This Sunday’s polling is mostly dominated by Scotland – even the YouGov/Sunday Times national poll mostly had questions about Britain’s attitudes to the Scottish referendum.

Let start with the Scottish polls though. Last weekend we had a Survation poll in the Mail on Sunday which was widely reported in the media as showing that George Osborne’s intervention in the referendum debate had actually boosted YES. This was mostly rubbish – the change appeared to be largely, but not wholly, the result of Survation changing their weightings. I concluded we should probably wait for more evidence before deciding what the impact from the currency row was.

Today we have a new ICM poll for Scotland on Sunday. Their topline figures with changes from a month ago are YES 37%(nc), NO 49%(+4). At first glance this poll would indicate the currency row had led to a significant boost to the NO campaign, but once again I’d urge some caution. Regular readers will remember that the previous ICM poll showed a big swing towards YES, far bigger than any other poll, so this one may very well just be a reversion to the mean rather than any meaningful change (in particular ICM’s last poll had an unusually pro-independence sample of young people, which I suspect may have vanished. On that subject this month ICM have apparently changed their method very slightly, changing the age bands they use to weight young people.)

In the rest of the poll ICM found that 63% of people in Scotland think it is in Scotland’s best interests to keep the pound, 12% think she would be better off with a separate currency. 47% think that an independent Scotland would be able to use the pound, that the main British parties are bluffing. On the issue of Scotland’s EU membership, 54% would like to see an independent Scotland be an EU member, 29% would not; 57% think Scotland would be able to join, 24% think membership would be blocked.

The SNP have also commissioned a new Panelbase poll. Now, the last time we saw an Panelbase/SNP poll they played silly buggers with the question ordering, but I’ve double checked with Panelbase and nothing like this happened in this one (though the wording is very slightly different to that used by the Panelbase/Sunday Times poll). The topline figures are YES 37%, NO 47%. The no vote is two points lower than the last Panelbase/Sunday Times poll, but Panelbase’s previous poll was a bit higher than usual – for most of the past year Panelbase’s polls have consistently shown a NO lead of between 8 and 10 points, this is wholly in line with that.

In short, looking at the post-currency row questions we’ve got some polls showing YES up, some showing NO up, some showing little change, all of them obscured to some extent by reversion to the mean after unusual results or methodology/wording changes. It’s a pretty confused picture, but I’m struggling to see any clear movement to YES or NO.

Meanwhile the weekly YouGov/Sunday Times poll is here and has topline figures of CON 32%, LAB 39%, LDEM 8%, UKIP 12%. In England and Wales 21% support Scottish independence, 61% are opposed and English & Welsh respondents are now slightly more likely to think E&W would be worse off (27%) than better off (23%) if Scotland left.

Also worth noting there is an interesting non-Scotland related question – YouGov repeated a question from last April about the government’s welfare reform package as a whole, freezes, caps, bedroom tax, etc. Back in April 2013 56% of people said they supported them, 31% were opposed. Now 49% support them, 38% are opposed – so still more in support than against, but a significant movement over the last year.

397 Responses to “ICM and Panelbase Scottish polls”

1 2 3 4 5 8
  1. @Old Nat
    “Lloyd George was on holiday in Inverness….”

    I hope you counted the spoons.

    Thanks for that. It was mildly interesting and a blessed relief.

  2. Crossbat11

    I agree with you re the cross breaks – indeed, some time back. it was a motto on here “Don’t look at the Scottish cross break!”

    However, L Hamilton has a point when you look at the Holyrood polling, where the SNP still lead.

    I do find the “personalisation of issues” aspect that the UWS study commented on interesting, though.

    It’s notable the number of times that the referendum is referred to (especially furth of Scotland) as “Salmond’s”, as if it was a campaign by an individual that none of the rest of the c. 45% of Scots who seemingly share that ambition have any input to.

    I can understand the politics of doing that. Indeed the reverse is also true. While the No campaign isn’t Cameron’s, the implication that Johann Lamont is just a Tory stooge probably doesn’t help Labour in Scotland.

    On the whole, politics isn’t a nice business. :-)

  3. Postageincluded

    “I hope you counted the spoons.”

    While I’m not THAT old a Nat as to have been there, I don’t understand the reference.

    Perhaps contemporaries might have been more concerned about the fate of the “four and twenty virgins” who allegedly “came doon frae Inverness”.

