourth seat providing the LibDems get less than 10% and the Tories get less than 15%. If they can make it to 6% then they will get the fourth seat if the LibDems get less than 12% and the Tories get less than 18%. That's well within the realms of possibility.

Because supporters of the other parties besides Labour should all vote for their first choice party on the regional ballot, everything will depend on those Labour supporters who are aware enough to realize that it is pointless for them give their regional vote to Labour. For their vote to count, they have to give it to their second choice party, and it's safe to guess that they will be much more inclined to vote either Plaid or Green, as these are the other left-leaning or—as Labour would say—progressive parties. Even if only some of that Labour switch goes to the Greens, it should be enough to get them a seat.


The Prediction

So now for my final prediction. I reckon that Labour will hold onto their six constituency seats; that the LibDems will hold on to Cardiff Central by a reduced but still comfortable margin; and that the Tories will just, but only just, hold on to Cardiff North.

Plaid will hold their two regional seats, irrespective of whether they come ahead of or behind the Tories; but the Tories will lose one of their two regional seats to the Greens.

Bookmark and Share


Anonymous said...

I'd hate to see David Melding lose his seat.

Welshguy said...

David Melding is my favourite Tory, but he's still a Tory. If it's a choice between keeping him and keeping Chris Franks, I know which one I'd choose (or Jake Griffiths, for that matter).

MH, I'm not sure Chris Franks is as safe as you suggest. If Labour win both Cardiff North & Cardiff Central (possible, likely even), and the Tories get more votes than Plaid (very likely) then the fourth regional seat will go to the Liberals/Greens/Ukip as long as they get half the Plaid vote. If the Plaid vote drops a lot then that's not so unlikely.

Also, you didn't even mention the Vale of Glamorgan! Jane Hutt's majority last time was only 83; if any of the formerly Labour seats change hands it will be the Vale. With polls as they are I suspect she will hold on; but it would certainly damage Labour's chances if she didn't. If the Tories won both Cardiff North & the Vale of Glamorgan then Chris Franks would be pretty certain to keep his seat. Arguably, Plaid / Liberals / Greens etc should vote Tory in the Vale as well.

Anonymous said...

Everyone likes David Melding, but at the end of the day it is his party that like Labour won't give Wales a single metre of electrified railways, would charge Welsh students thousands of pounds for their education, and would target the vulnerable rather than those who could pay as a way of getting out of the recession. Melding isn't in tune with the majority of opinion in the Conservative party in Wales, they are still fundamentally a British party and the vast majority of their supporters and possibly activists oppose Welsh national ambitions. Nationalists need to be careful about the Tories- they might use Welsh on their manifesto cover but at the end of the day they won't even appoint any of their Welsh MPs as Secretary of State. It should take more than that to fool us!

Welshguy said...

I agree, anon. - I'm glad I don't actually live in Cardiff North / Vale of Glam and have to face the difficult choice of tactically voting.

I'm increasingly coming to the conclusion that the Additional Member system, at least as it's operated in Wales, is a pretty crap electoral system. People complain that AV is complicated; but at least people don't need to understand how it works in order to use it properly; wheras the tens of thousands of Welsh people who keep voting Labour on the list are testimony to how ridiculous AM is.

MH said...

Yes, I'd agree that David Melding would be a loss to the Conservative Party and to the Senedd. He is a thoroughly decent politician. But we should remember that he used to be first on the list, and it was the Tories themselves who decided to demote him. They obviously prefer the "attack dog" technique of Andrew RT Davies ... and probably think that Melding has shown too many signs of "going native".

It's sad. But the reasons why we non-Tories tend to have time for David Melding are probably the very same reasons that the Tories themselves have for demoting him. John Osmond wrote a nice article about him here.


Reading that article again does raise the question about the Vale of Glamorgan. For John seems to think its winnable for the Tories, as does Welshguy in his comment.

I know something about the seat through my friendship with Ian Johnson, the Plaid candidate. A few years ago he put together a plausible scenario under which Plaid could win the seat with less than 30% of the vote in a tight four way contest. But I think the ground has now shifted. Labour are gaining votes because of anti-ConDem sentiment, both from voters who usually swing between Labour and Tory and from those who swing between Labour and the LibDems. So what was a highly marginal seat now appears to me to be a marginal-to-safe Labour seat.

