The Electoral State of the Parties, 2: the Conservatives


In the second of these reviews of the electoral state of the parties in Wales, I turn to look at the official opposition in the National Assembly, the Conservatives. In my discussion of the party this time last year, I suggested that “[t]he first fifteen years of devolution saw steady Conservative electoral advance from the nadir of 1997. The next few years may well see the party much more on the electoral defensive.”


In that context, Welsh Tories could well be forgiven for viewing the past twelve months with some degree of satisfaction. Despite having been in government in London since 2010, and taking the lead in implementing public sector austerity, the Conservatives have continued to display a rather impressive electoral resilience. Although it was never remotely likely that May’s European election would see the Tories repeat their 2009 feat of coming first in the popular vote in Wales, the Conservatives held their seat in the European Parliament, and lost only a modest amount of the vote share they had won five years previously. Moreover, they managed this in the face of a surge in support for UKIP that might have been expected to hit the Conservatives hardest. The opinion polls, too, have seen the Conservatives’ support holding steady at a level that, while a little below the vote share they won in 2010 and 2011, is still by historic standards a very respectable performance for the Welsh Tories. There is certainly no sign that the growth in UKIP support is yet causing substantial harm to Conservative support levels in Wales.



General Election

Assembly Constit.











However, the Tories’ prospects for further electoral advance currently seem limited. While they seem to have a strong grip over the electoral allegiance of roughly one-fifth of the Welsh electorate, among much of the rest of the population they are simply not a viable option. Writing in 1984 about Plaid Cymru, Denis Balsom and colleagues contended that “Plaid Cymru must exist, but cannot grow; it is a sturdy, dwarf plant” (‘The Red and the Green: Patterns of Party Choice in Wales’, p.323). What was true about Plaid Cymru in the 1970s and 1980s now seems a more appropriate description of the Welsh Conservatives as an electoral force: a party that has a strong and highly resilient base of support, but great problems reaching out much beyond that base to the rest of the electorate, much of which views the party with hostility.

Furthermore it is not only in terms of electoral advance that the Welsh Conservatives’ prospects appear limited. The party also appears to have rather limited prospects of actually doing anything much in Wales with the electoral support they do possess. In a multi-party system, such as we have for the National Assembly, it is not only success in electoral competition that matters; also of great importance is where you position yourselves in relation to the other parties. In the latter respect the Conservatives undoubtedly have problems.


Under the current regime, the Tories in the Assembly seem to wish not only to perform the classical opposition role of scrutiny and criticism of government, but more broadly to challenge much of the centre-left consensus they perceive across the other parties currently represented in the Assembly. In many respects this is probably healthy for Welsh politics: consensus assumptions should be subject to challenge. But is it healthy for the party’s own prospects of ever wielding power in Wales?


There currently seems no chance of the Tories being in power in Cardiff Bay under any remotely conceivable circumstances. It would be no surprise if, in the 2016 Assembly election campaign, Labour seek to shore up their support on the left by reviving the bogeyman of a Plaid Cymru-Conservatives coalition government. But as I commented last year, under the current leaderships of the respective parties, there seems no basis for thinking such a coalition a remotely realistic proposition.


The prospect of permanent opposition is a source of evident frustration to some Assembly Conservatives, at least some of whom could rightfully regard themselves as at least a match for their Labour counterparts in talent. Perhaps that frustration helps to account for the internal difficulties that the Assembly group experienced in recent months? For the moment, though, the Conservatives’ little local difficulties within the Assembly seem to have had minimal public resonance.


The party’s main task over the next few months will be the general election, where they are defending several seats gained in 2010. There seem few obvious possibilities for further seat gains (although in the event of a real Liberal Democrat meltdown Brecon and Radnor could well come into play); but in a close election overall, retaining as many as possible of the eight seats won last time could have a UK-wide importance. Cardiff North, won by fewer than 200 votes last time, and where the retirement of Jonathan Evans deprives the Tories of any incumbency advantage, will be very difficult to hold. In all their other seats, however, the Conservatives would appear to have at least a decent chance of holding their ground against a Labour tide now looking rather less overwhelming than had seemed likely a year ago. The potential joker in the pack is UKIP: their support could reach levels that, while not being likely to win UKIP any seats in Wales, might very well affect who does win some seats.


A good UK election performance will not, of course, resolve the Tories’ longer-term strategic problems within the National Assembly. But it could just make an important difference at UK level. The Conservatives may well be on the defensive in the coming general election, but their prospects of it being a largely successful defence look distinctly stronger than they did twelve months ago.

The Electoral State of the Parties, 1: Labour

As I did last year, I’m going to use this period – roughly the end of the summer holidays, with the Assembly due to reconvene soon and the main party conferences also approaching shortly – to take stock of the electoral performance and prospects of the main political parties in Wales. How have they fared over the preceding twelve months, and in what state are they to face the challenges that lie ahead – most obviously next May’s general election?


I begin with Labour. In Wales, electoral and party politics always begins with Labour. I’ve discussed at various points on the blog (most obviously here and here) the long history of Labour hegemony in Wales – now extending to almost eight full decades. In my review of the party’s electoral state this time last year, I also observed how, after a few rocky years up to 2010, Labour’s dominance in Wales appeared wholly resurgent. In that context, Labour’s electoral prospects looked very bright. Looking forwards, I asked:


So will the good times simply continue rolling for Labour? In the short term, the answer is probably Yes. The party is likely to finish a good first in the 2014 European Parliament elections in Wales, and is currently also on course to make several seat gains at the next UK general election.


Well, it hasn’t quite worked out like that, has it? There have been two rather significant blows to Labour’s dominance in Wales over the last twelve months.