  4. @CB11

    I also occasionally bring to peoples attention if Lab are doing something unusual. It’s the unusual poll that stimulates the debate away from the polldrums.

    Got that statistician’s name yet?

  5. According to my reckoning the Westminster VI for the Survation poll:

    gives a VI of

    Con 17%
    Lab 33%
    LD 6%
    SNP 38%
    Other 6%

    However they didn’t weight the poll according to how people voted in 2010, but according to how they voted for Holyrood. And this may have biased things in favour of SNP.

    There may be a more basic problem with the sample as well. When asked how likely they were to definitely vote for Westminster 65% said so. For Holyrood 67% gave an LTV of 10. But we know from all previous elections that turnout for Holyrood is around 50% but it is 60%+ for Westminster. So a sample that says it is more likely to vote for Holyrood is very possibly unrepresentative in some way.

    Incidentally the LTV=10 for the referendum was 74%, which suggests that the turnout will be high, maybe 70%+

  6. Roger Mexico

    Interesting post.

    “But we know from all previous elections that turnout for Holyrood is around 50% but it is 60%+ for Westminster. So a sample that says it is more likely to vote for Holyrood is very possibly unrepresentative in some way.”

    That’s true, but it might just also be the case that the very high proportion of Scots (in every poll) intending to vote in the referendum, might just have regenerated the idea of democracy in Scotland.

    I’m not convinced, but it might also be true.

  7. @ Roger M

    I referred to ‘panel packing’ in my earlier comment. I think Survation could be suffering from the effects of this; (IIRC Panelbase admitted at one time that it had concerns about its own panel being unrepresentative).

  8. Amber

    That was a concern raised by No campaigners at the time of the first Panelbase poll in the current series.

    Your comment about Panelbase being concerned about it, however, lacks any basis in what might be called ‘reality’.

    What they actually said was, “There has been some discussion on Twitter of an organised signup of Yes campaigners, hoping to influence Panelbase polls. Not all of the quoted posts actually relate to political polls, but the point is still one to consider. In fact new joiners have had
    no significant effect on our results but we do recognise the potential for abuse of the system if people on either side were able to coordinate a mass sign up of new members. As a precaution, between now and the referendum we will not be including results from panel members who joined more recently than June 2013. This only applies to political polls. We may still invite people to take part for our own internal analysis, but their data will not be used in the published results.We are also considering other measures to ensure that sample sources have no undue effect on results.”

    As they also said in that release, “we entirely understand that the Scottish referendum is an emotive issue”. I’m sure that you didn’t intend to allow your emotions on the issue to override reason, doubtless, you simply misremembered the actualite.

  9. Hence the IIRC If I Recall Correctly. And given that Panelbase felt they needed to ‘take precautions’ against it happening, I feel my comment was on point.

  10. Amber

    Except that you “didn’t remember correctly”. Panelbase found no evidence at all that new entrants had affected the results.

    That they recognised the potential for abuse was sensible. IIRC YouGov took similar action at some point.

    if Survation has been a victim of panel-packing by No supporters, then you could simply ask them if there was any evidence of that being the case.

  11. @Old Nat
    Sorry, you are quite correct. I meant “virgins” not “spoons”. Or perhaps “peerages”.

  12. PostageIncluded

    Must have been peerages that you meant. All those ex-Labour Scottish MPs forced to complain about their tiara-bedecked wives being unable to find a decent table in the Lords Dining Room.

    Scandalous, I call it! They sold everything for that privilege, and see how they are treated!

  13. I’m not getting into emotive arguments. Roger M makes a good case that Survation’s sample seems unrepresentative. Perhaps they failed to ‘take the precautions’ which Panelbase decided were necessary.

  14. Amber

    An unrepresentative sample is no evidence of panel-packing!

    It may well be (indeed it is demonstrably the case) that all pollsters at various times simply get things a little wrong.

    While I find conspiracy theories bizarrely interesting, it doesn;t mean that aliens landing in Stoneybridge are affecting our decent British pollsters!

  15. Crossbat et al

    The reference to the crossbreaks in YouGov was -as I have said – a mere debating point and no more.

    The fact that Survation had the NATS ahead for WESTMINSTER is a more substantial point. Now it may be that there are some questions with this poll. But they had the NATS at the same (HIGH) level for the Scots Parliament as did ICM a couple of weeks before. All polls are vulnerable in some respect and Survation are a respectable pollster.

