MP for Windsor
Working Hard For You
Migration data speech

Adam Afriyie (Windsor) (Con): Thank you for calling me to speak, Chairman Walker—if that does not sound too much like Chairman Mao. It is a pleasure to speak in this debate.

First, I say a huge thank you to my hon. Friend the Member for Harwich and North Essex (Mr Jenkin) for the presentation of the report by the Public Administration Committee. Given that immigration is one of the top three issues in the country, it is absolutely vital that we have a look at the underpinning foundation of the statistics and data that inform the debate about it.

I also thank the Minister, before he even makes his closing remarks, because I know that he is working exceptionally hard; he has worked exceptionally hard in all the jobs he has had. I am aware that we have the same agenda. We would like to see better statistics but sometimes in coalition it can be tricky to get these things through at the pace that he may wish for.

I will try to keep my remarks as brief as possible, but nothing I say today is to be taken as criticism; it is merely observation that there are better ways in which we can collect the statistics and more intelligent ways in which we can present them.

As a former shadow Minister for science with a degree in a social science—a semi-science—and as chairman of the Parliamentary Office of Science and Technology, I think that evidence is vital to any decision that we make in the public domain. Indeed, although I speak as a Back-Bench MP, I say as chairman of the Parliamentary Office of Science and Technology that its mission is to ensure that any parliamentary debate and policy formed on the Floor of the House is informed by the data and evidence, and that those are behind the decisions.

The data are so important in the immigration debate—that has largely been highlighted by the migration report—because it is a sensitive issue in the public domain. If we consider it in a political sense, a Government could stand or fall on the basis of their immigration policies and how the data are presented, and upcoming parties not currently in this place could succeed or fail on the basis of immigration data. Our national cultural-social cohesion could implode on the basis of a lack or misrepresentation of immigration data, and our future economic prosperity and relationship with the European Union could depend on the data and their presentation. It is an overwhelmingly important and sensitive issue.

One of my key concerns is that when I say “immigration”, people think it means different things. Some people think it is all about illegal immigrants and get upset about it, some think about genuine asylum seekers and others think that we need more immigration, saying, “Surely we need more, because we need more work visas for those important jobs in society that we are currently unable to fill here in the UK. Of course, those people go back.” Some think specifically about Romanians and Bulgarians because of the recent public debate and some think about investors, bringing millions and sometimes billions of pounds here to the UK, buying nice houses in the centre of London. Others may be thinking about visitors coming to join their families for a period who may return a year or two later, and some may—admittedly, mistakenly—think that tourists are part of the immigration debate.

The word “immigration” and immigration data are proxies for so many other things. That is why it is important that we have great clarity about the data we collect and what they mean. We must ensure that they are statistically significant.

Keith Vaz: I agree with everything that the hon. Gentleman has just said. There are too many myths about immigration. That is why we need official statistics that people can sign up to before we can even start a debate. We are not saying that there should not be a debate; the hon. Gentleman knows that, having attended many debates in the House about immigration. Issues have been raised with him as they have with me. Once those data are available, the big issues that concern the public can be tackled.

Adam Afriyie: That is absolutely right. Clear, accurate and granular information, data and statistics will enable groups with a view on each category of immigration to take a reasoned view.

I often think that politicians’ use of statistics—I confess that this may include me in my early days—is like a drunk’s leaning on a lamp post, less for illumination and more for support. I do not mean to criticise the ONS or even the passenger survey, which is doing what it is told to do in the best way it can, but the danger of the Government’s or any politician’s leaning on the immigration data and statistics is that they are weak and will just fall over. Yet the public animosity and disharmony that can be created by the misuse or misrepresentation of the data are all too well known.

Mr Andrew Turner: Does my hon. Friend agree that the £2 million, which my hon. Friend the Member for Harwich and North Essex said the Government are not willing to spend, is a tiny amount when it is considered that the same sum is proposed to be spent on a quarter of a mile of A road in my constituency? Why cannot the Home Office find the £2 million?

Adam Afriyie: The Home Office is doing difficult work in difficult circumstances of coalition. I agree with my hon. Friend. It would seem that spending £1 million or £2 million—or even £5 million or £10 million—to deal with such a vital issue at the heart of a current national debate, which could unsettle an entire nation, is a small sum, if that is what is required to put this matter right.

I suspect that only small sums and adaptations in how we use existing data and how we conduct the passenger surveys would be needed, and that those would assist enormously, in addition to the exit checks.

If we are to plan our public services, we need to have a good idea about what the immigration statistics and data are. It is interesting that the ONS said that the data at the moment “should not be used as a proxy for flows of foreign migrants into the UK”.

The Oxford Migration Observatory stated: “sampling errors are too large to measure with a reasonable degree of accuracy the number of migrants to a single region. Within the UK, or from a single country of origin”.

Yet if we listen to the public debate, including in my constituency, assumptions are already being made about particular areas and the effect of immigration. I have to admit that sometimes assumptions are presented by Departments, senior politicians and political leaders on the level of Romanian and Bulgarian immigration, for example, although the data just are not there to justify the statements.