The first came in the only major electoral test facing the parties during 2014, the European Parliament election in late May. Until very close to polling day, there seemed little doubt that Labour would top the poll in Wales by a substantial margin; nor did there seem much doubt that Labour would secure two of Wales’ four MEPs. Indeed, some of the opinion polls suggested that Labour was even in with a chance of winning three of the four seats. In the event, though Labour did secure the most votes of any party, its 28.15% vote share was a mere 0.6% ahead of UKIP, and was the second worst vote share for Labour in Wales at any major election since 1918. (Only in the European election of 2009 have they done worse). Labour won only one of the four Welsh seats in the European Parliament.


Of course, we must remind ourselves that European elections are very much not the same thing as general elections, or even devolved ones. Indeed, across Europe, elections to the European Parliament seem increasingly to attract votes for parties that are not serious contenders for domestic power. But those Labour supporters tempted to dismiss their disappointing electoral performance in the European election for such reasons should also bear in mind a second feature of politics in Wales over the last twelve months – the significant decline in Labour’s position in the opinion polls. Thankfully, we now have more regular polling in Wales, which makes it easier to chart trends in party support. The most obvious such trend over the last 12-18 months – more obvious even than the advance made by UKIP – has been the slippage in Labour support levels. Here are the yearly average ratings from 2012 for Labour, for both the general election and Assembly constituency vote:



General Election

Assembly Constit.











Labour’s performance in the polls in 2014 has hardly been terrible. What would any of the other parties in Wales give to have levels of support even close to those experienced by Labour through 2014?! But Labour’s aura of invulnerability has been shaken. Some Labour advance on the 2010 result in Wales at the general election must still be likely (and are perhaps made more likely by the findings of polling in marginal seats, which suggests that Labour is doing better there than across Britain as a whole), but the sweeping gains that looked plausible only a few months ago are now in far more doubt. In the context of a UK general election that could be very close and where every seat could matter, Wales does not now appear to be on course to provide Ed Miliband with as many gains as would have been expected this time last year.


Labour’s poll rating for the Assembly must also be giving them some concern. The three most recent polls have each put Labour support several points below the level secured in 2011. Labour’s policy record, particularly in health and education, has come under increasingly strong criticism – some of it directed from the Conservative party and their media supporters in London – and the evidence from the polls suggests that at least some of this criticism has hit home. Public evaluations of Labour’s performance in office in Wales are not glowing.


Labour still retains some substantial advantages for the next Assembly election. They have Carwyn Jones, still by some way the most popular party leader in Wales. The semi-proportional electoral system for Assembly elections works distinctly in Labour’s favour. Most importantly of all, Labour currently have no very obvious challenger of the strength that Labour in Scotland has faced with the SNP. Nonetheless, the political context of the next Assembly election may be far more difficult for Labour than in 2011. They still look marginally the more likely of the two major UK parties to emerge as the largest in parliament after the next general election. Labour in Wales thus face the possibility of fighting the Assembly election while being linked to a potentially rather weak UK government. This would be much less promising terrain for Labour than in 2011, where they could position themselves in contradistinction to a UK government that had limited support in Wales.


So the electoral state of the Labour party, and its immediate prospects, look rather less rosy than they did twelve months ago. The long-dominant party of Wales has performed rather poorly in recent months, and seems likely to face more difficult times ahead. Still, Labour’s opponents might want to reflect that you don’t get to be the dominant party in a nation for almost eight decades just by chance. And though fewer people in Wales now seem convinced by what Labour has to offer, most of them are even currently less impressed by the other parties.


Coming Soon…

Starting next week – though probably not on Bank Holiday Monday – I’ll be publishing a series of blog posts assessing the Electoral State of the Parties. As with the equivalent series last year, I’ve timed this set of pieces for what I consider the start of the new political year: the end of the summer holidays, with the Assembly recess due to end soon and the main party conferences also looming on the horizon. Blwyddyn newydd dda i chi i gyd.

One difference with last year is that this time around there will be five pieces posted in the series, rather than four. That tells something of a story in itself. In August 2013 it still seemed credible not to write about UKIP when discussing the main parties in Wales; it no longer seems to me credible to ignore them.

I’ll start the series with Labour. In Wales, we always start with Labour. The rest of the posts will appear over the following 2-3 weeks.

This series of posts may also be inter-woven with a few bits and pieces on Scotland. You may have noticed that they have a referendum coming up next month? Whatever happens in that vote, it is likely to have some sort of impact on us here in Wales. So although the main focus of this blog is very much intended to remain on Wales, I hope you’ll forgive some minor Scottish encroachment over the next few weeks.

A brief note on constitutional insanity


We know that, barring something truly extraordinary, there will be a UK general election on 7th May 2015. That election will consume much of the attention of politics watchers over the next few months, and of course there will be discussion of various aspects of it here on Elections in Wales.

We don’t yet know what will be the result of the general election. But if the result is a clear Labour win – a result that has come to seem less likely over the past twelve months, but remains far from implausible – then there will be change of Prime Minister and installation of the new government the following day, Friday May 8th. If there is not a clear win for either of the two largest parties – a result that has come to look rather more likely over the last year – then there will be pressure from both the media and other politicians to move very quickly either to the formation of a minority government or for the successful conclusion of negotiations over a coalition government.

The rapid change-over between governments in the UK has been termed Removal Van democracy: symbolised for those of us of a certain age by Ken Clarke, the hitherto Chancellor of the Exchequer, driving a rental van the day after the 1997 general election, in which to load his and his wife’s possessions from No. 11 Downing Street.