    However most polls ICM, PANELBASE, SURVATION are now converging with the Yesers in the low to mid 40s. Six months ago they were -at best- in the mid 30s. I would say that YES is gaining ground.

    Most polls show the SNP still flying high after ages in Government – a truly extraordinary feat.

    Therefore my point is that if Salmond is heading for a fall according to you then it is from a range of high achievement.

  16. Before my fellow Scots start hyperventilating about the cabinet meeting in Scotland, remind yourselves of recent history here –,d.ZGU

    This has been happening since 2008, under two PMs, with one other Scottish meeting in Glasgow on 16th April 2009.

    This weeks meeting appears quite in line with recent political management of the cabinet.

  17. ALEC

    Cabinet meetings held throughout Scotland are quite normal. They are always followed by an open question and answer session. Portlethen will be no different from that pattern.

    Why would anyone “hyperventilate” about such a normal process?

  18. Interesting story coming out of the EU regarding the fears that part of any post independence EU negotiations could be that Scotland loses rights to 54 VAT exemptions currently enjoyed by the UK as long standing membership concessions.

    On the face of it, possibly a scare story, but probably not. Given that we know there needs to be some form of negotiation, and even if a seamless membership is agreed, even by the SNP’s admission treaty changes are required, meaning national vetos come into play, so there is bound to be an element of horse trading. However, the extent of the impact of this is very hard to pin down – some concessions are likely in some areas, but a carry forward of all UK concessions is highly unlikely.

    Some of the more outlandish scare stories are frankly ludicrous. There have been stories that Scotland won’t be able to watch Dr Who or join Euro Vision if they vote Yes. Such naff reporting is driving sentiment towards Yes.

    However, we know that the economy is the key salient subject, and we also know that while it probably hasn’t shifted votes directly, the currency debate has found a large majority of Scots disbelieving Salmond’s claims.

    Were another issue with direct salience – like VAT increases – to come along, and Salmond’s confidence once again be questioned, then maybe we move from another run of the mill scare story to a much more significant worry for Yes.

    Incidentally, while I’m not the kind of Scot who jumps at every bump in the night, thinking those nasty English are out to get me, there are times when I wish national public organisations could inject a bit more political correctness.

    BBC News 24 reported the English group draw for the Euro’s next year in the sports headlines, and told viewers we would have to wait another 20 minutes for the three other home nations.

    Editorially, there is an argument about numbers, with far more England fans than any of the other nations, but you might have thought an editor in Salford could have thought about adding twenty seconds of time to maintain a better sense of equality for a ‘British’ Broadcasting Corporation. These little things do get some people agitated, and they do have a point.

  19. Watching BBC News, I’m impressed that the Russians only showed 4 Olympic rings in the closing ceremony.

    That unexpected tribute to Team GB’s four medals would bring tears to a glass eye.

  20. “Your comment about Panelbase being concerned about it, however, lacks any basis in what might be called ‘reality’.”


  21. ALEC

    ” the currency debate has found a large majority of Scots disbelieving Salmond’s claims. ”

    I’m not sure that I am aware of the evidence for that. The Survation poll asked whether people thought that the Uk was “bluffing” or not. 37.0% thought that they were, while 37.3% thought that they weren’t.

    Incidentally “Salmond’s claims” is exactly the kind of personalisation that the UWS study identified in their analysis,

    It’s a tactic that should be used by campaigners, not those on UKPR trying to analyse events on a rational basis.

  22. Another problem with that Survation poll was the order in which they asked the questions.

    The question on Westminster VI came after several on the referendum, and on the different votes for the Scottish Parliament. The clue is in the number – Question 14.

  23. Phil Haines

    Of course, that would only matter if you think that Westminster VI is the most important matter and should always come first.

    Had that been the case in this poll, it might have influenced the answers to the other questions.

  24. The ICM poll also contained a “Wisdom” question.

    “The poll of 1,004 Scots showed that, on average, voters expected the result of the vote on 18 September to be 53 per cent against independence and 47 per cent in favour.”

    Now, there is very little evidence that people actually “have wisdom”!!

    However, people’s perceptions can have an effect on their ultimate behaviours, so it’s an interesting (if not particularly important) pointer to how Scots are thinking that the debate is going.

