1
Select Committee on the European Union
Home Affairs Sub-Committee
Corrected oral evidence: Brexit: UK-EU movement of people
Wednesday 21 December 2016
10.30 am
Members present: Baroness Prashar (Chairman); Baroness Browning; Lord Condon; Lord Jay of Ewelme; Baroness Massey of Darwen; Lord O'Neill of Clackmannan; Lord Ribeiro; Lord Watts.
Evidence Session No. 4 Heard in Public Questions 50 - 59
Witnesses
I: Mr Ian Cope, Acting Deputy National Statistician, Population and Public Policy, Office for National Statistics; Mr Paul Vickers, Head of Population Statistics Outputs, Office for National Statistics.
Mr Ian Cope and Mr Paul Vickers.
Q50 The Chairman: Good morning and thank you for your time. This is a public session, which is being webcast. After the session, you will be sent a transcript. If you wish to make any factual corrections, feel free to do so. If, during our discussions, you find that you want to give further information, we would be grateful if you could send it to us. Thank you for your written evidence, which we greatly appreciated. As you know, we are looking at the nature and flows of migration to try to understand how many people are involved, where they are from, the number of people coming to study and so on. Against that background, could you give me a brief overview of your role in the collection of migration statistics and how you work with government departments in conducting your work?
Ian Cope: I am delighted to appear here this morning. The ONS is the national statistics institute for the UK. We produce a range of statistics on economics, population et cetera. On migration, we produce estimates on long-term and short-term international migration, from the international passenger survey. We produce information on—
The Chairman: Would you mind speaking up a bit? The acoustics in this room are really bad. And please go slowly, as we are looking at figures.
Ian Cope: Sorry. We produce information on long-term and short-term international migration, mostly from the international passenger survey. We also produce information on non-UK nationals resident in the UK, from our annual population survey, which is an annual boost to the labour force survey. We pull together information from ONS sources and from other government departments—HMRC, the DWP and the Home Office—in the migration statistics quarterly review, which gives the latest snapshot of migration. We believe that the international passenger survey is fit for purpose for producing those estimates at the national level, but we recognise that the demand for information on migration has increased significantly over the last few years. It cannot answer all these questions—for instance, local impact on employment, housing or schools—or meet all those new needs.
For the future, having greater access to administrative data already held by government and unlocking the power of information already held within government by the DWP, the Home Office and HMRC would provide an opportunity better to measure the wide range of aspects of migration in which people are interested these days. The Digital Economy Bill, which is going through Parliament, provides the potential to make that much easier and faster than it currently is.
To go back to migration, it is important to recognise that migration is of interest in its own right. It also feeds in to our population estimates. We take the census estimates and we measure natural change—births and deaths—and we adjust for international migration. Following the UN definitions of long-term international migration is important for the population estimates as well as for the migration estimates.
The Chairman: Do you want to add anything, Mr Vickers?
Paul Vickers: No, thank you.
The Chairman: You said that the IPS is not fit for purpose, Mr Cope.
Ian Cope: No, I said that it is fit for purpose for measuring migration at the national level but that a wider range of new demands and questions are being asked about migration. For instance, the IPS cannot tell you about the local impact of migration on schools and unemployment.
The Chairman: What do you see as the general shortcomings? What changes would you like to see to get a better picture?
Ian Cope: In terms of data sources, the ability to access the administrative data already held by government at the individual level would help.
The Chairman: You cannot access that at the moment.
Ian Cope: No. We have access to aggregate data and we include those in the migration statistics quarterly review, but legal and other barriers make it difficult to share data across government, which is why we are very excited about the Digital Economy Bill opening up access to those individual-level datasets. You could, for instance, link information on visas for people who are studying to understand more about whether students are going into the workforce after they finish studying. For example, if they are not paying national insurance, that is evidence that they have left the country. That could help to improve our understanding of student migration. Do you want to add anything on that, Paul?
Paul Vickers: There has been a change in the landscape in the demand for migration statistics, certainly over the last few years. Ten years ago we were concerned about the flows of migrants into and out of the country. As Ian said, the international passenger survey does that quite well. When you want to look at the characteristics of those migrants who are in the country—what they are doing and their impact on the economy in terms of the taxes that they are paying and the resources that they are using, such as hospital and school places—that is where this power of linking together a lot of the administrative sources will help us to unpick that information and provide a greater picture of what is going on.
The Chairman: Do the statistics on the country of birth complicate the picture? If so, how relevant is that to EU migration in particular?
Ian Cope: When you look at migrants, you often look at either country of birth or nationality/citizenship. Someone born abroad could be the child of people in the Armed Forces who are serving in Germany; they will have been born in Germany but will be British citizens. It depends partly on the question that you are interested in. Country of birth is relevant in that, if those people now living in the UK were born abroad, at some point they will have come into the country, but in terms of eligibility to public services et cetera, nationality/citizenship is relevant. The statistics give you different aspects and it depends on the question that you are asking.
