Migration: The needle in the data haystack

With migration continuing to make headlines in the media, we unpack what actually defines a "migrant", and how the UK's largest producer of official statistics goes about counting them. We also shed light on the misuse of migration figures.

Transcript

MILES FLETCHER

Hello and a very warm welcome to a new series of Statistically Speaking - the official podcast of the UK Office for National Statistics. This is where we hear from the people producing the nation's most important numbers, about how they do it and what the statistics are saying.

Now it's hard to think of one statistic that could be said to have been more influential these past few years than net international migration. Suffice to say it's the one ONS statistic that probably draws more media attention than any other.

But to fully understand the migration figures, and the swirling debate around them, we'd say it pays to know a little about how they are put together. And the first thing you need to know about that is what, or who, is a migrant in the first place. As usual, to unpack and explain the migration statistics we have the top experts from the ONS and beyond. Mary Gregory is director of population statistics here at the ONS. Madeleine Sumption is director of the Migration Observatory at the University of Oxford, and new chair of the National Statistician's Advisory Panel on Migration. And to help us understand how the numbers are used and abused in public debate, we're also delighted to welcome Hannah Smith, senior political journalist at fact checking charity Full Fact. Welcome to you all.

Madeleine, to start with you if I may, with that fundamental question, quite simply, what is a migrant? MADELEINE SUMPTION

Well, there are actually lots of different definitions of a migrant and we use different definitions at different points in time. The standard definition of migration that we use in this country is long term migration, so people moving for at least 12 months, and so the Office for National Statistics figures on immigration, emigration and net migration are all using that definition. And people in that data, they're migrants regardless of whether they are British or not British. So you could have a British person who's gone to live overseas for a few years and coming back they would be counted, in theory, at least in the data, as a migrant. There are other definitions though that are very useful for policy. So sometimes people talk about migrants, meaning people who don't have British citizenship, and the value of that is that these are people who are subject to immigration control, effectively that the Home Office is regulating their status. But it's also sometimes quite useful not to look at whether someone's a citizen now, because of course people can change their citizenship, and many migrants to the UK do become British citizens. So it can be useful to look at whether someone has migrated in the past. The standard definition for that is whether someone is born abroad. But now we've got all these exciting new data sets from administrative data, and so there's a new definition that's creeping in and being used a lot, which is someone who was a non-citizen at the time they registered for their National Insurance Number, regardless of whether they've subsequently become a British citizen. So it's a bit confusing sometimes for the external user, because for various reasons, we have to have all of these different definitions. You just have to know which one you're looking at at any point in time. MILES FLETCHER

But the basic headline definition, as far as the ONS is concerned -and I guess internationally too because it's important that these figures are comparable- is that it is a person traveling from one country to another for a period of 12 months.

MADELEINE SUMPTION

That's right

MILES FLETCHER

And I guess that is something that is perhaps not widely understood. People understand that migration has a degree of permanence, so they move from one country to another, and yet you can be a migrant in quite a sort of transient way. MADELEINE SUMPTION

That's right, we have short term migrants as well. So we have a lot of people who come to this country to do seasonal work. For example, they spend up to six months in the country. Then you have people who are long term migrants by the ONS definition and they may spend two to three years here, for example, if they're a worker or an international student. So you're right. I think in people's minds, often when they think about who is a migrant and who comes to mind, they will typically think of someone who is moving permanently. But actually a lot of migrants to the UK only stay for a couple of years. MILES FLETCHER

And none of these people, when it comes to measuring them, none of these people arrive Paddington Bear style with labels around their necks saying "I am a migrant". The ONS in measuring migration has to classify whether these people qualify or not. MADELEINE SUMPTION

That's true, and that is very tricky. And it's something I think the non-technical user of the statistics finds it difficult to appreciate quite how hard it is for ONS to work out who is a migrant or not. Because we have millions of people crossing our borders every year, most of them not migrants. We have tourists or people who come in to visit family members. There are all sorts of people and reasons why those people come and go, so ONS is really looking for the needle in the haystack, and a relatively small share of people who are crossing the borders are actually migrating. MILES FLETCHER

Well, that seems a good moment to bring in the person who is in charge of finding that needle in the haystack statistically. Mary, tell us how we approach this task. Perhaps start off by explaining how we used to do it.