  25. I wonder what impact the sale of our NHS hospital records to insurers, along with sufficient indicators of identity, may have on the polls?

    Probably none, little does. However, it seems that the delayed proposal to collate GP records was intended to dove-tail with the NHS records, in spite of government ministers saying that such usage would be illegal.

    IMO this is a story that should alarm us all.

  26. @OldNat
    “Of course, that would only matter if you think that Westminster VI is the most important matter and should always come first.”

    The order of questions in the Survation poll matters if people are going to consider the results of the Survation polling on Westminster VI, Since you, Roger Mexico and Amber have been discussing it, it clearly should matter to you.

    If we’re not interested in the results of the Westminster VI poll from Survation, it clearly doesn’t matter where in the order of questions that question came.

  27. I should have made clear that the story refers to the English NHS.

  28. Phil Haines

    Old age is clearly creeping in, since I don’t remember discussing the Westminster VI.

    Not that it particularly matters. Voting never happens in a vacuum. Lots of things are in voters minds as they stroll into the polling booth,

    Since a Westminster election will happen months after the referendum vote and so, it seems likely, will be affected by that result, then asking about referendum questions first has the advantage of chronological logic behind it.

    Unless polls only ask one question at a time, then there might be an inevitable tendency for the first question to affect the answers to the subsequent ones.

    That you haven’t previously seen a problem with polls asking multiple questions (AFAIK), that you see it as a problem on this occasion seems a little bizarre.

  29. @ Oldnat
    Obviously I stopped reading yr posts when you were unmasked as the No Campaign’s secret weapon but I did like the phrase — new to me — “would bring tears to a glass eye”.

    It’s probably a household saying in Scotland?

  30. Syzygy

    “I should have made clear that the story refers to the English NHS.”

    While the clarification is welcome, it is sadly the case that most of us outwit the “care” of the NHS in England tend to take it for granted that references to “the” NHS do only appertain to the system in England.

    As Alec noted above about the announcements of the football groups, that can be a minor annoyance. Alas, even within a post No BBC service that included a “Scottish Six” or similar, Alec would continue to be disappointed, unless he chose to tune into the broadcasts from his roots. :-)

  31. RobbieAlive

    I have no idea of the origins of the phrase. It’s one I’ve always known. If it is a Scotticism, then fine. It’s a very descriptive one.

  32. I have looked at the “wisdom” article, and I don’t rate the writer’s arithmetic skills. Getting 47% is said to be losing by 3 percent.

  33. OLD NAT
    “North of the border” So that would be Iceland or Norway?

    Good point. No, the existing border seems to me to be perfect;y valid and satisfactory, tho’ it used to be more apparent when we had to wind our way over Scap Fell and trundle up northwards after a trucker’s tea and bacon sandwich at Scotch Corner.
    Once you have had your No vote, all you need to do is blow up the M1 and the M6 to renew the element of impoverishment and misery essential to the independent psyche of the Scots. Leave us the airport and the East Line, though.

  34. @SHEVII

    “Re Atos

    I think people are missing the point here. It doesn’t really matter who introduced it- it depends on how it is used and whether it is fair.”


    Well, it depends what the goal is. If the focus is on the needs of those enduring the Atos process, then maybe who introduced it is secondary to the issue of having a good process and outcome.

    But given we are concerned also here with VI, who introduced it (and who is perceived to have done so etc.) is rather germane to the impact or otherwise) on VI.

  35. @John Pilgrim

    The A1. The M1 goes nowhere near Scotland. I’m constantly shocked that Scots pay so little attention to anything that happens in the ancient kingdom of Northumbria.


    “Looking at some of the comments that rightly got caught in moderation, can I just remind people think this is NOT a place to discuss whether government (or opposition) policies are any good or not, especially not controversial ones.”


    One can sympathise without what you have to deal with at times, but it can be tricky all round really. One difficulty with this is that it is the controversial which tends to be of interest, as opposed to just stating the obvious. This is amplified a bit because not unnaturally, you tend to mention quite a lot of the more straightforward observations in your initial post on a poll, leaving the rest of us to explore the more tenuous and controversial.

    This is before we get to the issue that the controversial at times may have bigger impacts on VI. The kicker is, that the stuff which is most toxic to a party, is liable to result in the most heated exchanges among the more party-politically inclined, precisely because it may have the greatest VI impact.