Paul Vickers: Just to illustrate, 40% of the 8.6 million people in the UK who were born outside the UK have British citizenship. If you just look at the country of birth, you could come to findings that are not relevant to the policy debate.
The Chairman: You said that the landscape has changed in terms of the demand for figures. In what way has it changed? What could you do to make sure that you meet the current demand for figures?
Ian Cope: The international passenger survey is large—700,000 interviews. As we said in our evidence, there are large flows in and out of the country, most of which are nothing to do with migration. We do not think that trying to increase the size of the international passenger survey is a sensible or cost-effective approach. We have made a number of improvements to the IPS in the past 15 years or so, but we do not think that further improvements or expansion to the IPS would be the right answer. The better solution is to be able statistically to link data held by the DWP, the Home Office and HMRC in a safe and secure environment inside the ONS in order to be able to unpick many of these complex questions.
Q51 Lord O'Neill of Clackmannan: One of the criticisms made of the IPS work in relation to students was that the bulk of students were on flights that arrived or left at times when you were not carrying out any surveys, because it was apparently too late at night, or something like that. Is that a resource issue or an issue of judgment that even then it would not have made much difference to the quality of the stats that you would be able to analyse?
Ian Cope: I will start and then hand over to Paul. You will see from the annexe that we have made a number of improvements to the IPS, including reviewing periodically where were are carrying out the survey. We have expanded the range of airports that we survey, because it was evident that we were missing flights coming in from, say, eastern Europe. We have also reviewed the times at which we carry out the interview shifts. We compare those shifts to Civil Aviation Authority flight data to try to ensure that we are getting the best coverage. I do not think it is primarily a resource question.
Paul Vickers: No. We released a report at the beginning of December on the IPS, which we can forward to the Committee and which goes into more detail about the quality of the international passenger survey. One thing that that looked at is whether the sample that we survey between 6 am and 10 pm is representative of all flights. Those findings suggested that it was, but that is not to say that we take that at face value.
One thing we are considering in the new year is carrying out a quality exercise, which we do periodically to make sure that the out-of-hours flight survey is representative. The last time we did that was at the beginning of 2014 when we were looking to see what impact these flights might have on migration from Romania and Bulgaria as they joined the EU. As a result of that we did not have to increase our out-of-hours survey because it was representative of the flights for the main survey.
The Chairman: So what is the reasoning behind including UK citizens arriving back in the UK when measuring net migration?
Ian Cope: With the international definitions, somebody who leaves the country for more than 12 months will be in an international migrant. If they are abroad for a period, a number of years, and then come back, they are an international migrant back into the UK. Paul has had some personal experience of that.
Paul Vickers: Ten years ago I went to Australia for two years, and while I was there of course I did not use any UK resources and was not part of the UK population. When I came back, there needed to be a way of including me in the population again so that the right allocation of resources could be made. It is important that British citizens are included when they return from being abroad for more than 12 months.
Baroness Massey of Darwen: Thank you for your explanation of the IPS. I am a little puzzled as to how it all now works. You mentioned a link with the DWP. Can you explain how that all fits together? You know that the House of Commons Public Administration Committee report, admittedly from 2013-14, was critical about some of the tools that are used. I assume that things have been updated in response to that report and are now quite different. Are you now linking with other agencies and other ministerial departments to make the tools more sophisticated? How is it working?
Ian Cope: The legal barriers that I talked about have not changed since 2013. We are very excited about the Digital Economy Bill, which, if passed by Parliament, will provide a legal gateway to give access to a lot of that data. At the moment, under the Statistics and Registration Service Act, all data sharing requires an information sharing order, which has to be approved by both Houses of Parliament. It is quite a slow and complicated process. The powers for access to administrative data for statistics and research purposes that are included in the Digital Economy Bill would put us in a similar position to Canada or Ireland, where there is a presumption that the statistics office can access administrative data held by the Government for those statistics and research purposes. Many of the things that were talked about in 2013 we still cannot do because the law has not changed. We make use of aggregate administrative data, and Paul can say how we incorporate that in the MSQR, for instance.
Paul Vickers: We get information from national insurance numbers visa information, which describes another aspect of migration and gives us a broader picture of what is happening. We also use information from the Department for Work and Pensions to help us with our population estimates. At a national level we take the information that we have from the international passenger survey, and then, to distribute that to get local authority estimates of international migration, we use distributions from administrative sources such as the DWP and the Higher Education Statistics Agency data.
