Newsnight has made a bold attempt at actually commissioning some polling to try and shed some light on whether there will be substantial immigration to Britain from Bulgaria and Romania once restrictions are lifted at the end of this year, commissioning polls from Vitosha in Bulgaria and Gallup in Romania to find out what proportion of Bulgarian and Romanian people actually are likely to move to Britain.

Let’s start with the Bulgarian survey. 37% of Bulgarians said that in the last 5 years they had considered moving to live and work abroad. “Consider” is a fairly low bar to begin with though, does sitting back and pondering whether it would be quite nice to live in the south of France count as “considering moving to live abroad”, or does it require serious consideration? To put this in context, in 2012 YouGov found 6% of British people are actively considering moving abroad and 42% would seriously consider doing so. Have half the British population actually upped sticks and left? Of course not, for most of those people it was an idle whim or a pipe dream. No doubt it is the same for most of those Romanians and Bulgarians surveyed.

To try and set the bar a bit higher, therefore, Vitosha and Gallup asked people whether they had any actual plans to move aboard, 31% said they intended to go and work abroad in either 2013 or 2014. 19% said they are looking for work abroad, 15% they have actually started to prepare plans to work abroad.

The next consideration is whether those people are looking to move to Britain – there are, after all, several other countries in the EU! Asked where they are planning to go, just under of third of those intending to go to seek work abroad in 2013 or 2014 said Britain (it looks as though they could say more than one place) – equating to 9% of the total sample.

This is still largely just measuring aspiration though, so the poll then asked if they have made any concrete preparations for the move, 52% said they had, 47% had not (in terms of what this meant, a majority said they made been in contact with someone working in Britain, half said they had looked for a job there through a recruiting agency, 16% that they’d looked for somewhere to live). What it boils down to is that just under 3% of Bulgarians say they have looked for a job in the UK through a recruitment agency, just over 1% without.

The working age population of Bulgaria is just under 5 million, so in the unlikely event that all those Bulgarians who have enquired about job opportunities in Britain find one (and the majority of respondents said they were only interested in moving if there was a firm job offer, hardly anyone said they were planning on moving speculatively), it would equate to something under 200,000 Bulgarians.

The Romanian survey was structured in much the same way, though there was less interest there in moving to the UK (the most popular destinations for would-be Romanian emmigrants were Italy and Germany) and those that did mention the UK as their favoured location were less likely to have actually made any concrete plans. Only just over 1% of Romanians had made any attempt to enquire about job opportunities in the UK. The working age population of Romania is about 15 million, so in the equally unlikely event that all those Romanians who have enquired about job opportunities in Britain found one, it would equate to something under 150,000 Romanians moving to the UK.

I would still urge a lot of caution with even these figures. The margin of error on a normal poll of 1000 is plus or minus 3%, so one should hardly read too much into figures of about 4% and about 1%. Equally people will naturally overestimate their likelihood of taking major life changing decisions – it is far easier to ring up a recruitment agency and ask if they have any jobs going in Britain than it is to actually uproot your life and move to a foreign country, far easier to look for a job than it is to find one. What we can say with some certainly is that bonkers claims about half the entire population of Bulgaria and Romania moving to the UK are, indeed, still bonkers.

261 Responses to “Newsnight polls of Bulgaria and Romania”

1 2 3 6
  1. AW

    I think there have been a number of responses to this in the previous comments .
    Did you want to move them across?

    On a different topic Ed Balls – Shadow Chancellor – 5:14:13 officially a faster finish than Mo Farah!

  2. Anthony,

    Before you parse the semantics of the questions too carefully, you need to look at the questions in the original languages to ensure that your ‘bars’ are not shifted up or down.

  3. Regardless of what the BBC’s poll show the one thing we can take from recent history that any attempt to quantify the number of immigrants likely to come into the UK from any given country is doomed to be wrong.

  4. Although (nerd alert) MoE actually decreases with response probabilities further from 50%. For a response of 4%, MoE is +/- 1.2%.

  5. So if the total is around 350K, times two for average family members coming too, and say another 20% for those who are not telling.

    Which could be as much as 850,000 people. Hoe easy it is to get into the UK and milk the system will decide if such a number increases after that. I’d guess at 1.5 million over 3-5 years.

  6. Jon – indeed, though in the absence of a copy of the survey in Romanian and Bulgarian and conveniently available native speakers of the languages, it’s not something easily done!

  7. Statgeek – you can’t times two for average family members.

    Firstly, the poll was of all working age adults, so spouses or partners coming are already reflected in the figures, you would only have to account for any extra dependants who are not of working age.

    The surveys actually asked this, and between 17% and 20% of people said they would take their children.