MARY GREGORY

Previously we used a survey called the International Passenger Survey, and there we would ask a sample of people as they came into the country, or as they left the country, what their intentions were, and we'd be able to provide very early estimates based on people's intentions to stay or to leave. MILES FLETCHER

This is people at airports and other ports of entry, ferry ports, for example, simply approaching people as they wander along the corridors, almost in a random sort of way. MARY GREGORY

Exactly that, you might have seen them. If you have travelled through an airport you may have seen a desk that sometimes says Office for National Statistics. And there would sometimes be people there with very carefully scheduled timetables to make sure that we collect a good cross section of people. MILES FLETCHER

So the International Passenger Survey is essentially a big sample survey. Nothing wrong with that, and yet, the number of people being stopped at airports who did actually identify themselves as being migrants was quite small, and that made for some very broad-brush estimates didn't it? MARY GREGORY

Yes, as you can imagine, people travel through airports or ports for many different reasons, and a lot of those people traveling will be traveling for a holiday or business or to visit family. And so the proportion of those people who are actually going to become residents or leaving for more than 12 months is very small, which makes it really difficult to pick up a good sample of those people. MILES FLETCHER

And because it's fair to say the International Passenger Survey was never set up to measure migration in the first place, and that was something ONS found itself pointing out for a long, long time before things actually changed.

MARY GREGORY

For a number of years we made clear that it was being stretched beyond its original purpose, and that it was the best measure we had at the time but now we think we can do better. And I think one other really important aspect of that is understanding that the survey was asking about people's intentions, and intentions don't always match reality of what we then see. MILES FLETCHER

Because you might arrive as a student, end up working, settling, starting a family... MARY GREGORY

Yes. Or you might find that you've arrived planning to stay for a year and then change your mind and you've left again. So it could go in either direction. MILES FLETCHER

So the case for change was strong. What has changed? How is migration measured now? MARY GREGORY

So now we have a variety of different ways to measure depending on the nationality of the people arriving. So for anybody from outside the EU, we have good data around visas from the Home Office, so we can use that to understand who is coming and what their reasons for travel are, and we can come on to that a little bit later. For people within the EU, that was a bit more difficult because prior to exiting the EU nobody needed a visa. And so at the moment, we use administrative data, so that's data collected for other purposes, and we use data from DWP, so the Department for Work and Pensions, to understand who has come into the country and who is staying in the country for more than 12 months. And for British nationals, we still, at the moment, use the International Passenger Survey, but we hope to change that very soon. MILES FLETCHER

And essentially, the last use of the IPS, as far as migration is concerned, is to capture British passport holders leaving the country because nobody else is counting them out. MARY GREGORY

That's right, and it's actually just stopped collecting that data. So we will move to the new methods very soon. MILES FLETCHER

Okay, so how successful would you say this shift has been? MARY GREGORY

I think we've definitely improved the data we can provide. It's a better reflection of people's behaviours. We know that because we've compared the different methods and looked against the census and how the population has changed there. But there are also other advantages as well. So we can now look a lot more at why people have come to the UK, or which are the people who are leaving, so we know more about the reasons for migration as well. MILES FLETCHER

Madeleine, you run what's recognized as one of the leading think tanks in this area. How much of an improvement is the current system? MADELEINE SUMPTION

I think the data that we have, particularly on non EU citizens, is much better than it was in the past. Just to give an example, back in the early 2010s there was a big political debate about what the right level of net migration should be, and the government had a target of getting net migration down to under 100,000 from where it was. It was around 200 to 300 thousand at the time. So of course, the biggest question was, how do we do this? And the problem was that, based on those old data, we had no idea how many students were leaving the country. If someone came on a student visa we maybe caught them coming in but there were various problems. We just couldn't measure accurately enough the students going out. And so this most basic of questions, how can the government meet its net migration target, was not possible to answer with the data. Fast forward now ten years, and at least for non EU citizens, we now have pretty detailed data so we can say, okay, this number of people came in on student visas, this share of them left, that share is actually a bit lower than people were expecting. So those are quite interesting data. At the moment we can do the same for work visas, family members, refugees - so that's a dramatic improvement. There's still work to do I think on EU citizens. The ONS is measuring quite high levels of both EU immigration and emigration, of people who don't seem to be coming on visas and probably have a status from before Brexit, but we don't have a great sense of that. And as Mary mentioned, we currently don't really have any data on British citizens migrating, so that will need to be fixed. But yeah, I think the overall picture on immigration and emigration is much better than it was. Separately in the system, there are some challenges, let's say, with the surveys that give us data on the population of people in the country, their characteristics and so forth. And that, I think, has deteriorated a little but hopefully will come back on track. MILES FLETCHER