Your point about the real benefit of linking all the data goes back to what we want to find out about the population and the migrants in the population. I am quite confident in using the international passenger survey to measure the flows into and out of the country, particularly at the national level. When you break it down, the quality is admittedly less because fewer people are sampled for those statistics. There is an increasing demand for information on the impact of migration on benefits, taxes and the use of services. Having access to this administrative data is really important. Our dialogue with DWP and HMRC is very positive and we are already getting secondments into those agencies so that we can look at and understand the data.
The real power will come, as Ian says, when we get the legislation and can get the information into ONS and then link it. We will get a huge amount of information to enable us to answer these questions.
The Chairman: Can I be clear on one thing? If I understand you correctly, what you are saying is that there are no legal barriers to the IPS and your judgment is that it is good enough for the job?
Ian Cope: For measuring flows in and out of countries at the national level.
Q52 The Chairman: My next question is on whether you can give me a general picture. Would it right to conclude from the written evidence that you have sent us that net migration has not been below 100,000 since the late 1990s but has been below 200,000 as recently as 2012?
Ian Cope: Yes.
Paul Vickers: Yes, that is correct.
Lord Ribeiro: Before I ask you my question, as a matter of fact I cannot get all these figures on page 3 to add up. I wonder if you can help me. In terms of immigration for work and for study, there seems to be a difference of 13,000, which is not accounted to any particular group. Can you help me on that?
Paul Vickers: The reason is that when we produce our overall immigration figures, the total figure is based on the IPS but adjustments are also made to flows into and out of Northern Ireland and for asylum seekers. We do that at the national level, but when we break it down we do not include those figures. So what you have is what we call our long-term international migration estimate as the total, but we use only the figures directly from the international passenger survey as the figures leading into that.
Lord Ribeiro: The 311,000 is an estimate?
Paul Vickers: Yes, but it includes the adjustments for asylum seekers and Northern Ireland. The figures below do not include those; they are pure IPS estimates.
Lord Ribeiro: So basically asylum seekers and Northern Ireland should be “others” in that context.
Ian Cope: That is why we put the note at the bottom. I am sorry if it is not clear.
Q53 Lord Ribeiro: The next question is about the difference between EU/EEA and non-EU/EEA citizens coming to the UK. Could you give an overview of how these stats are collected and their relevance?
Ian Cope: They are collected from the international passenger survey. Sorry, what else was in the question?
Lord Ribeiro: You have these statistics on the numbers of EU and non-EU/EEA coming to the UK. I want to know what the differences are between these groups.
Ian Cope: At the broad level, EU citizens predominantly come into the UK to work, while the largest group of non-EU/EEA citizens comes to study. We break them down in one of the tables showing the reasons for immigration.
Paul Vickers: Ian is absolutely right. Table 1 on page 15 of the briefing pack that we gave you shows the distribution of the different reasons for migration between the two groups. As Ian says, the majority from the EU come in for work purposes, while most people from outside the EU come in to study.
Lord Ribeiro: We have actually seen a fall-off overall in students coming from the non-EU. Can you give any reasons for why that drop-off is occurring? It seems to be quite dramatic in some respects, compared with the other data.
Paul Vickers: Basically, there was a big drop-off around 2011, which was when there was a tightening of the rules on coming into the country going to further education colleges and English-language schools. However, the number of students coming into higher education institutions has remained flat over that period, and we have not seen a fall there. The main fall has been focused on further education and-language schools.
Q54 Lord Jay of Ewelme: I want to drill down into one or two of the further points that you raise. I thank you for the paper, which was extremely helpful. As you will have discovered, it raised certain questions in our minds about the validity of the data. Could you say a bit more about trends in migration from what one might call the old and the new EU member states and how they compare with public perceptions, if that is a legitimate question? On that point, I notice at the top of page 12 of your very helpful paper that you talk about the net migration of EU2—that is, Romanian and Bulgarian—citizens going up by 25%, but that that difference is not statistically significant. I have to say that if a 25% increase is not statistically significant, it makes you wonder about the statistics as a whole.
Ian Cope: Paul, do you want to handle the first bit about trends in the old and new member states?
Paul Vickers: Basically, between 2006 and 2011 the immigration from all the different EU groupings and countries within the EU remained relatively flat. Since 2011, however, we have seen some changes. In fact, from 2012 we have seen an increase in the old EU migration from the traditional EU countries, but not surprisingly there has also been an increase from Romania and Bulgaria. The interesting thing there is that migration from the eastern European countries, what we call the EU8 countries, has remained relatively flat over that period.
Lord Jay of Ewelme: So the old EU, as it were, and the eastern Europeans, have remained flat, but Romania and Bulgaria have gone up?
Paul Vickers: Romania and Bulgaria have gone up, and there has been an increase in the old EU, too. That is shown in figure 5 on page 14. The blue line is the figure for the old EU, which you can see has gone up in the last few years, while the red line is for Romania and Bulgaria, and that has gone up too. You will see that the amber line has stayed between 60,000 and 80,000.