  8. Anthony, please could you clarify what has happened to the YouGov poll on the county council elections, which we had been given to understand had informed R&T’s estimates of gains and losses. Does it exist?

  9. @lefty

    Re: Moore’s Law etc.

    Firstly, props to you for flying the flag for engineering in an era of services and banks preferring to lend elsewhere etc.

    I dunno what you have to do for structural analysis but if it involves stuff like tensors and the finite element thing, or other nasty maths like that, that stuff is really rather hard.

    On Moore’s Law, I remember in the mid-nineties there were some things I wanted to do (to do with audio etc) that at the time required waiting for things to render and I realised that I may as well wait a while and Moore’s Law would mean I’d be able to do them in real time soon enough.

    Terahertz computing makes a lot of things possible in principle. The difficulty at present as I’ve no doubt you are aware is it’s a huge ask to get that speed of calculation on a single chip (for those not down with the terahertz thing we are talking about calculations a thousand times faster and more than currently) because of the amount of heat generated and how small the circuits on the chips need to be (leading to quantum effects that can screw things up).

    There’s talk of using graphene, and long-term… quantum though I’much not up to speed on the feasibility… will it happen? Alternatively they can run lots if chips and cores in parallel though that makes coding rather harder. I suppose the modelling you are talking about can be done that way?

    The thing that really fascinates is the potential for AI, cost they reckon the human brain runs at… is it around 10 Terahertz I think? But of course having the chip is one thing, having the program to run on it and simulate intelligence is something else. I wonder if having the speed might make some approaches more viable, like something emergent. Or some vast neural network. But given we already have some – admittedly vast – terahertz machines maybe we would already have figured that out.

    Still, might add something to robotics. Running their own real time models? …

  10. Regarding the “national share of the vote” projections of R&T from local elections, I’m quite sceptical as to how much reliance can be put on them, given that there must be some quite heavy judgement calls involved.

    That’s particularly so this year. The local elections are being conducted almost exclusively in what have always been the strongest Conservative areas of England, which since the creation of unitary authorities in the 1990s and 2000s have become even stronger by being largely shorn of their more urban Labour supporting parts (e.g. Nottingham, Derby etc). Only a handful of unitaries are holding elections at the same time.

    What I expect to happen is something like this. In the quarter or so of the parts of counties where Labour is competitive (the likes of Corby, Ipswich, Cambridge, Gravesham etc), it will secure significant increases in its share of the vote compared with 2009. That’s also much the pattern we’ve seen in the areas that aren’t polling this year. In the remaining areas, which are far more typical of county councils, Labour’s gain in vote share will be quite modest. At least, that’s the sort of pattern we’ve seen in by-elections over this parliament.

    So how do you make up a national projection if that is the pattern? Labour is indeed competitive in most of the areas that aren’t polling, that is the met districts, remaining unitaries, London and the majority of Scottish and Welsh seats. I don’t think you can project forward results in the deepest Conservative rural heartlands when assuming how other very different urban areas would have voted. What matters above all is going to happen in the Con/Lab marginal seats.

    The question is, when making national projections of equivalent vote share, how much notice is taken of differential swings in different types of seats? Because if it’s all done on the basis of uniform averages, I’ll be pretty sceptical of accepting it at face value. And if it’s more sophisticated than that, it would at least be reassuring to look inside the black box to see exactly what assumptions being made.

    So does anyone know whether the full methodology of the national vote share projections is published in enough detail for it to be subject to critical appraisal. Or do we just have to take these figures and the judgements made at face value?

    To put it in context, imagine if on this site we had to frame the debate around the results of one large but very infrequent opinion poll, conducted across a very biased sample but then radically reweighted by “experts” who in passing also put in place very significant other adjustments, none of which were properly disclosed let alone supported by published tables showing calculations. The expects would just tell us to stop fretting and rely on their good judgement. But we might in such circumstances be quite dismissive of the value of their occasional opinion polls and unsure of how to interpret it..

  11. Turk,

    “we can take from recent history that any attempt to quantify the number of immigrants likely to come into the UK from any given country is doomed to be wrong.”

    I don’t think we can say that at all and it is a particularly daft thing to say here.

    Much of the discussion here is about methodology and changes to improve accuracy. As Anthony has repeated often when people mention polls being wrong in the nineties, a lot was learned from that and big changes were made. Internet polling alone has lead to big strides forward.

    To say that two polls conducted with sound methodology to try to more accurately predict what were previously little more than guesses can’t do any better than what went before hardly bares scrutiny.

    I’d be happy to here why you think these polls are flawed but I think few will listen to the argument that it can’t be predicted or you can do better with better methodology.