My hon. Friend the Member for Rochester and Strood (Mark Reckless) made it clear on “Newsnight” a few weeks ago that, when looking at current data collection, a handful of Romanians or Bulgarians—four, five or six—making a certain statement could lead to a difference of 4,000 in our estimate of the number of Romanians or Bulgarians coming into the country. It is clear that the data are currently insufficient to draw conclusions or create policies from.

The current data are vague and self-selecting. People who go to another country wanting to stay there, knowing that there were no exit checks and that they could probably get away with it—not that I would do this—would, if they were desperate, answer appropriately to a question in a passenger survey about how long they intended to stay, to ensure that it looked okay. There is a lot of self-selection in who answers the survey and there will clearly be, if we are all reasonable human beings, an understanding that people will answer questions to serve their purpose, although I would hope that everyone is honest.

Although we want to get immigration levels down to tens of thousands, rather than hundreds of thousands, with the data and statistics that have been available for the past four years our current estimates could be 200,000 or 250,000, one way or the other: these numbers are enormous and the statistical significance of the data really needs to be examined and reined in as soon as possible.

Keith Vaz: Is the hon. Gentleman telling the House that, because of the problem with statistics, the Government will not meet its target or that we should not have a target because of the problem with statistics?

Adam Afriyie: A very apposite question. No, I am saying that we do not have any idea about whether these targets may be met. If the immigration figures came in at 30,000 or 40,000 a year, it could be argued by one side that this goal had been magnificently achieved. It could also be argued that the goal had been achieved if the figures came in at 200,000 or 150,000. Conversely, political opponents may argue that, given the statistical significance of these data, the goal has definitely not been achieved. Above all, in the absence of accurate, robust data, it is impossible to have a sensible public discourse.

Again, if we had the granularity of data that I and most hon. Members would like to see on the various categories of migrant in and out of Britain, we could have a robust argument about each category of immigration. Perhaps, in years to come it might not be necessary to have a blanket limit or target; we could take a more granular approach. For example, I would be in no way unhappy if we had an extra several hundred thousand students studying in our universities, paying tuition fees, helping to fund our great institutions and spreading around the world a deep well of good will on British culture, British language and the British way of being. That would be fantastic, because the evidence is fairly clear that the majority will return to their countries of origin and spread the message that Britain is a great place. That could enormously enhance our standing in the world and our economic performance. If we can get down to the granularity of debate, we will come to better policy solutions. In getting to the granularity of debate, I commend the Committee’s report, which makes a huge step in the right direction by observing that we need more granular data.

What is the solution? I do not purport to have all the solutions, but I put forward a few suggestions on data collection, based on the report, my experience and what is possible with technology. First, a little more data need to be collected in the international passenger survey, possibly as a short-term measure, until we resolve the whole situation. Having had a chat with the Minister, I accept that we may well need more data collected by the international passenger survey in the short term, but the location in which the data are collected might be relevant in coming up with better numbers.

It is also important that exit checks come in sooner rather than later. As Conservatives, we would have loved to have seen them come in very early in this Parliament, but sadly, in coalition, other priorities often get in the way. It may well be that there are elements among our current political friends, but usual foes, who believe in completely liberal border control, where people can move around without any checks at all. I recognise that that might be an element in the challenge of bringing the exit checks in sooner rather than later, but it is perfectly achievable by 2015.

When we talk about exit checks, we need to be mindful that someone would not necessarily have to queue and answer questions to exit the country. Indeed, when I visit many African and middle eastern countries, I have virtually no conversation at all. I simply walk into the country, they scan my passport, frown at me and ask one question. Then, as I leave the country, I hand the passport over, they scan it and say, “Have a nice day, sir”, and that is it. With technology, the idea that there would need to be intrusive surveys and so on is not necessarily right.

We also need to bear in mind that the airlines and travel companies hold an enormous amount of electronic data, which raises the question of why we do not use those data. We type in all the details on easyJet when we fly abroad, as do others when they travel in from overseas. Why are those data not used—not all of them, but a reasonably relevant or statistically significant sample—to check people when they are coming in and going out? I am sure that parts of the data may be used for certain purposes, which the Minister cannot discuss. It would be such an easy win, however, to open up access to those data, which people are freely providing when they travel, to get a better grip and understanding of the various types of migration and whether people are leaving the country.

Finally, I know that there is a will among Conservative Members of Parliament, many elements of the Labour party and the Liberal Democrats in Parliament that we get the issue resolved. I urge the Minister, for the sake of national harmony and a sensible national debate in the run-up to the 2015 election—when people must make their minds up about all sorts of things, including, hopefully, in the not too distant future, whether they want the UK to remain in the EU—to put some further measures in place, so that people are clear on the various categories of immigration data before that election. We would then not need broad, sweeping statements about immigrants in general; we would have a precise and targeted understanding of each of the groups that we approve or, in some cases, disapprove of, and the debate can become more rational.

I urge the Minister to take another look at the report and to bring forward more speedily some changes or suggested changes for the IPS, the ONS and the speeding up of exit checks from the United Kingdom. We can be a happy nation. The British people deserve better immigration data, particularly given that they will not be that expensive to collect.