Lord Jay of Ewelme: That has remained pretty constant, has it not?
Paul Vickers: Yes.
Lord Jay of Ewelme: That is helpful. Thank you. And what about the statistics?
Paul Vickers: The more you break down the international passenger survey, the smaller the numbers of contacts they actually represent. So when you work out your confidence intervals and whether something is a statistical change, you need quite a large figure in order for it to be significant change. We are merely reflecting that.
Ian Cope: So on the overall level we say that net migration is about 335,000, plus or minus a confidence interval of 40,000. As you break it down into smaller components, those confidence intervals get much wider, because, as Paul says, you have a smaller sample. This is really saying that that change is within the confidence interval.
Lord Jay of Ewelme: So how confident can we be that there has been an increase in net migration from Romania and Bulgaria if the increase is not statistically significant? Can we even conclude that there has been an increase?
Ian Cope: Those numbers are our central estimate. We give the information about confidence and so on to help users, parliamentarians and other commentators to understand the confidence we have in those numbers.
Lord Jay of Ewelme: On the second question, could you say a bit about the recent rise in EU inward migration from jobseekers as opposed to those arriving with a definite job? That is also an important distinction. Perhaps you could also point us towards the relevant chart.
Paul Vickers: Charts 6 to 8 in the attached document on pages 16 and 17 of our evidence show the increases of those people coming in to look for work. We have found that 189,000 EU citizens as a whole come for work. Of those, 57%, or 108,000, have a definite job to go to. This is all outlined in the paragraph below table 1 on page 15. Some 82% of those—43% of migrants—arrive looking for work. Over the last year, that increase has been a significant change. The chart shows that the EU8 has again stayed relatively flat over the last year. The biggest increase has been in the EU15 and the EU2.
Lord Jay of Ewelme: The chart on the top of page 16 suggests that there has been quite a sharp increase in people looking for work rather than having a firm job. Do you have any explanation for that? It is quite marked.
Paul Vickers: It would be speculation at this point.
Lord Jay of Ewelme: Have a go.
Ian Cope: All we can really do is give the evidence and the statistics. For the underlying reasons, I think we would need to interview people and try to understand what changes there have been. We would have to talk to people because we do not have that evidence, and I do not think it is appropriate for us to speculate.
Lord Ribeiro: Are the interviews done by questionnaire or directly face to face?
Ian Cope: They are face-to-face interviews as people are going through the airport.
Baroness Browning: I know the answer from what you have just said, but in the stats do you disaggregate people coming in to an existing job or looking for work and those coming with the intention of setting up their own business?
Paul Vickers: No, it is just whether or not you are coming for work. There is information on migration from the labour force survey that can be broken down. I do not have those figures here, but you can break down the people who are in employment and those who are self-employed. So we can look at the people already in the population, but we do not ask that question when they come in.
Baroness Browning: Just for clarity, what about non-EU immigrants coming and saying that they are setting up their own businesses?
Paul Vickers: Again, we can get that from the labour force survey.
Baroness Browning: It is in there, is it? Okay, thank you.
Lord O'Neill of Clackmannan: Given that the Romanians and the Bulgarians are among the poorest people who are coming and among the greatest number who do not have jobs to go to, how confident are you that the means whereby they arrive in the UK are by plane? I know that many buses come from Poland into London, but I could not say with confidence that the same prevailed for south-eastern Europe. If that were the case, could your stats be underestimating the numbers, or is there a built-in compensation factor?
Ian Cope: I will start and then hand it over to Paul to give more detail.
Lord O'Neill of Clackmannan: There are a number of qualifications.
Ian Cope: It is one thing that we have looked at a number of times. We do not look just at airports. We might have talked about airports, but we also measure by train—Eurostar—and by Dover and other ports where people might be coming in by coach. We are looking to cover all those routes.
Paul Vickers: That is exactly what I was going to say. We do not cover just airports; we cover the seaports and the Channel Tunnel.
Lord O'Neill of Clackmannan: Do you have any information on whether people come in by bus or by plane from the two poorest areas, where people have no job prospects at present and probably very limited financial resources anyway?
Ian Cope: I am sure we could get some information just from the IPS on the ports that people are coming in from. I am afraid that we do not have that with us, so if you would like that information we can send it to the Committee afterwards.
Lord O'Neill of Clackmannan: If it is not statistically significant and it would be at disproportionate cost, obviously it would not be too helpful, but it might be useful to get a general note, because not all migrants are the same; rich Austrians are a wee bit different from poor Romanians in terms of the—
Paul Vickers: —in terms of the sample and the way we do it. The report that we produced on the IPS policy that came out on 1 December will cover many of those questions, so we might forward that to the Committee if there is anything further in it. We are more than happy to send that.
The Chairman: Lord Jay had a couple more points to make.