  12. Phil,

    I have no idea what’s actually inside the black box, but they have the vote shares (or at very least, the percentages of councillors returned from each party) from previous council elections in those counties, and the equivalent national poll numbers. Couldn’t they use that to train the model?

    How they translate that number into a general election projection is a really good question, since people behave differently in different types of elections. But I don’t think the biased sample is necessarily a problem, because you’re comparing it to previous biased samples.

  13. AW


    Yes but something of a straw(or yellow trousered?) man.

    Irrespective if what I or other cerebral analysts of the +/- of immigration say, it will be about 350,000 too many to the majority of voters for all 4 of the main UK parties, surely ?

    It is almost beyond belief, even IMHO as a Labour supporter that one of the 4 main EU countries can or should be compelled to accept this when we have an economy which is not growing, which to me means that a third of a million new arrivals will take a lot of jobs which would otherwise be available to existing UK citizens or long term residents.

    Part of me wonders whether the time had come to warn the EC that we will not accept this, and dare them to expel us if we dig in. Not that I particularly want to give DC good campaigning lines , but ….

    I know some posters will try to say this is prejudice vs fact eg that most immigrants work hard and contribute to GDP. But try that on the doorstep. You do have to listen to the people in a democracy. It’s certainly time for Labour to take a much harder and more pro-UK line on the EU. As indeed icislists in France and other member states have done for years.

  14. @Welsh Borderer
    “which to me means that a third of a million new arrivals will take a lot of jobs which would otherwise be available to existing UK citizens or long term residents.”

    Why do you think the UK employers will give the jobs to the new arrivals rather than the people living here? Surely if we address that point, our residents will have nothing to fear from new arrivals.

  15. AW

    You mention immigration (even on a forum that is about polling) and you get responses which have little to do with polling and all to do with perception.

    However, on the subject of polling “Concerns” about “swamping” by migrants are at their greatest in areas where the actual number of migrants are at their lowest and the ethnic diversity of the population is below the national average.

    In inner city areas where the population is diverse and has been for generations such as Inner London where I lived and worked for decades the concerns are far less as reality proves the worries are primarily based on ignorance and are massively inflated.

  16. Regardless of whether 350,000 new arrivals take jobs or live on the dole, the infrastructure required to support them would be massive. It’s equivalent to a city the size of Leicester.
    Immigrants often tend to live close to each other for mutual support, so wherever they congregate there will be massive strain on schools, hospitals, utilities etc.

    I wonder what the results of a poll of UK residents about the predicted influx would say?

  17. peter cairns


    “we can take from recent history that any attempt to quantify the number of immigrants likely to come into the UK from any given country is doomed to be wrong.”

    I don’t think we can say that at all and it is a particularly daft thing to say here.


    I think you are being unfair Peter: that’s not particularly daft for turk surely?

  18. I think the most salient point from this poll is that most won’t move to the UK without being sure of employment

  19. Liz H

    The answer is surely about relative wage levels – minimum or sadly/illegally even below minimum wage level in UK is much higher than Bulgaria or Romania, where they will remit a hefty chunk of their wages.

    This is sadly inherent in EU – a continent-wide free wages market where free movement drives down real wage levels. Have we yet heard how EM intends to counter this ? Realistically how would progressive parties ever have a majority in EU at Council and Parliament level to insist on measures such an EU minimum wage ? Never in a hundred years ?!

  20. Regardless of whether 350,000 new arrivals take jobs or live on the dole, the infrastructure required to support them would be massive. It’s equivalent to a city the size of Leicester.

    -It would only be equivalent to a city the size of Leicester if they all moved to the same place!

    As you rightly pointed out most economic migrants move to areas where there is likely to be work and community support, primarily urban areas and as most migrant workers are perforce of working age and normally without any dependants their is very little actual additional pressure on services apart from in a few pinch points such as Dover.

    I defy even the most hardened UKIPER to point out a single council or service provider that has engaged in massive infrastructure construction because of migration.

    A shame really as it might get the economy moving a bit.

    Regarding the “dole” as I am sure AW will confirm Economic Migrants are far less like likely to claim any benefits at all compared to a cohort of UK born contemporaries


    If it is about wage levels and UK residents wont take up a job unless employers pay more, then stopping migrants is not going to resolve this issue. Instead of generating fear of migrants, it would be far better to campaign for better wages which would also reduce our welfare bill. We should not be subsidising employers who are not paying their staff a decent wage.


    What about stopping people leaving the UK when the economy is doing well, or banning anyone from moving to places like American where they can earn more draining skills from Great Britain.