Do many people in the UK realise just how unusual this way of doing things is? In most countries there is some sort of transition period between an election and the installation of a new government or administration. In the United States, the election is held at the beginning of November, but the President is not inaugurated until late-January (it used to be March!). This is an unusually long hiatus. But in most places there is some transition period. In 2012 in France, for instance, the second round of the Presidential election was held on 6th May, and Francois Hollande was sworn-in on 15th May. In Ireland – which like the UK has a parliamentary system of government, the 2011 general election was held on 25th February; the Dail approved Enda Kenny as Prime Minster on 9th March 2011, and he took office later that day.

The immediacy of the change-over in Downing Street is also unusual even within the UK. Within the devolved administrations, the newly-elected chambers meet some days after the election and appoint a First Minister (and a Deputy First Minister in Northern Ireland). For instance, when the SNP replaced had Labour as the largest party in the Scottish Parliament in 2007, Alex Salmond was elected by the Scottish Parliament to replace Jack McConnell as First Minister on 16th May, and sworn in on 17th May, which was some 14 days after the Scottish Parliament election.

I believe that this sort of situation used to be the case in London as well. I understand – although it is surprisingly difficult to find definitive information on the matter, and I’d be grateful if any blog readers can point me in the right direction – that in the 19th century it was the convention was for any new government not to take office until the new parliament met – even in a situation where the sitting government had very clearly been defeated at the general election.

This is one instance where other countries definitely do, and the UK in the past did, things better than the UK does today. Why do I say that? Well, let us consider the situation that political leaders are in immediately after a general election. They have been through a very draining and stressful general election campaign for the previous month (or more). They have then, very probably, been awake more or less all night after the close of the polls, following the national results coming in and attending their own constituency declarations.

Is that really the ideal situation in which to have party leaders and other senior politicians moving immediately into government, forming a cabinet and beginning to take other major decisions? Or, in the event on an inconclusive election outcome, are these the ideal circumstances in which to have leaders moving straight into coalition negotiations, and to be confronted with substantial media pressure for a near-immediate resolution of those negotiations? (On the latter point, recall the indignant tone from much of the press about Gordon Brown ‘squatting’ in No. 10, and their increasing impatience and even fury as the coalition negotiations dragged on for a whole five days).

Personally, I definitely do not think at my clearest, or tend to make my best decisions, when I am exhausted, sleep-deprived, or have just been through highly emotional and draining events (such as an election campaign and election night must surely be for leading politicians.) I very much doubt that many people do tend to think clearly, or are most likely to make good decisions, in these circumstances. (And I’m fairly sure that there is a welter of evidence from physiology and cognitive psychology that supports me here). The mental, physical and emotional state that party leaders and their close colleagues will likely be in immediately after an election is just about the worst possible condition in which to be making major political decisions. We should not be surprised if things go wrong in these circumstances; it’s more remarkable that just about anything ever goes right! It is utter insanity to expect our governments to be formed in such circumstances. Current UK practice is not merely unusual; it is also pathological.

I would propose a minor amendment in governing practice. As now, there should be a period of more than a week between a general election and the first meeting of the newly-elected parliament (in 2010 the election was held on 6th May; the new parliament met on 18th May. For most parliaments the gap has been slightly shorter than that). A new Prime Minister and government should take office on that day. The new Prime Minister could go and meet the Monarch after an affirmative vote in the new parliament. Or the new PM could ‘kiss hands’ in the morning before parliament meets for the first time in the afternoon. It doesn’t really matter.

What matters is that there should be a 10-12 day transition period after the election. This would allow some time for an orderly transition when there is a clear change of government from one party to another (as in 1997). At the moment when the new government then took office, all new ministerial appointments could be ‘ready and waiting’; rather than us having the present situation where the filling of junior posts can drag on through the first week or two (or even longer) of a government’s life.

In the event of a non-conclusive election, as we had in 2010, one might be able to have slightly less frantic coalition negotiations – the mandatory 10-12 day transition period would impose a natural timetable on events, and allow just a little more time for discussions to be conducted and reach some sort of conclusion.

In either case, it would give everyone just a little more time to play with. Most importantly of all, it might allow leading politicians more scope to do what is probably the single most important thing for them to do in the weekend following an election: get some sleep.

Plaid Cymru’s Strategic Dilemma


For members and supporters of Plaid Cymru, the devolution years have been a strange mix of achievement and failure, fulfilment and frustration. The positives are considerable. First and foremost must be the creation of an elected Welsh legislature and the significant extension of its powers after the 2011 referendum. Also of great importance is the 2007-11 coalition government, with Plaid moving from being a party of protest to one of power – a role to which it adapted with perhaps surprising ease and in which it generally performed competently. But on the negative side of the ledger, after its annus mirabilis of 1999 when Plaid twice came close to beating Labour in the national vote, the party’s electoral performance has been consistently disappointing. Plaid are currently only the third party in the National Assembly, and finished fourth in Wales in the most recent UK general and European Parliament elections. The contrast between the recent electoral fortunes of Plaid and its sister-party in Scotland (who Plaid actually out-performed in 1999) is stark.

The last year or so has produced some signs of electoral improvement for Plaid. The party’s opinion poll ratings have begun to edge upwards, both for Westminster and the National Assembly. The polls have also shown some advance in public ratings of their leader, Leanne Wood. Meanwhile, real elections have also produced a few  successes. The most striking, by far, was Rhun ap Iorwerth’s Assembly by-election victory in Ynys Môn in August 2013 (which itself followed a strong performance in the island’s local election the previous May). However, retaining Jill Evans’ European Parliament seat, in the face of strong advances from both Labour and UKIP, was also a fair achievement. With Labour’s poll ratings in Wales having moved downwards significantly over the last 12-18 months, Plaid Cymru can look forward to the 2016 National Assembly with at least cautious optimism.