Lord Jay of Ewelme: We talk about flows. I have a question about stocks. I think you said that the figures are rather less reliable on stocks than on flow, but can you say any more about the geographical distribution across the UK of the existing stock of EU migrants?
Paul Vickers: Yes. The information in table 4 on page 22 talks about the household population and the number of migrants across the country by region. Not surprisingly, most migrants, both EU and non-EU, live in London; 11.8% of London’s population comprise EU citizens.
Lord Jay of Ewelme: 11.8%?
Paul Vickers: Yes. That compares with the north-east, where only 1.6% are EU citizens. So we can break down that side of things. The table before you shows the distribution. Most people are in London and fewer people are in the north-east.
Lord Jay of Ewelme: So if you add London and the south-east, you will get up to 20%, roughly, will you?
Paul Vickers: Yes, you do.
Lord Jay of Ewelme: Okay, thank you. I have one other question, which is about what we know about EU migrants arriving into the UK who are coming neither for work-related reasons nor to study.
Paul Vickers: Again, table 1 on page 15 shows that breakdown. There are 24,000 EU citizens coming in to accompany and join relatives, and 17,000 people coming in for other reasons, which can be very varied; some people come in for religious pilgrimages, medical treatment, wanting to learn English. There are a whole range of different reasons that people put down. They are included in that other category.
Lord Jay of Ewelme: It is quite a lot, is it not? Well, it is 6%.
Paul Vickers: Yes, it is 6%. The majority are coming in for work.
Lord Jay of Ewelme: That is very helpful. Thank you.
Q55 Lord Condon: Could we turn to emigration briefly? What statistics do we have on the number of UK citizens emigrating to the EU? How many of them are there? Where do they go to? What do they do when they get there? Will the new exit checks make any difference to the quality and reliability of that data?
Ian Cope: Measuring people when they leave, where they are going and what they will do is harder than measuring people coming in, because they may be less certain. Paul, do you want to handle the first part of the question?
Paul Vickers: We measure them going out, and we ask them their nationality. In 2015, 124,000 UK citizens overall emigrated to the rest of the world. Of those, 35,000 went to Europe. The most common countries that they went to were France and Spain. We currently have no breakdown of what they do when they get to the EU, but we looking at different sources, such as UN sources or the European labour force survey. We are doing an assessment of those at the moment, which should be able to shed more light on that.
Ian Cope: We are looking at the labour force survey results from other countries.
Paul Vickers: Yes.
Lord Condon: You have no sense of whether they are lifestyle retirees moving to France and Spain, people going to work temporarily, or—
Paul Vickers: There will be some of that. My sense is that most of the people going are going for work purposes rather than retirement, but a certain element will be retiring.
Ian Cope: On the second part of your question, exit checks apply to people who are coming in and out of the country under visa arrangements. Obviously UK citizens leaving the UK will not be subject to visas, and exit checks will not tell us anything about UK citizens moving to the rest of the EU.
Lord Condon: I am not asking you to make a political judgment, but in terms of reciprocity and so on as we negotiate, would it be more helpful if we had a bit more information on what UK citizens who are going to the EU are doing? We seem to have a lot of information coming the other way but very little about our own citizens going to the EU.
Ian Cope: In general, the Office for National Statistics’ approach is to be helpful and to respond to new and emerging policy requests. Clearly, as I said earlier, the debate and the interest have shifted, so that is obviously something that we can look at.
In terms of any new system that is put in place, it is important that the statistical tail does not wag the government policy dog, but equally it is important that any new system takes account of measurement and the statistical need to produce statistics as a result.
Q56 Baroness Massey of Darwen: Could I return to the issue of students? What are the pros and cons of including students in the net migration figures? Would it be correct to infer that this debate is less relevant to EU migration, because the vast majority—72%, I am told—of those who arrive in the UK to study are non-EU citizens?
Ian Cope: On the first part of that question, long-term international migration, as I explained at the beginning, is defined as people coming in to the country for 12 months or more. Obviously students here at universities et cetera will be here for more than 12 months and should be included in the long-term international migration numbers. If they are studying in the UK for less than 12 months, obviously we do not include them in that number. Long-term international migration is defined in a way that includes, and should include, students here for more than 12 months, but as you rightly say in the second part of your question, the majority of students coming in to the UK for study—higher education in particular—are non-EU students.
Lord Jay of Ewelme: But 13% of the EU citizens who come here come to study, so it is a reasonably significant proportion.
Ian Cope: Yes.
Baroness Massey of Darwen: What about people who come to language schools to study? Do we have records of them?
Ian Cope: They are potentially included in the international passenger survey; it depends how long they will be here for. Typically, however, language courses are less than a year.
Baroness Massey of Darwen: Yes, they are.