    What exactly is this ultimatum;

    “You can’t come here to live and work, but you still have to open your markets to our goods and let your people invest their money in London to avoid a Tobin Tax, oh and we’re still allowed to close factories in Britain and move jobs to Poland where Labour is cheaper…..”


  23. Folk forget that it was britain that was most enthusiastic about extending the eu eastwards. Apart from germany the rest of the eu were not so keen, maybe because they saw these problems coming?

  24. Richard,

    We were hoping to shift the balance towards free market free trade policies away and a more business focused Europe from on about social protection and regulation.

    People getting on their bikes and coming here is a direct consequence of the Open Europe we wanted.


  25. 350,000 is, from Antony’s article, clearly a top whack number, and extremely unlikely. For all I know it might be a tenth of that. Does anyone know of any data about what percentage of people considering getting a job outside their home area actually do so, anywhere?

  26. I’m broadly unconcerned with Romania and Bulgaria for one reason and one reason only: we’ve already had doors opened to Poland, Hungary, the Czech Republic, Slovakia, Slovenia and the Baltic states. And whilst it was planned poorly, the UK’s welfare state hasn’t collapsed under the strain. I really don’t see another two being a problem – at least, not one we can’t easily cope with.

    My main concern is that the Cheeky Girls might carry on being interviewed for their apparently comprehensive insight for all things Romania.

  27. I think some of the contributions here prove that the ploy of claiming ‘swamping’ by immigrants is a very effective one electorally, so not so ‘bonkers’.

    If some of the contributors to UKPR are in this framne of mind, would it not be fair to state, as one did above, that would succeed ‘on the doorstep’?

    They don’t read the BBC poll ‘on the doorstep’ and most voters would not have either the intelligence or the interest to do so.

    Let’s have a poll on how many have read and understood the BBC poll to see if I am wrong!

  28. @Howard

    Claims of immigrants swamping the UK may very well be very good at convincing people who were never going to vote Labour or Lib Dem that they must vote for UKIP. For the rest, Ed has already made the suitable noises about understanding the issues. Which he does, he understands that there isn’t really an issue. But he’s being nice to the people who think there is one, and promising reviews and better enforcement… But if someone is a voter who wanted concrete commitment to keeping the foreigners out as their vote deciding issue, they were going to vote UKIP anyway and it’s not worth Ed’s time chasing too hard after them.

  29. Petercairns

    Labours estimates for polish workers was some 50,000 that turned out to be over half a million.
    I would have thought that if you indeed understand polling then taking a single poll on anything is a waste of time, polls need to be taken over a period of time and fairly frequently otherwise all you have is a single view expressed on a particular day so no it’s not daft to suggest the BBC poll will probably be pointless.

  30. @Spearmint

    It’s not necessarily an insurmountable problem. But it will mean that outcome of the model will be much more dependent on the the assumptions made, should there be a lot of variation between different types of seats.

    A parallel with opinion polling might serve to illustrate my point. Suppose that AW conducted a YouGov poll only in rural Conservative seats. Even with heavy weighting to be more representative of the UK population, it would still be risky to predict the outcome in most Con-Lab marginals on that basis..

  31. Paulcroft

    I realise you think we all hang on your every witty word, but it’s not good manners to throw insults at people who post on these pages, please feel free to disagree with what I say but but keep it respectful.

  32. Turk
    That’s like saying tonight’s YouGov poll is a waste of time because opinion could change a week later.

    Could you attempt to say something new on this matter please?

    This was a poll not an ‘estimate’ by some bygone administration. Did they poll?

  33. @Turk
    “Labours estimates for polish workers was some 50,000 that turned out to be over half a million.”

    It wasn’t “Labour’s estimate”. It was an estimate arrived at in research conducted by the relevant parts of the civil service, which was then presented to ministers and which ministers relied on when forming their policy. That’s hardly the same thing.

    Likewise, the estimates no doubt now being drawn up on the back of research including the likes of this poll won’t be “Conservative estimates” (let alone “Con-LD estimates” ). There is one difference, however. Current ministers unlike their predecessors do have the benefit of hindsight when choosing whether or not to dismiss the possibility that official estimates on EU migration from accession states can prove to be way out. And if they plan on a best case scenario, they deserve to be held accountable if things go wrong.

  34. @AW

    I was thinking about older or elderly relatives too, and those who are not considering it now, but will if a family member gets established in the UK.

    I knew a family with plenty of prospects who migrated en masse to Australia when one of them recommended it (siblings, parents, six in all). It’s not that uncommon.

  35. @BillPatrick

    While I don’t see a problem with this scale of migration, the 350,000 isn’t unlikely. There are about 100,000 here from these two countries today.