However, in looking to advance, Plaid Cymru faces a strategic dilemma. That dilemma can be simply stated: that there is a fundamental tension between Plaid Cymru’s long-term objective of challenging the Labour party’s dominance of Welsh politics, and what is clearly the most sensible short-term strategy for it making a significant advance in the 2016 National Assembly election.

Leanne Wood has stated that Plaid’s long-term strategic objective is to challenge Labour as the dominant party in the National Assembly. To achieve this, Plaid will obviously need to raise their overall vote share well beyond the 18-19% won in 2011. But in addition to simply stacking up more votes, challenging Labour dominance in the Assembly will require Plaid to capture a significant number of constituency seats from Labour in south Wales. Labour won 22 of the 23 constituency seats in the three south Wales regions in 2011; whereas even a strong performance by Plaid on the list vote could plausibly secure it only two list seats from each south Wales region, or six in total. While Labour continues to dominate the south Wales constituency seats so totally (in South Wales West Labour have never lost a single contest for an Assembly constituency seat) it is nigh-on mathematically impossible for Labour to be displaced as the largest party in the Assembly, or indeed for any other party even to approach them in terms of number of AMs. For Labour’s dominance of the Assembly to be challenged, serious inroads must be made into Labour’s dominance of the south Wales constituency seats: there is simply no alternative.

But let us remind ourselves of where Plaid starts the campaign for the 2016 National Assembly election: as the third party in the Assembly, with only the 11 seats won in 2011. A general rise in Plaid’s vote share might plausibly win the party some additional regional list seats. But the scope for gains there is distinctly limited – probably at most to one additional AM in each of North Wales, South Wales West and South Wales Central. A more substantial advance will require some constituency gains. So at what targets should Plaid be aiming?

Sensible strategy is generally for parties to target the most clearly winnable seats. The three clearest target constituency seats for Plaid in 2016 are the following:

  • Llanelli, which requires only a 0.2% swing from the 2011 result for Plaid to capture
  • Carmarthen West & South Pembrokeshire, for which, although they narrowly came third last time, Plaid require only a 3.2% swing to win; and
  • Aberconwy, for which Plaid would need a 3.9% swing.


These are the only three constituency seats that look obviously ‘winnable’ for Plaid in 2016: the only seats that Plaid can capture with a percentage swing significantly below 10%.

None of these seats is in one of the three south Wales regions.

The strategic problem facing Plaid begins to come into focus. Do they focus on the most obviously winnable constituency seats? These offer the clearest potential for immediate Plaid Cymru gains. However, winning these seats – two of which are currently held by the Conservatives – would have little impact on Labour’s overall dominance of the Assembly. Furthermore, because both Llanelli and Carmarthen West & South Pembrokeshire are in the Mid and West Wales region, success in gaining both these constituencies might well mean losing Plaid’s Mid and West Wales list seat, thus producing a net gain of only one seat in that region.

Where is comes next on the list of potential Plaid targets? The next two constituency seats requiring the smallest swings for Plaid gains are Caerphilly, where Plaid would need a 9.7% swing on the 2011 result to win, and Clwyd West, which would need a 10.2% swing for Plaid to come from third place to win. These are the only other two seats where Plaid can win with swings around 10%. Neither of these two seats would exactly be easy wins for Plaid Cymru (to put it mildly). And only one of these two seats is in south Wales (Caerphilly is in South Wales East).

In a very good year for Plaid Cymru (with their national support level at around that won in 1999) and a bad year for Labour (with their support falling to the sort of level won in 2007), it is possible to see a pathway to Plaid winning 18 seats in the National Assembly. (That is not, please note, my prediction for how well Plaid will do in 2016.) Achieving this would require Plaid to capture all its obvious target constituency seats (up to and including Caerphilly), and the mathematics on the list seats working in ways that are as favourable to them as seems even vaguely plausible. On such a scenario, though the parties would be pretty close in terms of vote share, Labour would still be some way ahead of Plaid in the Assembly, at around 24-25 seats.

Moving any further forward than this would require Plaid to be achieving some truly remarkable swings – in some places even to approach the sort of swings they managed in several seats in 1999:

Clwyd South would require a 12.0% swing for Plaid Cymru to capture it;

Neath, a 13.5% swing

Preseli Pembrokeshire, a 13.5% swing

Cardiff West, a 13.6% swing

Wrexham, a 15.5% swing

Swansea West, a 15.8% swing

Rhondda, a 16.9% swing

Torfaen, a 17.1% swing

Cynon Valley, a 17.5% swing

Islwyn, a 18.2% swing


In pondering the task ahead of them, Plaid Cymru strategists might be wise to reflect on one aspect of the experience of the Liberal Democrats in 2010. Prompted by the eruption of ‘Clegg-mania’ after the first leaders’ debate into thinking that they might make considerable gains across Britain, the Liberal Democrats diverted precious resources from their original limited list of target seats into attacking across a broader front. They ended up, on a somewhat increased vote share, actually making a net loss of seats overall (including losing Montgomeryshire in Wales).

For Plaid Cymru in 2016 to put resources into targeting seats that are crucial to achieving their long-term objective would risk sacrificing more obviously winnable seats. Yet prioritising the seats where Plaid clearly could win in 2016 would mean, in practice, accepting that they will not seriously challenge Labour’s status as the leading party in the Assembly until some point in the future.

It is clear where Plaid Cymru wish to get to. Their dilemma is that the pathway for them actually getting there is much less clear.

