Skip to main content
Coronavirus (COVID-19): Latest updates.


Scottish Independence and Wales

27 March 2017

Whatever else we might think about politics at the moment, it isn’t dull. Just in the last few days, we have seen two (related) announcements with potentially momentous implications for the future of the UK and the people of these islands: the news on the imminent triggering by the UK Government of Article 50, and Nicola Sturgeon’s statement on a possible second Scottish independence referendum.

Among the questions that have been asked of me in recent days have been a number related to the Scottish First Minister’s statement, and its relevance for Wales. I’ll try, in this blog post, to address some of the main issues that people have raised.

First, a number of people have asked me about support for independence in Wales. As I have discussed at various times on the blog in the past, the general picture has consistently been one of very limited support. The precise level of support reported can vary from poll to poll, but it also varies according to the type of question that is asked. Thus, when people ask me (as a number have done recently) “how much support is there for independence in Wales”, I’m afraid I have to resort sometimes to the standard academic response of suggesting that the question is too simple.

If we give people several constitutional options for Wales to choose from then we generally see support for independence at or below ten percent. In the most recent such poll, that conducted by ICM for BBC Wales, then six percent of respondents favoured the independence option – an entirely typical level of support. If, however, we ask about independence directly, in some form of Yes/No choice, then the proportion of respondents who choose independence is rather higher. But it is still very much a minority taste – in the most recent Welsh poll of which I am aware (see here and here) in which such a question was asked, even on this straight binary choice, support for Welsh independence was still well below twenty percent.

While it would be interesting to see if opinion has shifted on this latter form of question, or on the issue of Welsh independence more generally, there is as yet no solid evidence of it having done so since the Brexit referendum.

Second, another matter that several people have asked me about is support in Wales for Scottish independence. This is not a question that has been asked frequently, and not – as far as I am aware – recently. The most recent data that I am aware of dates back to just after the Scottish independence referendum. This showed very limited support for Scottish independence in Wales.

Third, I have also been asked about whether support for Welsh independence would change significantly in the event that Scotland voted for independence in a referendum. I think the most important thing to say here is that asking about such matters in any survey is difficult: you are asking respondents, most of whom are not political obsessives, to make a huge leap of the imagination into a hypothetical future scenario. I’m not at all confident that we can take too seriously the responses that we get to any such question. However, where polls have asked about this in Wales (e.g. here), the response has been that it doesn’t seem to change attitudes in Wales very much at all. Most people in Wales do not want Scotland to leave the UK. But if Scotland does do so, that doesn’t appear to change many people’s minds about whether Wales should follow a similar path. This may mean simply that looking forward to a scenario in which the UK as we know it no longer exists is too big an ‘ask’ for most survey respondents. Or it could reflect something deeper – that for most Welsh people the union that really matters to them is the ‘M4 union’ between Wales and England.

Wales is in a very different place from Scotland in terms of the constitutional debate. This is emphatically not to say that things in Wales are either static or dull. Indeed, in many ways Wales has been through the most interesting devolution journey since the 1990s of any of the three devolved nations. There have been a series of models of Welsh devolution – though some might view them as a set of bodged jobs – and almost the only constant has been change. Devolved Wales was definitely not, as Lady Gaga might put it, born this way. In less than two decades since the Assembly was created, Wales has acquired substantial primary law-making powers, is now developing significant taxation and borrowing powers, and in the latest iteration has moved from a conferred to a reserved powers model of devolution. This is not small stuff.

Just as Gigi Hadid is, so I am informed, International Model of the Year, this year’s model of Welsh devolution is the one incorporated in the 2017 Wales Act. Very few outside observers expect the latter to be remotely enduring or sustainable – indeed, if it outlasts the career of the estimable Ms Hadid it may exceed many expectations. Yet amidst all this flux, one thing that has been stable for pretty much a decade and a half has been public attitudes. All polls and surveys throughout that period have shown limited support for independence; they have also shown only modestly higher support for abolition of the Assembly. The normal picture, evident in just about all polls since early in the century, has been a clear majority in favour of some autonomy within the UK. Within that, most polls show a rather greater number favouring ‘more powers’ than support either the status quo or more limited devolution. And, curiously, support for ‘more powers’ has largely remained stable even as the powers of the Assembly have been changed and, generally, increased substantially. Whatever powers the Assembly actually has, plenty of people in Wales seem to support more.

This raises questions about how much people understand what powers actually are devolved in Wales. But that is another whole set of questions – which I have discussed at various points on the blog, and will do so again, doubtless, in the future.


  1. Graham Hathaway

    I have recognised fully the landscape of politics verses independence and the role of our assembly outlined in the blog. It’s certainly the case.
    It’s what drives the cause of Indy. Has the case properly been made and has Westminster acted houourably in creating parity with other devolved parts of Wales. I think not.
    The latest Wales Bill has even clawed back powers originally devolved. The centre has been less than serious about Wales and this has been hard fought territory. It gives no confidence in often contested areas of devolved areas with challenges and prevarication from W/M.
    The waters are muddied and tensions are lurking on the surface of devolved responsibilities.
    In short there is no warmth of ideas and support for Wales on its journey to greater powers.
    It explains confusion with the public on devo, and uncertainties.
    The question of laws and law making is as vexed now as hitherto.
    Until the question of a separate jurisdiction for Wales is bestowed we shall always be yoked and manacled. The fallout from Brexit will leave Wales very vulnerable as a consequence.
    But by coincidence it may shake the chains. As you say Roger, these are not dull moments in our turbulent diet of daily politics. And with this possible change.