If the Tories had been cleverer, I think they would have chosen David Melding as their candidate for the VoG precisely because of his wider appeal to non-hardline Tories. But as John Osmond put it, they appear to want to be Tories rather than Welsh Tories. But that's their choice. However if the Tories were to pull off a victory there it would certainly be a good thing in helping to deny Labour an overall majority (for they still wouldn't get a regional seat) but would also mean the Tories could only get one regional seat.

MH said...

As for Plaid's second seat, the scenario Welshguy outlined is possible. But it would be the only way that Chris Franks could lose his seat, and any one of the other four scenarios I outlined would see him safe. So I'd put his chances at 80-20.

I don't think Plaid's vote will drop much, if at all. It was 16.7% and I think it's more likely to go up a bit than down. 18% seems about right to me, and I would be very surprised if it went below 15%, but that means one of the crucial factors will be whether this is more than twice the LibDem vote.

Although I have said that Plaid do not stand a chance of winning a constituency seat in this region, that doesn't mean that Plaid won't do well. This time round, the name of the game is for the Plaid constituency candidates to realize that their main job is the increase the regional vote. None of them should allow themselves to think they can win a seat. Now of course I don't expect any of them to put it in those terms in this campaign; it is probably easier to say "vote Plaid twice" providing they don't say that to Plaid supporters in Cardiff Central or Cardiff North. But whatever else they do, they must stress the importance of people giving their regional vote to Plaid.


Finally, I fully agree with Welshguy about the difficulties of this voting system. Tactical voting is a difficult concept to understand ... and an even more difficult thing to base a campaign on. It will not be immediately obvious to the average Plaid or Green voter who lives in Cardiff North that the best chance for their party to get a regional (or another regional) seat is by voting Tory ... especially when maybe only a couple of streets away in Cardiff Central the message is for them to vote LibDem rather than Tory.

The additional member system we currently have increases the necessity to vote tactically for the FPTP constituency seats. All of this tactical complexity would be done away with at a stroke if we elected the Assembly by STV.

Welshguy said...

Re: voting systems - The real measure of the complexity of a system is how easy it is for a voter to understand how she/he must vote, in order to best bring about the scenario that he/she most desires. In this respect, STV isn't that complicated at all; no more so than AV.

I'm hoping that, in the meantime, when we go down to 30 constituencies at Westminster, the same happens in the Assembly (with the extra 10 seats reallocated to the list) - the regions would have 6 constituencies and 6 list members, making the overall results pretty much proportional.

Re: Chris Franks; are you sure the Plaid vote will go up / hold in South Wales Central? All the polls seem to indicate otherwise, although I suspect they probably over-estimate the Labour vote-share (and possibly correspondingly under-estimate Plaid's).

Even if the Tories keep Cardiff North, the fourth seat could well fall if the Lib Dems lose Central but otherwise don't collapse altogether.

MH said...

Sad to say, WG, although your solution of reducing the number of FPTP Assembly seats to 30 and increasing the regional seats is logical and easy, it isn't what's planned. The idea is to "decouple" and have two different sets of boundaries ... which is the current situation in Scotland following their reduction in MPs from 2005.

I think it will be a mess. Yet the Tories and LibDems were so anxious to get the reduction through that they didn't think through the consequences (or thought they didn't matter) and Labour want as many FPTP seats as possible because they benefit from it.


On the level of Plaid's vote in SWC, the latest YouGov poll puts Plaid at 18% in SWC. It's a small sample size, and therefore isn't that reliable. The previous poll put us at 16% and the Tories at 20%. I think we'll get less than the Tories, but we'd have to get a lot less (i.e. half of our vote would need to be less than a third of theirs) before it could affect us. I can't see the Tories getting more than 22.5% or us less than 15%. Obviously we can't take anything for granted, but I don't think we should be pessimistic.

Welshguy said...

Let's hope you're right...

Anonymous said...

I don't understand why Plaid and Lib Dem voters should be so squeamish about voting for Tories.

The issue at this election is whether voters want a Labour or Tory government. A vote for Plaid or Lib Dem entails the possibility of a coalition with the Tories. If a coalition with the Tories is viewed as a positive outcome then how can voting Tory be such a painful process?

The only way of ensuring Wales does not have a Tory government is to vote Labour.

MH said...

The "only way"? That's a rather blinkered view, Anon. The earth may revolve round a big orange ball in the sky, but not everybody shares Peter Hain's view of the world revolving around him and his "it's either us or the Tories" mantra.