    Of course, if there is hard polling data on the more controversial, then that can neutralise some heated objections. This depends on the willingness of polling companies to ask the hard questions though…

  37. (sympathise without -> sympathise with)

  38. “Senior Labour figures have serious questions to answer about past links to vile paedophile support group”

    I see the Daily Mirror is raising the same concerns as the Daily Mail, so not just a Mail smear campaign.

  39. @Carfrew,

    I agree up to a point about the controversial issues being potentially the ones that shift VI, but I do think this is slightly overblown a lot of the time.

    If you added 0.2% to Labour VI for everything they did that (by the consensus on this board) was “likely to appeal to the electorate” and subtracted 0.2% from Tory VI for everything the Tories did that was “likely to lose them votes in the long run” then the lead would be about 30% not 6%.

    Of course, noone can ever be proved wrong, because there are so many factors, allied with MOE, that determining the effects of any given policy is nigh-on impossible.

    For me, the fall in Tory VI can be adequately explained by a combination of “UKIP surge”, “massive cuts in public spending” and “omnishambles budget”. The rise in Labour VI can be adequately explained by “former LD voters who didn’t like them going into coalition with the Tories”.

    I struggle to find any grounds to suspect any other issue of having had a lasting, measurable effect on the overall lead. Most of the things we debate furiously here (to AW’s dismay) are probably barely even noticed by the wider public.

  40. @The Other Howard,

    The involvement of the NCCL in supporting the “rights” of pa*dophiles isn’t remotely news (although I didn’t know that Harman had written specifically on the subject).

    The political ground has shifted considerably under our feet on issues to do with children’s sexuality. Once upon a time it was considered a legitimate political cause to seek to lower the age of consent (even radically so). It was a minority view, but that was sort of the point of the NCCL – supporting radical minority views.

    Of course there is a distinction between “pa*dophile” (a psychological state / disorder) and “sex offender”. Although with hindsight, PIE was very definitely the latter (I know, at one point I investigated some of them).

    Interestingly, if you look at the contributors to the Home Office guidance on sentencing for indecent image offences, one of those named is a convicted offender and advocate of such “rights”.

  41. @OldNat

    “Since a Westminster election will happen months after the referendum vote and so, it seems likely, will be affected by that result, then asking about referendum questions first has the advantage of chronological logic behind it.”

    The 2015 general election will take place some 8 months after the referendum but a year before the 2016 Holyrood elections. Asking a Westminster VI question last a few seconds after questions on the referendum, Holyrood constituency vote and Holyrood list vote doesn’t even meet satisfy a (pretty spurious) chronological logic either.
    “Unless polls only ask one question at a time, then there might be an inevitable tendency for the first question to affect the answers to the subsequent ones.
    That you haven’t previously seen a problem with polls asking multiple questions (AFAIK), that you see it as a problem on this occasion seems a little bizarre.”

    Oh I can assure you that I have previously seen it as a problem. And I trust that you are aware that AW raised the general issue of question ordering in the header to this thread: (“Now, the last time we saw an Panelbase/SNP poll they played silly buggers with the question ordering…..”). So it seems a little bizarre to me that you should be questioning my raising the issue of question ordering in this thread of all places.

    My view on this is fairly straightforward. For responses to questions on Westminster VI to be treated as reliable, they have to be the first substantive questions asked in a poll. If a pollster tacks them on at the back end of a poll focused earlier on something else, which is what Survation did, the results can’t be treated seriously.

  42. the comments here and the contributors are clearly those of people who care albeit they have different views on what caring is.
    I remember Cameron once saying that we need to be at the heart of Europe after all we don’t want to end up like Switzerland.
    at some stage it might be interesting to have a thread on that poor sad country according to Cameron. perhaps it can be explained to me how come when I went there with my first swiss wife I got 12 swiss francs to the pound. yes 12. so what puzzles me is how much do we not want to be like Switzerland as I cant even bear to look at how few swiss francs to the pound we get now. in fact it isn’t plural it is one point something.
    so mr Cameron what is it that is so wrong with poor Switzerland ?
    what system do they have for their government and can we learn anything from it?
    perhaps not but it is certainly worth a look instead of us assuming that whatever our system is it must be the best.