And bedding in the new system has brought about the need for some pretty big revisions. And that, of course, brings challenges doesn't it. Around confidence in the numbers when you have to revise by several hundred thousand the number of people that have been classified as migrants. And you get these sort of headlines about the ONS, you know, missing the population of Cambridge or wherever it was. But it wasn't a question of missing people as such, was it? It was just getting better data to understand which of the people coming and leaving should actually be classified as a migrant. MADELEINE SUMPTION

Yes. There were a number of issues there. There were a couple of cases where it was a case of missing people. There were some Ukrainians, for example, that got lost in the data. But that was a relatively small part of the overall revisions. Mostly, I think there's a challenge, and this challenge is not going to go away entirely but I think that the situation is improving, that when people's behaviour is changing the ONS still has to make assumptions about how long people are going to stay if they want to produce the data quickly. And so when you have a big policy change, you've got new groups of migrants coming in who don't necessarily behave, you know, leave and arrive after the same amounts of time as the previous groups of people who came in. Then you're more likely to have some revisions. And that's one of the things we've seen over the past few years. MILES FLETCHER

Let's trace the story of migration, if we can, just over the course of this century so far because it's been one of, if not the biggest, political stories. And you might argue, one of the factors that has determined the course of political events in this country. Obviously the ONS is not a political organization, but its figures do tend to have an enormous influence in that direction. Migration really became a big issue in that sense around about the early part of the 21st century when countries were joining the EU from the old Eastern Bloc. And suddenly there was a perception not only that there were large numbers of people arriving as a result of EU enlargement, but that the ONS was struggling to actually keep track of them as well. MADELEINE SUMPTION

Yeah. I mean, I would say that the increase in migration was even a few years earlier than that. The EU enlargement was one of the biggest events in migration in the last 30 years but there had already been a bit of an uptick in non EU migration, even from the late 1990s, and that is something that we saw across a number of countries. So the UK has been a major destination country, and is, if you look at comparable European countries, towards the top of the pack. But interestingly, we've seen some broadly similar trends in quite a lot of high income countries towards higher levels of migration. And that, of course, you know, as you've said, it's made migration much more salient in the political debate, and it's greatly increased the demand for accurate migration stats. And not just stats on the overall numbers, which of course are important, but really understanding who is coming to the UK. You know, what kinds of visas are they on? What do we know about their characteristics, their nationalities? How do they do when they get here...So I think that the demand for good migration statistics is just much higher than it was at a time when the UK experienced relatively limited migration. MILES FLETCHER

It's arguable that it was indeed rising EU migration that actually led to the events that led to Brexit. What has changed in terms of migration flows because of Brexit? MADELEINE SUMPTION

Well, the changes have been really big actually. I mean before Brexit quite a substantial share of all migration was from EU countries. After the referendum, even before the UK left the EU but after the referendum vote had been taken, there was already a decline in EU migration for a host of reasons including the exchange rate and so forth. So in some ways Brexit did what it was expected to do in reducing EU migration to the UK, because when free movement ended we saw quite a dramatic decrease in EU migration. And net migration from EU countries is now actually negative. So we've got the EU citizen population in the UK shrinking. But what was unexpected about Brexit was that then there was quite a big increase in non EU migration for various reasons. So partly policy liberalisations that at the time didn't necessarily look like a massive liberalization, but I think that the take up from migrants was much more enthusiastic than perhaps the government had expected. Lots of things came together. More international students, more workers, the war in Ukraine of course and lots of Ukrainians coming to the UK. And all of those came together at the same time and meant that we then ended up unexpectedly with these record high levels of net migration, peaking at just over 900,000 between 2022 and 2023. And now, of course, the numbers are coming right down again. So we had a record increase, we've then had a record decline to back to what are actually still pretty high levels of over 400,000. So we've really been on a roller coaster ride in terms of the migration patterns in the last few years. MILES FLETCHER

Yes, and statistically the contrast between what's happened recently is that these migrants have become much more conspicuous and much more measurable because they're being covered by visa data, whereas previously, the EU migrants in the early part of the century weren't actually picked up until the until the census in 2011 were they? MADELEINE SUMPTION

Yes and my hope is that because we're now measuring migration using visa data, when we get to the next census hopefully it will mean that those revisions - especially given that we'll have planned revisions over the next few years to the data - the hope is that that will mean we won't need such big revisions at the next census because we will have had a slightly more accurate measure between the censuses. MILES FLETCHER