    The problem is the underinvestment in social services. For about 3 years 600,000 people from EE paid extra tax…

  36. Sorry my mobile corrected what I typed. I meant the previous post for PatrickBrian. Strange correction. Should switch it off.

  37. @Paulcroft – have you nothing more to say? I’m hanging on your next witty word. Do please say something.

  38. Quite interesting to see some non funeral comment reaching escape velocity in the media now. There has finally been a bit of comment on the HoC report into the housing loan scheme. the more I see of the report, the more devastating it gets for the scheme. Half baked seems mild, but to see a central part of a very political budget roundly trashed by his own MPs must worry Osborne.

    On another issue, the Markit Household finance index for April was published this morning. It’s remarkably bad news, and runs oddly counter to proclamations that confidence is returning.

    The index has been below 50 for months now, but April saw the index worsen for the first time this year. Income from employment fell back on the month, and the figures, if accurate, suggest that the raising of the tax threshold has been offset by falling earnings.

    I haven’t been following this index for long enough to get any feel for either how accurate it is or how it feeds through into future economic numbers, but I would intuitively have thought that with the change in dynamic of the economy to a position where we are now totally reliant on the service sector for growth, rather than for manufacturing, these numbers really are quite bad.

    Unless household incomes start to rise, it’s hard to see any sustained service sector growth, especially if we are now also seeing unemployment start to shift up as well. With manufacturing and construction slumping and exports declining (including sharp recent falls to non EU countries – can’t blame Brussels there) if household incomes are starting to be squeezed again, Osborne really won’t have anywhere to hide.

  39. So I guess this is going to be in the running for worst headline in the UKPR awards –

    “25% of Bulgarians want job in UK”…

  40. Wes – that’s what papers get for taking the PA feed. It’s reporting of polls is often poor (the Standard itself, in contrast, normally does a good job with them)

  41. Two out of three British families considering emigrating….

    That would leave space for about 40 million if it happened.

  42. @sen5c from the last thread

    “I’m very surprised,incidentally, at how poorly the Greens appear to be performing, particularly with such a yawning gap available beyond Labour’s more-of-the-same-but-slower position and the Lib Dem’s no-Liberal-belief-so-sacred we won’t abandon it in office.I know Green politics feel to an extent an indulgence people can’t afford in times of economic stress but still, I’d expected a lot more than their current declining fortunes.”

    I don’t see any evidence that we’re doing poorly. In my region we’ve gone from three principal authority councillors to fifteen over the last two years and are expecting that number to go up again next week. We’re making slow, but steady, progress in a political environment where Labour are at their strongest point for over a decade. We are also doing it whilst being marginalised by the media (at least in comparison with other parties).

    Basically, we’re not making massive progress or major breakthroughs, but we are definitely not in any kind of decline.

  43. James E
    That would be tough on the new immigrants who would be working to keep our baby boomers alive (includes me) instead of our own children.

  44. I haven’t been able to track down a reference to it, but didn’t the person who did the estimates of potential immigration very recently appear in front of a Commons select committee.

    As I recall he explain in detail that his estimates were taken out of context and that it was the press that had highlighted the low 50k figure rather than the Civil Service or the Government.


  45. Of course, it is likely that those who have already taken steps towards moving abroad are more likely to move here.

    But that doesn’t mean some of those who haven’t yet taken steps won’t also move to the UK.

    Is there any indication as to how many who haven’t yet taken steps may actually move here, based on previous figures?

  46. I mean, most people don’t apply for a job a year before moving do they? His many vacancies are held open that long?

  47. We need the Monday Yougov poll so we can start a new thread.
    Wonder which way the 4 Yougov polls will point this week
    Will the Labour lead narrow even more, widen again or stay the same?

  48. turk

    A very slightly thicker skin and a slight sense of humour can help a lot I find. Sorry your feelings were injured but my “witticisms”, as you call them, are just little things that vaguely amuse me: I have no expectation at all as to how they are received – but certainly not in the way you describe. I shall desist in future though in your case and hope that helps. You never told me how the mice are by the way…..


    I am chastened and am turning over a new leaf. I shall aim to make my jokes more ironic and less funny.

  49. Sometimes I really wish to know what would happen if all these foreigners (including myself) disappeared from the UK…

    I guess not even that would change public opinion. The public wouldn’t recognise why they would have to pay higher taxes for the same level of public service (or why it would deteriorate).

  50. @ Laszlo – your point is purely accademic as the ‘damage’ has already been done – mainly by Labour Administrations!

    It was blatant Social Engineering by Labour and everyone else just burried their heads in the sand!

1 2 3 6