Re-introduction of universal exit checks

Adam Afriyie (Windsor, Conservative): To ask the Secretary of State for the Home Department what recent progress she has made in introducing universal exit checks; and if she will make a statement.

Karen Bradley (The Parliamentary Under-Secretary of State for the Home Department): The Government is committed to reintroducing exit checks. By April 2015, comprehensive exit checks will apply on scheduled and commercial air, sea and rail routes.

We have recently introduced new powers in the Immigration Act 2014 to support embarkation checks at the border, and we continue to work with carriers and port operators to explore the least burdensome way of delivering the exit checks commitment.


Last year’s migration statistics could be wrong by as much as 50,000

Are Britain’s migration statistics accurate? Certainly not as accurate as I might like. That’s why yesterday I wrote an article for the Huffington Post urging the Home Secretary to reintroduce entrance and exit checks on UK borders as a matter of urgency – to count people in and out of the country.

If we’re going to write effective immigration policy that deters health and benefit tourism and attracts investment, we must have a detailed statistical picture of the number and category of people entering and leaving the country.

Read the article on the Huffington Post


Accuracy of immigration statistics

Adam Afriyie (Windsor, Conservative): To ask the Minister for the Cabinet Office what assessment he has made of the accuracy of the International Passenger Survey for estimating migration flows; and if he will make a statement.

Nick Hurd (The Parliamentary Secretary, Cabinet Office; Ruislip, Northwood and Pinner, Conservative): The information requested falls within the responsibility of the UK Statistics Authority. I have asked the authority to reply.

Letter from Glen Watson

As Director General for the Office for National Statistics, I have been asked to reply to your recent Parliamentary Question asking the Secretary of State for the Home Department, what assessment she has made of the accuracy of the International Passenger Survey for estimating migration flows; and if she will make a statement.

ONS publishes estimates of long-term immigration, emigration and net migration each quarter. These are largely derived from the International Passenger Survey (IPS). The estimates are published alongside their margin of error which refers to the 95 per cent confidence interval, and is a measure of the uncertainty associated with making inferences from a sample.

The latest IPS estimate for long-term immigration for the year ending December 2013 was 485,000, with a margin of error of +/- 29,000. The latest IPS estimate for long-term emigration for the year ending December 2013 was 295,000, with a margin of error of +/- 19,000.

ONS has published an overview of the quality and reliability of the International Passenger Survey (IPS) in relation to producing estimates of long-term international migration flows, that is, flows of migrants intending to remain in or out of the UK for twelve months or more. This can be accessed at:

Furthermore, ONS has recently published a review into the ‘Quality of Long-Term International Migration Estimates from 2001 to 2011’. This review found that a substantial amount of immigration, particularly of EU8 citizens, between 2004 and 2008 was missed by the IPS, prior to improvements to the design and coverage of the survey in 2009. Revised net migration estimates, which are consistent with the results of the 2011 Census, were published as part of the review. The review can be found at:

The improvements to the IPS have reduced the relative error around the estimates, as well as the balance of the sample between EU and non-EU migrants.


We still have no idea how many Romanians and Bulgarians are in the UK

Over recent months press speculation about how many Eastern European migrants are now living in the UK has run riot. Huge figures have been used to back up wild stories and tiny figures have been used to strike back.

But the fact of the matter is, right now, we just don’t have an accurate count of migrants from any country in the European Union; and recently published figures on Romanian and Bulgarian migrants are, ironically, an excellent example of our lack of hard numbers.

Last week it was reported that the number of Romanian and Bulgarian-born people working in the UK had dropped from 144,000 to 140,000 between December and March, defying general predictions of a flood of migrants from Eastern Europe.

These results are based on the Labour Force Survey, carried out by the Office of National Statistics. While these surveys are great for uncovering general working habits, they’re not so good for telling us about small groups of people, like Romanians and Bulgarians. That’s because the number of Romanians and Bulgarians who take part in these surveys is absolutely tiny.

Fewer than 200 Bulgarians and Romanians took part in the survey, meaning that the drop in 4,000 people working in the UK is based on the finding that four of those people had left the country. Just four people – that’s the evidence for the drop in the number of Bulgarians and Romanians in the UK.

As my Conservative colleague Mark Reckless pointed out: “This survey cannot possibly tell us how many Romanians and Bulgarians started working in the UK during the first quarter of this year.” In fact, the ONS warned us similarly, saying that their data “should not be used as a proxy for flows of foreign migrants into the UK”.

This is not to say that journalists and the media are strictly to blame. Journalists can only report on the available figures, and that’s exactly what we lack.

So what does this story tell us? Rather than reveal much about the number of migrants coming to the UK, it tells us a lot more about the sorry state of the UK’s data.

We need to calculate migratory flows more accurately (see last week’s article in The Spectator). This will involve universal entry and exit checks. Only once these are in place can we know the true number of migrants entering and leaving the country.

Both sides of the immigration debate must accept there are severe shortcomings in the way we quantify migration. The Government should find a way of collecting reliable migration data and resist any measures from the EU that prevent us from doing so. If we want answers and full, accurate picture, we need to have proper border checks.