Paul Vickers: Those who stay for longer get picked up, and they will be included in the study element of the international passenger survey. In addition, we have information from the Home Office and its visa data, which we publish alongside the IPS every quarter in the migration statistics quarterly report.
Q57 Lord O'Neill of Clackmannan: Your very helpful information is a wee bit bald in one or two areas. Could you talk us through page 18, table 2? Could you explain a bit more about the growth in employment for EU nationals and how it compares to growth in employment for UK and non-EU nationals?
Paul Vickers: I can. It is important to explain before I go into that that the information presented in table 2 is the number of people who are in employment at a particular point in time. If you look at the change over time, you have to realise that there are lots of flows in and out of employment. This just looks at the stock of information and the difference in the two stocks.
Ian Cope: So people may retire or become unemployed and come out of the labour force, while other people join the labour force. This looks at the net difference. You are just talking about the numbers.
Paul Vickers: You see the difference in employment between July and September 2016 and a year before: employment has gone up by 454,000. Of that increase, 213,000—just less than 50%—are due to an increase in British nationals, while 221,005 of that increase are EU nationals. The remaining small element of the increase is the non-EU nationals. So there is a bigger increase in the number of EU nationals in the labour market compared with a year ago.
Lord O'Neill of Clackmannan: The next table shows EU nationals in employment by industry. We were a wee bit concerned. Are these headings internationally agreed, or are there UK-specific ones for various areas of activity?
Ian Cope: As the footnote says, the industrial breakdown is from the standard industrial classification. So it is a UK classification but it is harmonised with the EU classification, because that is important when you are making international classifications. One of the things that with hindsight we could have caveated in this information is that it is based on the annual population survey and the labour force survey, which is a survey of household members. It does not cover what we call communal establishments—managed accommodation such as student halls of residence or hostels—so it may miss agricultural workers, for instance, who are living in a farm hostel. With hindsight, we could have made a bit clearer for you in our note.
Lord O'Neill of Clackmannan: From what we could see, manufacturing and a combination of food services are sourcing more than one in 10 workers from the EU, and it is only in those areas that that is the case. However, we have also had information from the NFU suggesting that it relies quite heavily on labour from the EU. Could that be accounted for by the discrepancy that you have just alluded to?
Ian Cope: It could be partly that. Particularly in agriculture, a lot of seasonal workers come in for a short period and then leave. By their nature, they are less likely to be picked up by a household survey. A combination of those two factors might explain the difference.
Lord O'Neill of Clackmannan: Even allowing for that fact that some of the agricultural workers who are engaged in fruit picking might be travelling around the country, they would still slip through a succession of nets?
Ian Cope: That is possible.
Paul Vickers: They are likely to, because they will be living in what we term communal establishments—hostels or farm accommodation. This goes back to the point about our wanting to find out more about these people, which is why we are keen to get access to administrative data—HMRC data, for instance, since people will still be paying national insurance contributions even if they are doing so on a short-term basis that will not get picked up by this table. Getting access to this kind of data will help us to answer some of these questions in more detail.
Lord O'Neill of Clackmannan: Has this never been the subject of a discussion between you and the NFU, which provides what seem to be statistics that diverge from yours?
Ian Cope: I am not aware of any conversation, no.
Baroness Massey of Darwen: At the very bottom of page 19, table 3 shows a high figure—90.8%—for “other services”. What do you mean by that? I am also quite surprised that human health and social work activities are at 90.3%, but education is 92.5%. I was always under the impression that health and social work activities were very high. Am I wrong about that?
Paul Vickers: That 92.5% figure is the number of UK citizens in employment in those sectors. So 90.3% of the people working in human health and social work activities are from the UK, but nearly 5% are therefore from the EU and a further 5% are non-EU citizens. There are more people from outside the UK working in health, which is what I think you were saying, than in education. So the table supports what you are saying.
Baroness Massey of Darwen: And the other services?
Paul Vickers: That is just a catch-all for anything that is not included in the above.
Baroness Massey of Darwen: It is pretty high.
Paul Vickers: You have to remember that that is the proportion of people working in that sector. Some 1.6 million people work in that sector out of nearly 28 million people overall, so it is quite a small proportion of the nation.
Baroness Massey of Darwen: Got it. Thank you.
The Chairman: You did say that these classifications are made by the Government and are reconciled with the EU?
Ian Cope: Yes, this is the UK’s standard industrial classification. It says that in the footnote at the bottom. We have classifications by industry and occupation.
Lord Jay of Ewelme: Do you keep statistics yourself on any other sectoral classifications than this one, or is this the only one that you keep? There are some slight oddities such as the fifth classification, which combines wholesale retail and the repair of motor vehicles. That seems to be a slightly odd combination.
Ian Cope: This is the top layer of the standard industrial classification. It is broken down into something like 4,000 different industries. For the census, which we carry out every 10 years, we can produce much more detailed breakdowns of industry. This just gives you the highest-level classification.