  43. @Neil A

    Well, you make some good points, and I’ve been through some of this before, so just to summarise:

    – Yes, it can seem as though only a few big events have significant VI impact. However, when AW posted the graph of the rise of UKip, it seemed to be associated with a succession of smaller events.

    – Bigger impacts may occur in tandem with more gradual trends, and the former may help to obscure the latter

    – some things may in fact shift VI, but are cancelled out by other things occurring contemporaneously.

    – What does not shift VI may be as interesting as what does

    – Small things may compound. The pasty tax on its own, could hardly be taken to be an omnishambles. Hard to see any one measure as an omnishambles… pasty tax on its own, maybe not much VI impact… but the pasty tax added to other things… becomes omnishambles…

    – Part of what determines salience is the info. people have, or whether something will become more salient down the line. Eg, I was concerned in the past about the flooding hostage to fortune. People weren’t concerned about services so much in 1992. By 1997 however…

    – polling can test what people WOULD find salient if they had the info.

  44. @Neil A

    I should add… that parties may respond to polling. So something does shift VI negatively, say, but then they endeavour to fix it. Whether the effect persists may depend on how good the fix. In other words, you get two significant VI effects, but which again cancel. One would wish polling to capture both, rather than assuming nothing actually happened to VI at all…

  45. Or to put it another way, because polling provides a feedback mechanism, maybe it helps foster polldrums…

  46. @ Neil A

    Re Harman & others in NCCL

    What the DMail are asking is people to comment on work they did nearly 40 years ago. How many people would have the memory of documents they produced or access to the complete set of documents produced. Perhaps NCCL wanted some of these minority groups to be out in the open, rather a hidden organisation. For all we know NCCL may have been working with the Police and Home Office at the time.

    While I think that it is legitimate for the Dmail to ask questions, their current articles appear to more of an opportunity to smear political opponents. Perhaps the Dmail should be contacting Liberty to ask them to investigate what happened in their organisation going back to the 1970’s and 1980’s. Shami Chakrabarti is the current director of Liberty, so I would have thought that all such enquries should be passed to her.

  47. Having heard Salmond’s latest comment about ‘Independence for Scotland means Scotland will never have to put up with another Tory government from Westminster’, that reads in reverse ‘that the rest of the UK after Scottish Independence will not have an election result where Labour and LibDem MPs from Scotland cause a bias in the UK election’. It would seem that comment is a win-win for both sides.
    From the opinion polls available, it appears that the Scots have already made up their minds one way or the other and nothing said from now on is going to significantly change the percentages for Yes or No, the ‘Don’t Knows’ appear to be ‘the Don’t Cares’. If the recent currency and EU discussions do not change the mind of the ‘Don’t Knows’ then it seems nothing will.

  48. [Snip]

    The dynamic of the debate is changing, with a sense of momentum behind the Yes campaign, and clear polling evidence that the status quo is the least favoured option. So in effect, Cameron has saddled unionist and BT with the single option of defending the least popular option.

    Almost by definition, this forces BT to take a negative stance against Yes, but it also enables Yes to get away with basing their case on an awful lot of ifs and buts. It makes any move towards discussing changes within the union, like devo max, difficult, because in truth, that isn’t what people are being asked to vote on.

    The fracturing of the No campaign is interesting to watch. There are different visions of independence, but in all practical terms, it’s the SNP running the Yes show. The No campaign, however, is altogether more complex. (Could that be the next slogan?).

    There are splits between and within parties on devo max, and because this isn’t on the ballot options, it’s very difficult to bring forward an ill defined devo max case within the confines of a binary choice. The Yes campaign can live by supposition, assumption and uncertainty (‘it will all be negotiated after a yes vote’) and get away with it, but without any other option but the status quo being on offer, it’s very hard for the No’s to offer a similarly vague prospectus successfully.



    I think that must have been a long time ago when you got 12 CHF to the £. When was that? For ten years now, indeed, the CHF has more than matched the € in trend against the £. The CHF was worth 42p and is now worth 67p.

    Sobering also to consider that 10 years ago the € (that ‘failed currency’) was worth 67p and is now 83 pence. Some failure!
    Interesting to note that just at the time of the bank crash the € was at parity with the £ and has now depreciated to 83p. That has, I assume, been to the benefit of the EZ member states in terms of exports. I suppose a ‘crossroads’ state such as Switzerland is immune to balance of payments issues.

1 2 3 4 5 8