On Uniform National Swing


One welcome development in Welsh political life and electoral analysis over the last year has been the growth in regular political polling. We now have more regularly reported measures of party preferences and public attitudes. In discussing the results of those polls, one topic that has cropped up quite frequently, both in Comments on blog posts here and also away from the site, has been that of how polling numbers (which report intended vote shares) are translated into possible election outcomes in terms of seats in parliament or the National Assembly. This is a topic that is easily open to some confusion, and one I’d like to discuss here.

 A first observation is that current polls should never be seen as a prediction of the next general or Assembly election. For one thing, as Sir Robert Worcester of MORI has often said, “Polls don’t predict; although pollsters sometimes do”. A poll asking about voting intention is a measure of party support now; it is absolutely not a prediction of what it might be at some time in the future. Of course it is true, as Nate Silver has observed, that the closer to the election you get the more confidently you ought be able to predict the final election vote shares from current polls. And polls conducted immediately pre-election should be able to get pretty close to the final outcome: if they don’t, something is probably awry with a pollster’s methods.

What I’d like to spend most of this post looking at, though, is how vote shares from polls are generally translated into potential outcomes in terms of seats: ‘what would happen if these findings were repeated across Wales in an election?’ These seat totals are also not predictions of what will happen, but rather projections of the current position as revealed to us by the polls. But how are such seat numbers generated? And how seriously should we take those numbers?

The method I use for all such projections reported here is that of Uniform National Swing (UNS). This method is also used by many others, such as the UK Polling Report site, and by the BBC. An obvious virtue of UNS is simplicity. To apply it you just compare the percentage support for each party in a given poll with the percentage support they received in the most recent relevant election. The percentage change (or ‘swing’) in support from the last election, whether positive or negative, is then applied uniformly to every constituency (and electoral region for National Assembly elections) in Wales. Repeat the process for every party, and see which party comes out on top in each constituency. Once this is completed, you have a full set of projected results for the whole of Wales. It really is as simple as that.

To illustrate, let’s use the recent BBC/ICM poll. This put Plaid Cymru support for the constituency vote in National Assembly elections at 24% (an unusually high figure in recent years); this compares with the 19.3% that Plaid won on the constituency vote 2011. Applying UNS from the poll to a projected Assembly election therefore means simply working through all forty constituencies and adjusting the Plaid vote upwards by 4.7% (i.e. 24-19.3). The Conservatives were at 19% in the same poll. This compares with the 25.0% that they scored in 2011. So for the Tories, UNS means working through all forty constituencies and adjusting the 2011 result downwards by 6% (i.e. 25.0-19).

For working out a projected general election outcome from a poll, all that needs to be done is to work through all forty constituencies for each party. For the Assembly it is just a little more complicated. I first work out the projected constituency results; I then apply UNS to the regional list vote for each party in each region, with the calculations allocating the list seats taking into account which parties are projected to have won the constituency seats in that region.

So that’s the method. (Try it some time at home; hours of fun for all the family). How good is it? Well, it is clearly not flawless. The most obvious and immediate flaw is that it occasionally projects impossible outcomes. The BBC/ICM poll showed the Liberal Democrats’ constituency vote on 5%. This compares with the 10.6% they won in 2011; UNS would therefore suggest applying a reduction of 5.6% to the Liberal Democrats vote share in all forty constituencies. OK – but what about somewhere like Llanelli, where the Liberal Democrats won only 2.1% in 2011?! Times are tough for the Lib-Dems at the moment, but I’m confident that their vote share hasn’t yet dipped below zero anywhere…

The other major limitation on UNS is that it does not allow for factors that are likely to produce local variations from national swings. We know that such variations exist, and indeed have generally been increasing in size: across the UK, the standard deviation from the national swing has risen. These deviations can be regional, such as we saw in the last Assembly election, where the swing to Labour was notably bigger in the three south Wales regions that in either Mid & West Wales or in North Wales. But deviations can also be particular to a single constituency, due to some specific local issue, to splits in a local party, or simply due to an individual local candidate who is unusually effective (or unusually poor). A common source of at least modest constituency-specific deviations is incumbency: i.e. the sitting member normally does accrue some sort of personal vote. Where a party’s candidate is standing for re-election for the first time there is normally a modest incumbency bonus; conversely, where an incumbent representative stands down and a party is defending a seat with a new candidate, they will typically experience worse than average swings.

UNS doesn’t account for any of these potential sources of variation. Why, then, do people use it? First and most obviously, analysts need some form of simple and neutral formula for projecting from polling numbers to an election outcome. We know that UNS is not perfect, but it is less flawed than any alternatives (for example, see the discussion here of ‘proportionate swing’).

Second, in the aggregate, UNS is normally pretty good in terms of projecting election results from the vote shares won by each party. Thus, in the 2011 National Assembly election there were seven constituency seats where the result differed from that which would have been predicted by UNS changes from 2007-11. Seven out of forty is quite a high proportion. But these local idiosyncracies largely cancelled each other out. Overall, the net differences between the final result and that predicted by UNS were small: the Conservatives won two more constituency seats than UNS would have projected and Labour one; Plaid lost two more than suggested by UNS and the LibDems one.

In my view, UNS provides us with a broad guide – a baseline gauge against which both the overall performance of the parties, and the results of individual seats, can be assessed. But it doesn’t provide us with anything more than that. And projections, using UNS, of polls conducted now are most definitely not any sort of infallible prediction of exactly what will happen at some point in the future. UNS is a perfectly reasonable tool, provided that we understand the limits of its usefulness.