And I guess the three elements in this recent wave of migration that have attracted particular attention, yes, people have come to work and people have come to study as previously, but in this latest wave, people were bringing more of their dependents with them weren't they? Perhaps because they were coming from further afield? MADELEINE SUMPTION

There's a bit of a puzzle about precisely why that increase in the number of dependents happened when I think it seems like there were probably two main factors. You've got international students bringing family members. We saw a shift in the countries as you mentioned, the countries that students were coming from. A lot more students from Nigeria, and they're more likely to come with their family members. We also saw a really big increase in the number of people coming to the UK as care workers after the government opened up a route for care workers. And so in one year alone, in 2023, there were visas issued to over 100,000 care workers, and they brought more than 100,000 family members with them, partners and children, that is. But that's now changing, because in response to these changes the government then introduced restrictions on the migration of family members, specifically of care workers and international students. So we've seen over the last year that fewer people are now bringing their family members with them. MILES FLETCHER

Interesting example of better data enabling a policy response in that sense. MADELEINE SUMPTION

Yes, and I think it has been very helpful that we've had these data on dependents. Ten years ago we would not necessarily have known. We would have seen that migration was high but it wouldn't have been very easy to distinguish whether people were coming as the partner of an international student or the partner of a health and care worker and now we do have those figures which is incredibly helpful for the policy debate. MILES FLETCHER

Oh, Mary, one thing we haven't spoken about so far is the impact of COVID. How did the system cope with that period? In measuring the negligible flows to start with, but then the turning on the taps again as things returned to normal... MARY GREGORY

I think in terms of measuring the statistics it was a massive challenge because the International Passenger Survey stopped and then it was agreed that it wouldn't restart measuring migration, and it actually accelerated our progress to what is now a better measure, but it happened under very difficult circumstances. So we very quickly moved to using administrative data. So data collected by government already to help us measure. And of course there was so much going on then that added to the challenge. Exiting the EU, changes to the immigration system etc. So it was really important we were very careful about how we make sure we understood what had caused the changes and how we measured it really accurately. MILES FLETCHER

There was indeed another test of the credibility of the ONS migration estimates when it was announced that a very large number of people had applied for settled status just as we were about to leave. These are EU citizens applying for settled status in the UK just as we're about to leave the EU. How do we reconcile those two very different estimates, because a lot of people use them to suggest that there were far more people here than you've been telling us for all these years. MARY GREGORY

What we can do is we can look at the data sources available to us. The census is a really valuable source in that respect because it gives us the most comprehensive view of the whole population for England and Wales, done by ONS. Obviously, Scotland and Northern Ireland are done by their own statistical offices but we can look at that to get a much better understanding of the full components of the population. But of course, it was really difficult. When there was free movement across the EU we wouldn't know for sure how many people have come and how many people have left. And that's actually become a little bit easier in terms of a statistical viewpoint, because now people do need to have visas in order to travel unless they've got settled status already. MILES FLETCHER

The new system has been bedding in these last couple of years, and you've had the unenviable job of announcing some pretty large revisions to the figures. Have things settled down now? MARY GREGORY

So I think we've made really good progress on people traveling from outside the EU as Madeleine already referred to. We know a lot more about them. We're more confident in that aspect, and we would hope therefore, that the revisions in future will be much lower in scale. There will always be some revisions because we are making assumptions about people. Just to pause on that for a second. We publish data five months after the reference period, but obviously it's 12 months before somebody meets the definition of being a migrant. So we have to make some assumptions about who will stay and who won't, but those are relatively small and should be small in terms of revisions. So I think with non EU numbers we have made really good progress and that is the largest part of the picture. So just to put that in perspective, in terms of immigration just over four in five people immigrating in in our latest data are from outside the EU, so that's positive. Where we do have more work to do is those people coming from within the EU and British nationals, and we've got plans to develop the methods for both of those so we will see revisions coming up in both of those areas. We will put out more information in the autumn about the progress we've made, and if they're ready and we think the quality is good enough, we will implement those methods in November. Otherwise, we'll wait until the following publication because for us it's really important that when we do this we do it properly. MILES FLETCHER

And important for everybody to remember that the ONS, in the job it does, can only make the best of the information that's made available to it at any given time. MARY GREGORY

Yes absolutely. And I think especially with the British nationals where there are a lot of challenges. Because, of course, if you're a British national you come and go as you please. The other things that we are looking to improve are going to be less significant in terms of the headline numbers but are also really valuable. So if we can change the methods for EU, for example, we should be able to do more on people's reason for migration. And we also hope to do more on breaking down those from outside the EU, to understand a bit more detail about how long people are staying and if they change visas, that kind of thing. MILES FLETCHER

Mary, thank you very much. That seems a good moment to bring in Hannah.

