Developing our international migration estimates

ONS is transforming the way it produces international migration statistics. It has done so against the backdrop of significant change in the UK and abroad. Here Mary Gregory describes how ONS continues to understand more, not only about new data sources, but also changing behaviours of migrants and how that affects the estimates.

Since 2021, long-term international migration to the UK has been at unprecedented levels. This has been driven by a variety of factors, including the war in Ukraine, the effects of the post-Brexit immigration system and pent-up demand for study-related immigration because of travel restrictions during the coronavirus pandemic. While remaining at a historically high level, we are now beginning to see a reduction, with immigration falling and emigration increasing.

For the year ending (YE) June 2024, we estimate long-term international net migration provisionally stood at 728,000. This is 20% lower than our updated net migration estimate for the YE June 2023 (906,000), which is the highest level we have seen.

In the 12 months to June 2024 we have seen a fall in immigration, driven by declining numbers of dependants on study-related visas from outside the EU. In our most recent data points, we are also seeing decreases in the number of people arriving for work-related reasons. This is consistent with visa data published by the Home Office, and in part reflects policy changes earlier this year. Home Office analysis looks in more detail at the reasons for these falls.

We are also starting to see increases in emigration, most notably for those who came to the UK on study-related visas. This is likely to be a consequence of the higher numbers of students coming to the UK post pandemic who are now reaching the end of their courses.

Innovation brings with it change and learning

These estimates are published as official statistics in development. This reflects the continuing development of our approach to measuring long-term international migration. Until 2020 we primarily used the International Passenger Survey (IPS). Through this survey we asked a sample of people about their intentions when they arrived in or left the UK. This survey was paused during the pandemic, so we accelerated work to use 'admin data' - data collected by government for administrative purposes, including visa information, higher education statistics and tax data. Using admin data enables us to see actual behaviours, rather than relying on intentions.

As part of our ongoing improvements, in today's release, we have updated our previously published migration estimates back to YE June 2021. This reflects new work around the assumptions we need to make for provisional estimates and additional information from admin data relating to Ukraine visas and individuals transitioning to new visas. Together, these updates have increased our previous estimates of net migration. For example, net migration has been revised upwards by 166,000 for YE June 2023 and by 181,000 for YE December 2023.

Improvements to assumptions

Long-term international migration refers to people entering or leaving the UK for a period of at least 12 months. To produce timely estimates, we produce an estimate of net migration before 12 months has passed. This means when we publish the provisional estimates we do not have all the travel information available and make assumptions about how long new arrivals are likely to stay in the UK. Over time we can see whether a migrant has interacted with admin data sources, which then enables us to update our estimates based on actual behaviours. As it does every time we publish, this new data has led to revisions to our previous provisional estimates. For example, in the year ending December 2023, having more complete and up to date travel information has added around 25,000 people to our previously published estimates.

For this publication, we have also improved our assumptions, so that future revisions resulting from these assumptions should be smaller.

Since the pandemic and changes to the post-Brexit immigration system, we have seen changes in who is coming to the UK. Instead of migration to the UK being driven by EU+ nationals (EU+ includes EU countries plus Norway, Iceland, Liechtenstein and Switzerland), we're now seeing more people from further afield, especially Indian, Chinese and Nigerian nationals. We are also seeing specific patterns of behaviour, for example Chinese students, including those who come on shorter Masters' courses, are more likely to leave before 12 months than Nigerian students. We explain this in our student paper. We now use our knowledge of past migration by age and nationality to improve our assumptions on whether those who have recently arrived in the UK will become long term migrants. This should improve our estimates and make them more resilient to any future changes in patterns of behaviour.

Additional information

We have improved our understanding of visas granted to people on the rapidly introduced Ukraine schemes. We now include additional Ukraine scheme immigrants, which has increased immigration by 43,000 for YE June 2023 and 30,000 for YE December 2023.

We have also made improvements to how we identify people moving to new visas (e.g. students moving to graduate visas) for non-EU+ nationals. Previously we could only see that an individual had a new visa if they travelled on that visa. This meant if a migrant stayed in the UK without travelling we would not see their change in visa and would assume they had left the UK (e.g. through the Common Travel Area) when their original visa expired. In the data we are provided with, we can now see when people move to a new visa as soon as it is granted. This has reduced non-EU+ emigration, especially student and work dependants. For example, for the year ending December 2023 we have reduced our emigration of non-EU+ nationals by 69,000 from 233,000 to 164,000, as our research paper explains. This increased knowledge and understanding should reduce revisions in future.

More information on the impact of these changes can be seen in our main bulletin.

What's next for migration?

We have seen a reduction in net-migration in the latest data, which appears to be the start of a downward trend. However, we will be able to see the extent of the impact of policy changes more fully in May 2025, when we release data for the year ending December 2024.

While we want to minimise the impact of revisions for our users, we are committed to continuous improvement in our approach to measuring migration. It is important we are transparent about changes and make those changes only once they represent improvements to the estimates.

We're increasingly confident in our understanding of non-EU+ migration and we expect revisions for this group to become smaller with time. However, we are intending to improve our methods for estimating migration for EU+ and British nationals in 2025. These are the two smaller parts of the picture of migration to the UK, together making up around 15% of immigration in YE June 2024.

For EU+ migration, we currently use the Department for Work and Pension's Registration and Population Interaction Database. We're exploring alternative sources, such as Home Office Borders and Immigration data, which should also enable us to better identify reasons for migration.

For British nationals, we remain reliant on the IPS. We know the challenges with this and are planning to use new sources in 2025. We've provided more information in our research update.

Further work

Today we have also responded to user feedback by incorporating an estimate of the non-UK born population into our release. Rolling on from Census 2021, it estimates the non-UK born population of England and Wales was 11.4m, 1.4m higher than in 2021. This is a temporary solution while we work to find an improved approach. We'll be providing an update on this work next year.

If you have any feedback on today's releases, please contact [email protected].

Public link

Please use the above public link if you want to share this noodl on another website.