Other ways of "ensuring" that Wales does not have a Tory government are for Labour not to have an unfettered majority, and therefore be forced to form a coalition with either Plaid or the LibDems.

Anonymous said...

I want a One Wales II govt - that't be the best outcome for Wales.

If I were living in Cardiff North I'd be tempted to vote Tory to make sure Wales doesn't become a One Party State under Labour - which it will. The turgidity of anther Labour State, constantly ground-standing and wind-bagging with anti-Tory racism is too depressing to comtemplate.

Like-wise, I'd vote LibDem in Cardiff Central.

Welshguy said...

Are you sure that the constituencies will be "decoupled", MH? Vaughan's just posted as if the more likely outcome will be bigger lists, as it's up to Westminster rather than Cardiff; & the Tories/Lib Dems will undoubtedly benefit from bigger lists.

Cibwr said...

As things stand, yes the constituencies have been decoupled. There is no mechanism for reviewing their boundaries though so I think this is a one election fix. It leaves the door open for either a full review and increasing the list component or to consider changing the system to STV. Don't hold your breath on this though.

MH said...

Interesting, WG. I need to look into it further. Section 13 of the Act does specifically decouple the Westminster and Senedd boundaries, as Cibwr says, leaving the Assembly boundaries as they are. I have to admit that I took that to be a permanent arrangement; but from what you, Vaughan and others have said, I now realize it might not be. It could equally be a "let's take a while to think about it" arrangement. That makes some sense in that the result of the referendum on AV is not yet known, and it would seem a little bizarre to keep FPTP for the constituency element in Wales if we vote for AV in Westminster.

Now without wanting to sound too despondent, I'm not hopeful about the AV vote. And there will be a real question about how to keep the LibDems happy if they lose on AV. STV has been talked about as a means of electing a new House of Lords, so the Tories in Westminster might not be implacably opposed to it for the Senedd, just for the Commons.

Of course you're right to say that the Tories and LibDems would want more proportionality in the Senedd, and that they could impose it ... for it's not a devolved matter. However it would help if Labour had fewer than 30 seats, for the other parties could then vote to endorse the decision, which will be happier all round.

However if things are going to change (and I repeat that things will remain as they are at 40 plus 20 unless Westminster does something to change it) then although 30 plus 30 is much better than 40 plus 20, it would be better still to press for STV. Let's go for the real McCoy.

Welshguy said...

While I agree that STV is what we should hope for eventually, let's face it, if people have this much trouble stomaching AV I don't hold out for STV any time soon.

The 30/30 proposal however doesn't really involve changing much in terms of the system, while it would make a real difference to the results and deal with a lot of the issues of tactical voting - Scotland has much less of this problem, as their 7-person lists mean the constituency results are less important. Best of all, it is such an elegant solution it would be difficult to argue against it; though no doubt Labour will anyway.

A difficulty would come if and when the Assembly is to be expanded, as then you'd be looking at having far more list seats than constituencies, or be forced to have different boundaries or STV. But I can't see that happening soon either.

How about a post showing possible seat totals for a 30/30 assembly? Possibly using the boundaries you posted a while back. One for us electoral-map-nerds...

MH said...

It was Penddu who came up with the good ideas on boundaries, WG, here and here. In a sense, there's no need to do the exercise again because the Boundaries Commission is going to do the job anyway ... but then again, the BC didn't come up with very convincing proposals last time.

But whatever, in time for the 2015 election we will have 30 Westminster constituencies, no matter how many fights and arguments there are along the way. Within this framework, if we have a 60 member Assembly and keep the AMS, then 30 plus 30 is not bad.

A 60 member Assembly under STV would, if it used that framework, either pair Westminster constituencies together and elect 4 AMs to each (or group them in threes and elect 6 AMs to each ... or a mixture of the two).

When the Assembly is increased in size (and I think this should only happen when more areas of responsibility are devolved to Wales to justify it, and does not have to be to 80) these 15 paired constituencies could each elect 5 AMs to make 75 (or 10 tripled constituencies could each elect 8 AMs to make 80, but I think such large constituencies should the exception rather than the rule).

Perhaps we should wait and see what the reaction to the 30 Westminster constituencies is. The tolerances are very rigid, and if the feeling is that being so rigid cuts right across our natural and perceived boundaries then (even though we'll have to live with it for Westminster) it might well be better to do something different for the Assembly. But if it turns out to be broadly acceptable, then it would be better to work within the same framework for both.

Post a Comment