While I Was Away…


As you may have noticed, I have been away from Elections in Wales duty for a bit. While I was off re-charging my psephological batteries, Lord Ashcroft released another set of polls in key Labour-Conservative marginal seats. (The full results of all the polling are here.) As with his previous round of such polling (which was conducted in April and published in May), this round (which was conducted in June, and published last week) included one Welsh seat, Cardiff North.

Cardiff North is a very important seat at the next general election. As well as enjoying the singular honour of being my own constituency, it is one of the most marginal seats in the whole of the UK – the Tories having captured it by a mere 194 votes in 2010. If they were to have any serious hopes of advancing from their current position in the House of Commons towards an overall majority, the Conservatives really need to hold Cardiff North. On the other hand, if Labour is to stand any chance of winning an overall majority, then they certainly need to gain the seat.

The prospects for the parties, however, are somewhat complicated by the factor of incumbency. We know from copious research that popular sitting MPs can outperform the typical swings experienced by their parties. That was probably one reason why Cardiff North was so close in 2010. On the average swings seen across Britain, and Wales, the Conservatives should have gained the seat fairly comfortably. However, Labour’s popular incumbent MP, Julie Morgan, kept the swing from Labour to the Conservatives down to a mere 1.5%. With Julie Morgan not standing next year – she is now the AM for the constituency, having won it decisively in 2011 – that ought to hand a significant advantage to the Conservatives. However, the Tories’ own position has potentially been weakened by the fact that the victor in 2010, Jonathan Evans, has already announced his decision to stand down at the next general election. So while Labour will have lost any incumbency advantage they had in 2010, the Conservatives will not gain from incumbency as they might have expected.

So how are things shaping up for the parties? As with his previous polls, Lord Ashcroft asked two main voting intention questions. The first was the standard “If there was a general election tomorrow, which party would you vote for?” The results for this question (with changes on April’s poll in brackets) were:

Labour: 38% (-3)

Conservatives: 31% (-3)

UKIP: 14% (+6)

Plaid Cymru: 5% (-2)

Liberal Democrats: 5% (-3)

Thus, the modest Labour lead has remained exactly where it was on this question, even as both leading parties have lost some ground to UKIP.

However, Lord Ashcroft also asked a second, following up question on voting intention: “Thinking specifically about your own PARLIAMENTARY constituency at the next General Election and the candidates who are likely to stand FOR ELECTION TO WESTMINSTER there, which party’s candidate do you think you will vote for in your own constituency?”. (Emphasis in original. Note that although the question refers to the ‘candidates who are likely to stand’, it did not actually name them, as there was not a full list of candidates available for all parties in all the seats that were polled). Results for this question (with changes from April again in brackets) were:

Labour: 41% (+1)

Conservatives: 30% (-3)

UKIP: 12% (+4)

Plaid Cymru: 7% (no change)

Liberal Democrats: 6% (-4)

The results here are clearly somewhat more encouraging for Labour: they have actually slightly increased their lead over the Conservatives since April.

The poll contained a couple of other interesting questions. One asked respondents whether they recalled having been contacted by the parties ‘over the last few weeks’. Here, the Conservatives were slightly ahead of Labour in Cardiff North (with 31% of respondents recalling having been contacted by them, compared to 27% for Labour); however, this represents a halving of the Tory advantage on this measure from the previous poll.

A second interesting question asked people if there were any parties that they would definitely not vote for at the general election. Here, the Liberal Democrats and UKIP tied in first (or rather last) place, with 65% of respondents each naming them; Plaid Cymru scored 58%, the Conservatives 48%, while Labour did best with only 39% choosing them. Particularly damning for the Lib-Dems is that more than half of the Cardiff North sample who indicated that they voted Lib-Dem in 2010 now said that they definitely would not vote for them in 2015.

As I mentioned in my discussion of the previous Cardiff North poll, we should exercise some caution in interpreting these polls. Individual constituency polls have a distinctly mixed record; moreover, they can only gauge voting intentions now, and can’t tell us what might change over the next nine months. But at the moment Labour are in the lead in Cardiff North. Although their advantage is nowhere near sufficient for a Labour victory here to be a certainty, Mari Williams must now be the clear favourite to gain the seat.

Who do we Trust?


Some time ago, I began a short series of Blog posts (see here and here) under the general theme of ‘hitherto neglected aspects of public attitudes to devolution’. These drew on detailed survey evidence gathered by the Wales Governance Centre here at Cardiff University, particularly the 2011 Welsh Referendum Study and 2011 Welsh Election Study. Although these studies are now slightly dated, they provided by far the most detailed information yet gathered in any surveys about various aspects of public attitudes towards many aspects of devolution and government here in Wales.

I had intended there to be three posts in the original series. I published two of them, and in the latter briefly previewed the third, but then got side-tracked by the fever pitch of political excitement that was the European Parliament election. Apologies for that. Anyway, making a somewhat belated appearance here is the third piece in the series. It concerns Trust. To what extent do the people of Wales have trust and confidence in their elected representatives, and those who govern over them?

The 2011 studies covered these issues in two main, slightly different ways. First, one question, which was included in the post-referendum wave of the Referendum Study asked the following:

‘How much do you trust the following to work in Wales’ best interests?’.

The table below summarises the (%) responses obtained for two levels of government – the UK and the Welsh governments – and also for two categories of representatives – MPs at Westminster and AMs at Cardiff Bay.


UK Government

Welsh Government



Just about always





Most of the time










Only some of the time





Almost never





Don’t Know





What is immediately striking about the table is the disparity in trust of government and political representatives in London and in Cardiff. Those in London are trusted at least most of the time by roughly one-quarter of all respondents, whereas around two-thirds are willing to give that level of trust to those in Cardiff. This is a very large difference.