  2. welshnot

    This is a good analysis of the failings of your own survey, seeings as you advised the BBC on the creation of the St David’s Day study.

    As you have identified, the survey provides a false impression of support for Welsh indpendenceby by mixing options for greater powers with a binary independence option.

    It creates a misleading figure, currently 6 percent, which is used constantly throughout the media as proof of shockingly low levels of support.

    As you acknowledge, the actual level is liable to be somewhere around the 20 percent mark.

    If I was an expert on polling, I would feel uncomfortable about the continued reliance on such a flawed survey (even if I did help create it).

  3. Graham BC

    one interesting extrapolation of the responses are, a wonder when the support for more powers might stop. Could it be a case of more a fear factor when it comes to independence and the more we devolve the less fear there is, so the support level might actually rise once we get a powerful devolved assembly

  4. Graham Hathaway

    You have again demonstrated eloquence as much as foresight. It’s the very crucible of identity and self belief. The power to make changes, to influence, and address perceived Welsh issues in a way that local circumstances dictate.
    You have a hand on the steering wheel of progress and authority.
    Those who preside over such matters have a status of game changers and innovators.
    This isn’t party political, it’s the shine of true democracy where opinions matter and where voters can identify with their leaders who can effect change. And be locally accountable.
    This is well accepted at Westminster. The two party dominance is a practiced art of power retention in a centralised fixated and self empowering feast. It’s demoralising and imperfect. The Brexit outcomes may compound these morals affecting Wales to dismiss us further with little enactments and adoption of EU laws to our benefit or a revolution of ideas of greater interdependence or federalism. The way it looks now I fear suffocation.

  5. J.Jones

    To come back to a favourite subject; I notice that Panelbase, when surveying Scotland for opinion on independence, weight their sample according to place of birth; Scotland or England.

    You will be familiar by now with my constant “whinge” that something is wrong with survey samples in Wales that consistently understate the percentage of adults who do not speak Welsh. The St David’s day poll had 62% able to speak Welsh this year, 64% last year and “ability to speak Welsh” is a very potent predictor of opinion on a range of matters…the language itself of course but also support for devolution/independence.

    ICM, who I had a dialogue with on this subject, declined to weight to Language ability on the grounds that it was too imprecise; it varies between the Census (19%) and the all Wales survey (24%).

    Nevertheless the percentage of adults born in Wales (69%) and the percentage born in England (23%) is well known and secure.

    In the future it would seem sensible to weight for country of birth when asking the “I” question. I suspect that, since only 9% of English born and 21% of Welsh born speak Welsh, then weighting for country of birth might go some way to sorting the Welsh speaking ability exaggeration as well.

  6. Graham Hathaway

    Lucy’s reference does add grist, even some clarity on the more philosophical issues of language and wider relationships of media influence. Wales faces unprecedented erosion of its core elements of those things most take for granted. To have influence over the way we wish to lead our lives
    There are existential threats posed to our Nation that have accumulated without challenge.
    It has generated an apathy of association as a passenger rather than a driver of change and progress. It needs to be talked about as the article suggests, and with haste. It’s never too late.

  7. J.Jones

    The recent YouGov Scottish independence poll also weighted to country of origin; Scotland, rest of UK, elsewhere.

    Why is it so hard to get proper weighting for YouGov Welsh barometer polls?

  8. J.Jones

    Here’s an interesting article by Prof John Curtice:-

    It’s interesting to me because it shows how polling can get it wrong when it comes to social attitude surveys.
    For most political surveys, methods used by Roger/YouGov are adequate and accurate because they seek out and find the “politically engaged”; those people who listen to the news, have a political opinion and are very likely to vote. They give the correct answer to the question “Which party will have the greatest number of politicians after the next election?” Or the relative standings of each party.
    They do this because they find the people who will swing elections.
    My position has always been that some questions, and particularly “social attitude” questions, need to find people who do not volunteer to answer polls.
    John Curtice shows how important this is when the big questions are asked in referenda.
    The Curtice poll asked those people with no interest in politics if they voted in the GE; few of them had. He then asked if they voted in the EU referendum…a high percentage had. Here’s the rub; a majority of those who voted in the EU referendum but were uninterested in politics, voted “Leave”. Amongst those interested in politics the majority voted “remain”.
    In this way we see just how wrong Roger is to trust YouGov’s inadequately weighted polls for the WES and for some barometer questions. ICM is similarly inadequate for its questions in the St David’s day polls.

    Play “What if?”

    What if those people who form more than half of the voting population (those who don’t vote in Assembly elections) suddenly voted on Welsh specific issues. What would the outcome be?
    What does the WHOLE population of Wales really think?
    Until psephologists like Roger remove the blinkers (or set aside their political affiliations) and look seriously at how and WHO they poll, we will never know.

Leave a reply

Your email address will not be published. Required fields are marked *