The Chairman: But you have others available?
Ian Cope: Yes. We can break down agriculture, forestry and fishing into those individual components for the census. As you get into surveys, the smaller sample sizes mean you can break things down less.
Lord Jay of Ewelme: Would it be easy for you to send us a breakdown of the wholesale retail and repair of motor vehicles?
Ian Cope: We can certainly take that away and come back with what we can.
Lord Watts: The Committee has heard a lot of statistics from different organisations, and what we are looking for from you to pull them together. It is interesting that, as you described it, you are looking forward to getting information from some of the other departments. Is that information being transferred to you now? You sound as though it is in the future rather than something that is happening now. I am a bit surprised, given what is going on, that that information has not flowed to you so far.
Ian Cope: As I said at the beginning, there are various legal constraints that make it very difficult for government departments to share those data with us, which is why we are dependent. We have been working with the Cabinet Office through its open policy process for the last two and a half years, leading up to what is in the Digital Economy Bill. That has the power to transform the richness and depth of statistical analysis that the Office for National Statistics can do. As Paul said, though, we have seconded ONS staff to work in other government departments so that we can analyse their data in their secure research environments, for instance. We have done that with the Home Office.
Paul Vickers: We have arrangements in place with the Home Office and HMRC.
Lord Watts: So the labour requirements are now removed and you can get access to all that information?
Ian Cope: No, that is dependent on Parliament passing those particular clauses in the Digital Economy Bill.
Lord Watts: Could you give us a timeline for when you think you will have that information?
Ian Cope: If the Bill is passed by Parliament, which we hope will be some time this year—although obviously we cannot pre-empt your decisions—that will open the doors to getting access to getting a lot of this data, and we might be able to start getting that towards the end of next year.
The Chairman: That leads nicely to Lord Ribeiro’s question on the extra tools and resources.
Q58 Lord Ribeiro: What extra tools and resources will be needed by you? You have already described your situation as hinging on the digital Bill in order to access some of this, but what resources do you think you will need from both the Home Office and the Government to introduce a more restrictive immigration regime for EU citizens?
Ian Cope: Ultimately, any new migration regime is a decision for the Government. We would want to be party to some of those conversations in order to ensure that the statistics could be collected and that we could continue to produce them. As I said earlier, the real prize is being able to access the administrative data so that we can do this record-level linkage, linking individuals in a very secure environment inside the ONS. I think we are a trusted organisation. We have been protecting census records for over 200 years, for instance, and I think we are widely trusted. The Statistics and Registration Service Act has very strict provisions about what we are and are not able to do, and there are penalties; you can go to jail for up to two years if you do something that you are not supposed to with data. So access to data is the big prize here.
Lord Ribeiro: It has been said that the IPS is not fit for purpose. Is that one of the ones that you might want to ditch if you had access to the extra data?
Ian Cope: As I said at the beginning, we believe that it is fit for purpose for measuring at national level but it cannot answer all the questions, such as many of those of you are asking today. We would want to continue with the international passenger survey. It is well regarded internationally and gives the high-level flows that you would probably still want to be able to capture, partly to benchmark your administrative data to.
Paul Vickers: I think it is fit for purpose at the national level, as Ian said. Sometimes administrative sources can give you misleading data on those flows. Certainly other countries struggle with some of their administrative sources to get reliable net migration figures.
Lord Ribeiro: The Government have consistently said that they want to get immigration down to the tens of thousands as opposed to hundreds of thousands, but the chart on page 10 that you showed us at the beginning was quite clear that we have not got below the 100,000 figure for quite some time. From the 1970s to 1995, net migration was below 100,000, but since then it has been well above. Mindful of the new accession states in the EU, and mindful of the need still to have a workforce to keep several service industries going, what is your own assessment of any Government’s ability to get these figures down?
Ian Cope: I cannot answer that question. It is a political decision for the Government. It is for the Government to set their target. It is our job to measure the statistics as best we can.
Lord Ribeiro: Yes, but looking at the trend, looking at the figures as they are here, peaking as they do at about 300,000 in terms of net migration, from all the statistical information that you have in your hands, how long would it take to reduce that figure to where we were before?
Ian Cope: I cannot answer that question. It depends on what the Government do in response to the migration regime. That is a political judgment and a political decision. Our job is to try to measure the numbers as best we can and to provide the information to inform parliamentarians and policymakers.
Q59 The Chairman: Can I move on to the question on overall migration? Non-EU migration, which the Government control, has been consistently higher than EU migration. Why is that the case, and what do the statistics tell us about that?