The initial inclination of social scientists looking at apparently interesting survey findings like this is nearly always to find ways in which we can ‘explain away’ the differences. There are several potential such reasons here. First, and perhaps most obviously, is our old friend question wording. It may well be that the wording of the question (‘to work in Wales’ best interests’) tends to lead people to offer responses more favourable to those within manifestly Welsh institutions. In addition we should consider partisan politics: at the time this survey was implemented, the UK government and the majority of members in the UK parliament represented political parties with only minority support here in Wales. Third, we might also bear in mind that in March 2011, when this survey question was asked, many memories lingered of the 2009 Westminster expenses scandal, which would hardly have helped elevate reported trust in UK-level politicians. Nonetheless, even with all these caveats entered, the difference in reported trust between UK and Welsh political institutions and those within them is stark.

Partly to compensate for any potential problems related to one particular question format and wording, elsewhere in the Referendum and Election studies another types of question was asked about trust. Here, respondents were asked to rate different institutions and those within them on a 0-10 scale, “where 0 means no trust, and 10 means a great deal of trust”. Four separate questions were asked, concerning levels of trust in people within those institutions to ‘Tell the truth’, to ‘Do what is right’, to ‘Be concerned with the problems of people like you’, and ‘to conduct their work with honesty and integrity’.

 Taken together with the question above about ‘Wales’ best interests’, these different questions potentially tap into several different dimensions of political trust: a concern with Wales, a connection between representatives and represented, and personal probity. We should not, therefore, necessarily expect that answers will be wholly consistent across the different questions: one could quite imagine people believing some politicians to be personally honest yet utterly out of touch, for instance.

To help place answers about politicians and governments into some sort of broader context, some of these questions were also asked about other institutions like The Courts and The Police. The table below shows the mean average ratings (out of a maximum of 10) obtained for the four questions:



Tell Truth#

Do Right#

Concerned Problems*



UK Government





Welsh Government





Westminster MPs





Assembly Members





Your local council



The European Union



The Courts



The Police



# Source: 2011 Welsh Referendum Study (post-referendum wave); * Source: 2011 Welsh Election Study (post-election wave).


It will surprise no-one, I suspect, that politicians and political institutions generally scored lower in terms of trust than those involved in the justice system. Indeed, if anything I might have expected the gaps to have been even greater. It will also surprise no-one that the European Union attracts low levels of trust.

What again stands out perhaps most from the findings on these questions, however, is the disparity in responses regarding MPs and the UK government on the one hand, and AMs and the Welsh government on the other. Although the differences on this types of question format perhaps look a little less stark than in the ‘Wales best interests’ question discussed above, they remain substantial. However the question is asked, it seems, those at the devolved level attract much greater trust than those at the UK level.

Why might this be? There is some general tendency for people to prefer political representatives who are closer to them; hence, surveys across Britain generally find greater levels of trust in local councils and councillors than the national government and MPs. Yet here we find greater trust in the devolved level even than in local councils. It may be that the proximity of the surveys to the 2011 Assembly election helped raise the reported standing of the devolved institution and its members somewhat; even so, the differences between levels of trust in the devolved level and the UK level is both so consistent and so substantial that it is very difficult to believe that conducting the survey at another time would have made very much difference to anything.

Greatest Hits, etc


It will soon be exactly one year since Elections in Wales was launched on an unsuspecting and defenceless world.

Since then I have been pleasantly surprised, even amazed, at the level of interest that the blog has attracted. We’ve set monthly records for readership and page views in eight of the last eleven months, and have passed 30,000 total page views.

Diolch o galon, i chi i gyd, am eich cefnogaeth a ddiddordeb. Thanks very much to all of you for your continuing interest and support. It might be, to misquote one of my favourite films, that the electoral politics of one little nation don’t amount to a hill of beans. But, as Leslie Nielson went on to say, this is our hill, and these are our beans…

I will actually be away for a few days when we celebrate our first birthday. (Though I expect to receive reports of the nation united in rejoicing, large and emotional crowds throughout Wales etc etc). However, I thought it might be of interest to some of you – and particularly those of you who have joined us over the last year – to share with you some of the most popular posts over the last twelve months. So, in no particular order…

Slightly to my amazement, one of the most popular posts I did in the first year concerned a survey question wording experiment. Trying to Get it Right reported findings of a mini-study that I ran with YouGov, where we were seeking to puzzle out why some recent YouGov polls in Wales had been showing surprising results for the regional list vote in National Assembly elections. The results we got back amazed us, and many of you clearly also found it interesting.

A second very popular post involved me playing my regular role of Mr Spoil Sport. (This is a role I started fulfilling rather early in life. I still remember when, as a young boy, I found out that Father Christmas wasn’t real. I eagerly shared this news with all my friends at school that day, who I thought would want to know this – only to make some of them cry, receive a bollocking from our teacher, and find that I ended the day with rather fewer friends than at the start). In this particular instance, I suggested that question-wording effects meant that the apparently striking findings of a constitutional preference question in the annual BBC/ICM poll were, in reality, maybe less interesting than they initially seemed.

Three other very popular posts (Part One is here; Part Two here; and Part Three here) looked at the history of one-party dominance in Wales. Sustained period of dominance by a single party have been a consistent feature of political life in Wales throughout the democratic era – and, I have argued at various times, a persistent pathology in Welsh political life. These pieces outlined the story from the 19th century until more-or-less the present day.