Hannah, then, from what you've heard, as someone who's in the business of tackling misinformation and ensuring that debates are properly understood, what is your assessment of how useful, how reliable, the ONS migration data are? HANNAH SMITH

Now as we've been hearing from Madeleine there's been some significant improvements in the way that the data is collected and published. I think another thing that can give people confidence is how transparent the ONS has been with not only the strengths of the data, but also the limitations and the work they're doing surrounding ongoing development with that. I think that's absolutely key when we're talking about access to good information –transparency- understanding what the data can tell us and what the data can't tell us, and what the ONS is looking to do to change that. I think ultimately this is, as we've been hearing, a really complex issue, and trying to reconcile that with the fact that it's of massive public interest. And, as Madeline has been saying, someone who is not a technical user of the statistics, it's really important for someone like that to be able to understand these issues in a straightforward way, and trying to find that balance between getting the right level of detail that can be understandable for a general user is difficult. But I think the ONS has been really open about the challenges with that, and this conflict between the idea of timeliness and completeness of data, as Mary was just saying, we don't have complete data at the moment that the first statistics are published, but obviously the alternative is just to wait a really long time until that full data is available. So I think trying to strike that balance is also key, and something that, like I say, just being transparent about that is the best way to approach it. MILES FLETCHER

In your work for Full Fact, what do you come across as the major misuses of migration figures, the deliberate misunderstanding of migration figures. And how well equipped Are you to combat those?

HANNAH SMITH

It's hard to know how much of it is deliberate misuse of migration figures, and how much of it is, as you say, due to just misunderstanding the data. I think there are obviously some things that we don't know, some information gaps. So, for example, the scale of illegal migration is something that's perhaps a bit harder to capture, just by the very nature of it. That's something that we found is a really common theme in the things that we're fact checking. You know, we've seen surveys that show that a quite significant proportion of the public thinks that the data shows that more people are entering the country illegally than legally. We fact check politicians who make similar claims. So we know this bad information does cause real harm, and I think that's why the information that the ONS is publishing is really, really important for reasoned debate, and just having that information available is the first step to help counter the bad information that's out there.

MILES FLETCHER

You mentioned illegal immigration or undocumented migration, that by its very nature is a tricky one, because it's difficult to accurately measure isn't it, and to come up with a robust estimate that can counter exaggerated claims. HANNAH SMITH

Yeah, of course. And we know that some of the people who are arriving in the country, either undocumented or illegally are captured in the data. So for example, the data we have on small boat arrivals, but it is ultimately, like you say, hard to estimate. And I think similarly to what Madeleine was saying earlier about the different definitions of what constitutes a migrant. Different people will have different views or different understandings of what constitutes illegal migration. So that's another thing that we have to bear in mind when we're talking about this issue. MILES FLETCHER

And do you feel you've got the tools to effectively combat the worst excesses of the Wild West that social media often is? HANNAH SMITH

Yes we do work with social media companies. So we have a partnership with Meta which allows us to directly rate misleading content that we see on their platforms. And we definitely do see a lot of content specifically related to migration which thanks to that partnership we are able to have influence on. But at Full Fact I think we're always calling for improvements in how better to combat misinformation, not only in this space, but just generally. So media literacy, for example, we think is a really vital step that's needed to ensure people are equipped so that they can spot what's fact and what's fiction. And we've been making a lot of recommendations in what can be done to improve media literacy to meet the public's needs. We also think that legislation needs to be strengthened to tackle this kind of misinformation and other sorts of harmful misinformation that crop up online. So yeah, we do have a lot of tools at our disposal, but we think that the information environment and the regulations surrounding it could always be strengthened. MILES FLETCHER

That's interesting. And what sort of areas do you think it could specifically be strengthened? As far as the production of statistics are concerned? HANNAH SMITH

I think, as I say, transparency and accessibility is key. I think perhaps trying to anticipate where misunderstandings could crop up. A lot of the work we do, or an approach that we can take with fact checking, is something we call pre bunking, which is trying to look at what topics are resonating with the public, what things we think might crop up, and then producing content that puts the correct information out there. Ideally, trying to get ahead of the bad information. I don't know if I'm going to butcher this saying, but a lie gets halfway around the world before the truth had time to put his boots on. I think that pre bunking is an effort to try and reverse that. And I think there's a parallel there with anticipating, as producers of information, where the misunderstandings might be likely to fall, and putting content warnings on or health warnings as prominently as possible, and also making sure that the people that are using the information, whether that's the media, politicians, other people, are aware of potential pitfalls to try and minimize the risk of that spreading to a wider audience. MILES FLETCHER