Paul Vickers: I will try to unpack that. The most helpful diagram to help you with that is figure 3 on page 12 of the evidence that we provided. The blue line and the amber line show what is happening as a comparison between EU migration and non-EU migration. That shows that 10 years ago there was a consistent gap between those two of 100,000, where between 2006 and 2011 non-EU migration was consistently above EU migration. Since 2012, that gap has narrowed so that they are now virtually the same. Part of the reason is that we have had the increase in EU15 and EU2 migration that I was describing earlier, which has given us an increase in EU migration, but restrictions have been placed on non-EU migration, particularly on the number of study visas for people English-language colleges and further education colleges. That is what caused the fall between 2011 and 2013 in non-EU immigration. You are right that non-EU migration is controlled by visas. Table 1 on page 15 shows that most of the people coming in are still coming in for study—113,000 is the latest figure for non-EU citizens. We are still getting 74,000 coming in for work. They tend to use the skills visa, so filling up shortages of skills in the UK.
Lord Jay of Ewelme: Looking at the figures for EU migration from 2006 to today on your helpful chart, is there a correlation between them and the performance of the British economy, going down during the years of the financial crisis and then going up as that came to an end, particularly with regard, as you were saying earlier, to EU15 migration going up again as people come back to the City? If you charted that alongside the performance of the British economy, would there be a correlation?
Paul Vickers: It is not just the performance of the British economy but its performance relative to other economies in the European system.
Lord Jay of Ewelme: Indeed, but is there a correlation?
Paul Vickers: We have done some work showing that that appears to be one of the drivers for the increase.
Lord Jay of Ewelme: So the better the economy does, the more people want to come here.
Paul Vickers: If the rest of the EU is not performing at the same level.
Lord O'Neill of Clackmannan: And do people leave when the economy is not doing so well?
Paul Vickers: In 2008, we saw a fall in the number of EU8 migrants coming in over that period. Figure 5 shows that from quarter 4 of 2007 there was a big fall in the number of EU8 migrants from just over 110,000 down to 65,000 in late 2009.
Ian Cope: I think it is generally recognised that there are push and pull factors in migration. If an economy is doing badly, that might encourage people to leave, and if it is doing strongly than it might encourage people to come. Figure 2 on page 11 goes back to the mid-1970s, and you can see that a lot of British citizens were emigrating at that time. We have not done those studies so this is speculation, but I do not think that is an unreasonable assumption.
Lord O'Neill of Clackmannan: It could also be the fact that with the negotiations on Brexit or the possibility of Brexit that Romanians and Bulgarians turned up in greater numbers in case anything prevented them from being let in later.
Ian Cope: We have no evidence on that, so I cannot speculate.
Lord O'Neill of Clackmannan: You really need to look at the next quarter, after June, to get that sort of figure.
Paul Vickers: At the moment, our figures do not show that. The flows into the country only go up to the end of June, so we are waiting for the next figures to come out in February and beyond.
Lord Watts: Sorry to pursue this again, but once you have these statistics, which you say you will hopefully have by this time next year, how long will it take you to do the analysis and bring that together before you can provide any detailed information on what it tells us?
Ian Cope: Our experience of other pieces of work, such as where we are doing work around administrative data census and bringing together data sources to measure the population to see if we can do that without a census, suggests that once you have the administrative data you have to understand its quality—are there “funnies” in it because, say, the regime has changed? There is a piece of work to link the data with our sources and to understand it, and it really depends on how that goes and what that shows us. We might need to ask for additional information or we might be able to produce something quite quickly. It will take time to understand the data, which is partly why we are seconding people to these departments already: so that we can get ahead of the curve. It is difficult to predict how long it will take though, until we get the data and can see what it shows and whether it is fit for purpose and links smoothly.
One of our problems in the UK is that we have a lot of registration numbers—national insurance numbers, visa numbers, passport numbers—so it is often a challenge to link these datasets together. In particular with the flows data, there are very large numbers of people coming in and going out. That is a long-winded way of saying I cannot give you a precise answer.
Paul Vickers: We are trying to be ambitious in what we are doing. On page 27 of the evidence that we submitted, we have set out what we hope to achieve in the next six months. There are some dependencies, as Ian was saying, such as if we can get access to the Home Office exit check data. We already have access to Home Office migrant journey data and HMRC data, but at the moment we are looking at those individually. That should give us some initial analysis, but the real power will come in at the end of this period. We have reports that we are hoping to release in the next six months, but the real power will come when we start to link it, which we can do only when the data comes to the ONS. When we do that, we will be able to get some interesting findings.
Ian Cope: I can promise that it will be transparent and that we will set out our plans as best as we know them. We are happy to keep the Committee informed, if you would like further information as we progress.
The Chairman: Thank you very much indeed. It would be helpful if you could keep us posted and if we could have a note on the limitations of the current data that you have and what you would like to see. It would also be helpful if you could refine this evidence, in the light of our questions, to have better explanations. We will then be able to publish that with the report. Thank you for your time this morning.