Another rather popular post was one analysing the electoral system used to choose the majority of Welsh councillors. This system – a multi-member version of First Past the Post – is, to my mind, a strong contender for the title of The worst electoral system in the world, managing to retain all of the weaknesses of First Past the Post without any of its redeeming features. Sadly, there currently seems little eagerness to get rid of it within the ranks of Wales’ still-dominant political party.

Finally, I as a taster before the night of the European election results, I ran a short piece seeking to explain the mathematical formula under which the seats in Wales – and the rest of Britain – would be allocated. The Fabulous Mr D’Hondt was a 19th century Belgian mathematician, and evidently one with a sense of humour…

 Happy reading. I’ll be back soon, with my psephological batteries fully restored, I hope.

Public Attitudes to the Political Parties


Elections in Wales has now been running for very nearly a year. During that time its been very pleasing to see the audience for the blog growing considerably. The many readers who haven’t been with us right from the beginning may not, therefore, be familiar with one survey question that was run on the YouGov poll conducted in July 2013 to mark the launch of the blog.

This is a question that was originally developed in the context of multi-party continental European political systems, and seeks to measure public attitudes to the parties in a rather different – and arguably more subtle – way from questions on current voting intention, or that long-standing political science favourite, party identification. The question follows this format:

“We have a number of parties in Wales, each of which would like to get your vote. Using a scale that runs from 0 to 10, where 0 means very unlikely and 10 means very likely, how likely is it that you would ever vote for…”.

This question can then be applied to all potentially relevant parties.

The question recognises that many voters do not have a simple and absolute attachment to one party and aversion to all the others, but varying degrees of attraction towards the options before them. It is thus particularly useful in multi-party systems, such as we have had for some time in Wales. And because the question has now been asked in several polls, we can compare attitudes over time.

This question was included in the latest Welsh Political Barometer poll, and was asked not only about the four parties represented in the National Assembly, but also about three others: UKIP, the Greens, and the BNP. So how did the parties do?

A first interesting thing to look at, I think, is the percentage of respondents who score a party 0 out of 10: in short, they really dislike this party, and cannot see themselves ever voting for them. The table below reports two sets of statistics for each party: the percentage scoring that party 0 out of 10; and the change in that percentage since July 2012, when this question was previously asked. (A positive score in the final column represents a bad result for a party: it indicates a rise in the percentage of people indicating they score their likelihood of ever voting for that party at 0 out of 10).



% 0 / 10

Change since July 20123







Liberal Democrats



Plaid Cymru













A few things stand out from this table. A first one is the BNP figure, which many may find reassuringly high. A second is the fact that attitudes seem to have become somewhat more negative for nearly all the parties; or perhaps we just got a rather grumpy sample! Other noteworthy things include:

  • It is striking that attitudes to UKIP seem to have become more clearly defined: their electoral support has risen in the last year, but more people also now seem to regard them as electorally beyond-the-pale.
  • As well as their electoral support levels falling, antagonism towards Labour seems to have increased. They are still the least disliked party, but their advantage over the others in that respect seems to have diminished.
  • It’s also notable that, in what was not an outstandingly good poll for them in terms of voting intentions, Plaid Cymru come close to matching Labour as the least disliked party. Relatively few people dislike Plaid; the party’s problem continues to be a failure to convince sufficient number of the rest of the electorate to positively support it.
  • While the Conservatives remain much more strongly disliked than either Labour or Plaid, they are the one party to have actually marginally improved in this respect over the last year.
  • The poll piles on yet more bad news for the Liberal Democrats. They attract much more hostility than they used to: when this question was asked immediately after the 2010 general election, only 17% gave 0 out of 10 for them. Now the figure is almost half of all the sample, and actually slightly higher than that for the Conservatives.

That’s the picture for hostility; what about the positive end of the spectrum? Those rating parties above the mid-point on the scale, in the 6-10 range, might be said to be broadly positive towards a party. So what percentages of our sample rated each of the parties in this range? The next table shows these figures, and again also gives the changes since the July 2013 poll. (For this table, a positive number in the change column is therefore a good thing for a party, indicating a rise in the proportion of people scoring them highly on likelihood to vote for that party).


% 6-10 / 10

Change since July 20123







Liberal Democrats



Plaid Cymru













Labour is some way ahead of the field: as well as attracting less hostility than the other parties, it also attracts notably more positive support. That’s not a bad position to be in! What may be a bit concerning for Labour, however, is that their position has declined so much since last year. As with voting intention, Labour are still in the strongest position, but they are no longer completely out of sight of the other parties.

Plaid Cymru are in a clear second place on this measure. Yet they will surely be disappointed that they have not improved on this measure over the last twelve months and have actually moved backwards slightly. Although in third place on this measure, the Conservatives may take more heart here. It is also noticeable, looking at the details of the figures, that much of the Tory support is very strong: 11 out of the 27% of respondents scoring them at 6-10 actually choose the 10 out of 10 option; for Plaid Cymru the equivalent figure is only 7 out of 31%.

The news continues to be unremittingly bleak for the Liberal Democrats. Not only has hostility towards them risen, but their potential pool of voters seems to be shrinking. The contrast with May 2010 is again striking: then, fully 42% scored them in the 6-10 range. On this measure, the Lib-Dems are now well behind not only the other parties currently represented in the Assembly, but also UKIP and even the Greens. For UKIP, what is notable is that their pool of potential support has not expanded; what has happened over the past year is that UKIP has started to convert much more of that potential support into votes.

Overall, these figures provide us, I think, with a useful supplement to those from polling questions on voting intention. Though in most cases they tell us a similar story, these questions add some interesting nuance – notably, in the case of this poll, for both UKIP and Plaid Cymru. So I hope we’ll be able to repeat this question in some future Welsh Political Barometer polls.