Survey information we have from our own sources - the public confidence in official statistics survey - suggests that people who have heard of the ONS tend to recognize the fact that it is independent, that we are not subject to political control, and therefore you might think people should have confidence in the figures. Is that corroborated by your experience? HANNAH SMITH

That's really interesting. I'm not entirely sure. I think from a fundamental point of view, I think trust in organizations like the ONS, knowing that you're getting impartial evidence, knowing that you're getting unbiased information that's been put through the most robust scrutiny that it can be, can only be a helpful thing. We know that trust in politics is at a very low level, so having those impartial producers of information that we know aren't subject to any political control or affiliation, I think can only be beneficial for that. MILES FLETCHER

Madeleine, this is where the National Statistician's Advisory Panel on Migration comes in, the body that you chair. Can you just tell us a little about its work? What its role is? MADELEINE SUMPTION

Yes. So this is a new body that will advise the ONS on migration statistics. Obviously, the ONS migration teams have been soliciting expert advice in various ways for several years. I know this because I've been part of that process as well, but the arrangements are being formalized now, actually building on the model that I think has worked quite well in some other areas of migration statistics, like labour market data, for example, to have a panel of independent experts who help advise on things like whether the statistics are really answering the questions that users have. Obviously ONS has a lot of excellent statisticians, but they're not expected to be deep in the weeds of the policy debate and really sort of understanding exactly how people want to use the data and so forth. So the idea of this panel is to have some of that independent voice to help ONS shape its vision of what kinds of data it can produce. How can it make them more relevant and accessible to users, that kind of thing? MILES FLETCHER

And I guess when you ask most people whether they think migration does have a role to play, particularly in modern economies, answering that question depends on having good data, having data that meets the needs of experts in economics and so forth. So we can see whether indeed, migration is having a positive economic benefit. MADELEINE SUMPTION

Yes there are lots of things that the data are needed for. So looking at the impacts on the economy is one of them that the Office of Budget Responsibility, for example, uses the migration data when it's making its forecast of how much money there is effectively. So you know, how big is the population? What are people likely to be paying in tax? What are we likely to be spending in addition on services? Because we have more migrants, more people in the population. So it's important for the financial impacts. It's important for planning public services. How many school age kids are we going to have? How's that changing? What do we need to do to plan school places? And yeah, then it's important for the broader policy debate as well, understanding different categories of migration, what should the Home Office do? What should other government departments do, and thinking about how to respond to the impacts of migration. MILES FLETCHER

Yeah. And you can't calculate GDP per head of population until you know how many heads there are, to reduce it to its most simple terms. MADELEINE SUMPTION

Indeed. Yep. And that's been one of the challenges. There are more challenges along those lines, when thinking about the impacts of migration, we're getting a lot more administrative data, so data from basically the records of different government departments and agencies around the country, that will tell us things like how many migrants are claiming benefits, or how many migrants are imprisoned, or any number of things. And it's really important that if you want to be able to interpret those statistics, you really must have a good idea of how many migrants are from different countries, different parts of the world, are in the country in the first place. Otherwise you might make your calculations wrong. And I think there is still more work to be done in that area, in particular looking at population. We've been talking mostly about migration in and out of the country. There's still a fair amount more to be done on making sure that we have really accurate statistics on the number of people who are here at any one point in time. MILES FLETCHER

Mary, finally from you then, do you support that good progress has been made, but important steps are still to come? MARY GREGORY

I think so. I mean, there's always improvements that can be made. No matter how good we get, we will always want to do better. But I think also it's such a privilege, but a huge responsibility, to work on something so important, and we don't take that lightly in ONS. We know that these numbers make a difference to so many people, and as Madelene said, the number of people in the country is a really important number, but so often the thing driving that is the migration figure. So without the really good migration data, we don't have the really good population data, and so we will keep working on that together as well as we can. MILES FLETCHER

And on that positive note we must come to the end of this podcast. Thanks to you, Mary Madeleine and Hannah, for your time today, and as always, thanks to you at home for listening.

You can subscribe to future episodes of Statistically Speaking on Spotify, Apple podcasts and all the other major podcast platforms. You can also follow us on X, previously known as Twitter, via the @ONSFocus feed.

I am Miles Fletcher, and from myself and producer Steve Milne, until next time, goodbye.

ENDS