Understanding international migration statistics

Collecting and analysing data to produce migration statistics is a complicated process. With lots of different information around migration, it is easy to misunderstand the data and the picture around migration. Here we aim to bring information together to support our users’ understanding of the main concepts behind international migration statistics produced by the Office for National Statistics (ONS).

The Office for National Statistics (ONS) produces estimates of who comes in (immigration) and who goes out (emigration) of the UK and therefore the population change because of migration (net migration).

We currently publish estimates of long-term international migration twice per year, at the end of May and at the end of November, at the same time as the Home Office release their immigration statistics and the Department for Work and Pensions release their data on the number of National Insurance numbers (NINos) allocated to adult overseas nationals.

We also publish articles throughout the year to provide further insight into international migration and to provide more context to our estimates in response to the needs of our users.

We have recently released a series of articles using data from Census 2021, which provide more information about the size and characteristics of the international migrant population (the number of people who live and are usually resident in England and Wales).

Before mid-2019, long-term international migration estimates were released quarterly and were accredited as a National Statistic in line with the Office for Statistics Regulation (OSR) guidance. As we moved to transform the way we produced migration statistics, we reclassified our statistics to Experimental Statistics in 2019. We will continue to research how to improve the way that we produce estimates and are moving towards having more stable methods and data sources. We have increasing confidence in admin-based international migration estimates and our methodology is beginning to stablise. We have ambitions for UK international migration statistics to be Accredited Official Statistics, which will include, at some point, removing the 'official statistics in development' label. We will continue to work with OSR on an appropriate timeline for this. The official statistics in development label continues to help our users to understand the stage of research.

We define a long-term international migrant (applying the United Nations definition) as someone who changes their country of usual residence for a period of 12 months or more, so that the country of destination effectively becomes the country of usual residence.

Net migration is the difference between the number of people coming to live in the UK (immigration) and the number of people leaving to live elsewhere (emigration), for 12 months or more. When more people are arriving in the UK than leaving, net migration is above zero and so adds to the population. Our long-term international migration statistics include estimates of immigration, emigration and net migration.

It is often thought that you can measure international migration by counting the number of people in and out as they cross the border. However, measuring international migration is a complex process. To provide estimates “on a timely basis and at intervals that meet the needs of the users as far as practicable” (as outlined in the UK Statistical System Code of Practice), we need to make assumptions about the likelihood of those who arrive in the country staying for 12 months or more. This could not be done by simply counting people in and out. For more information see our International migration research, progress update article.

Even using passport stamps as people cross the border is an unreliable method as some people hold multiple passports, and use different passports for inbound and outbound journeys. The UK and Ireland also belong to a free travel zone called the Common Travel Area; people can travel freely between the two countries and movements across the land border between Northern Ireland and the Republic of Ireland are not tracked.

Some countries have systems in place where everyone registers on immigration, and de-registers by law on emigration, but there is no system in place that counts every arrival and departure in the UK in this way.

Our methods are based on administrative (admin) data. These are data collected primarily for operational purposes, but we can use them to estimate the number of people entering and leaving the UK; this is why we refer to statistics as "estimates" rather than "counts" or "measures".

The data used include:

• borders and immigration data from the Home Office

• benefits and earnings data from the Department for Work and Pensions (Registration and interaction population database (RAPID))

• university data from the education sector (Higher Education Statistics Agency (HESA))

• asylum and humanitarian data from the Home Office

We continue to use survey data from the International Passenger Survey (IPS) for British nationals. More information on what data went into the last estimates can be found in our methods paper.

To provide up-to-date data (within five months of the reference period), we initially provide provisional estimates. These provisional estimates provide an early indication of migration until we have seen whether people really did stay in the UK or left the UK for a full 12 months. Initial adjustments are made to the data based on averages from previous years.

For example (not real data):

• we see 100 people arriving holding a long-term study visa

• previous year's data show 10% of those arriving on a long-term study visa do not stay for 12 months or more

• we would initially include 90 of these people in the immigration element of our long-term international migration estimates

• once we have the full dataset from this period, we may see that 15 of these people left the country before 12 months and we would revise our estimate to 85

An estimate is necessarily provisional until we have a full 12-month dataset where we can see in the data how many people stayed or left.

It is likely that our early provisional estimate will be revised once we can observe a full 12 months of activity for people. As set out in our revisions policy, estimates may also be revised as methods are refined, or we obtain more complete data about the period of interest.

Long-term international migration flows and stocks estimate two different things. We can use different data sources to estimate migration flows and stocks.

Flows estimate the number of people moving into and out of the UK. Our long-term international migration estimates provide data on migration flows for a given time-period.

Stocks estimate the number of non-UK-born or non-British citizens resident in the UK at a given point in time. The Census 2021 data provide an estimate of non-UK-born or non-British citizens resident in England and Wales at the point of data collection.

The stock of the migrant population is built up over a period from long-term migration flows. When fully developed, the dynamic population model (DPM) will be able to estimate population and, in time, will produce fully coherent estimates of population counts and changes because of births, deaths and long-term migration. This means that we should be able to estimate a population at any point in time by taking the previous population, adding births, net migration and subtracting deaths to create a new estimate.

Nationality refers to country or countries of legal residence and can be collected through self-reporting, or is based on providing information from an individual's passport. Country of birth refers to the country in which a person is born.

It is possible that an individual's nationality may change, but the respondent's country of birth will always remain the same. This means that country of birth gives a more accurate estimate of change over time. We have published guidance on using country of birth, nationality and passports held data.

In August 2020, we announced we would not return to producing official migration statistics for non-UK nationals from the International Passenger Survey (IPS) and instead transform our methods using admin data to estimate long-term international migration. For decades, the IPS was the best data source that the Office for National Statistics (ONS) could use to provide estimates of both long-term immigration and emigration, and therefore net migration estimates for the UK.

However, this method is based on a respondent's intended plans to move to or from the UK, thus has certain limitations, because people can change their plans and therefore their stated intentions would not reflect their actual behaviour. This limitation of the IPS is more prominent during periods of high uncertainty, such as during the coronavirus (COVID-19) pandemic and following the EU referendum.

As a result of the limitations of the IPS, and the fact the IPS was paused during 2020, the ONS accelerated its approach to changing the way migration statistics are produced from surveys to admin-based migration estimates (ABMEs) to measure migration.

Our Estimating International Migration: 2012 to 2021 further shows the benefits of developing our methods using admin data to deliver improved estimates of migration.

We have developed our methods in-line with our move towards producing estimates based on actual observed activity in admin data. Admin data refers to any collection of data that has been primarily collated for operational purposes, for example, visa information and travel records.

Admin-Based Migration Estimates (ABMEs) bring data together and estimate how many people arriving and departing in a given period are long-term international migrants. Our ABMEs are used to estimate EU and non-EU nationals migrating to and from the UK. For British nationals, we continue to use the International Passenger Survey (IPS), supported by findings from the 2021 Census while we explore the best admin sources to estimate this group. For more information on our methods and transformation, please see our Improving migration statistics using administrative data article.

As described in Section 7: Publishing provisional estimates, provisional estimates are updated once we have 12 months of data. For example, an improvement to our methods in 2030 may mean that estimates published in 2028 will be revised. This could mean that estimates of net migration increase or decrease, depending on the method being improved. More information on what data went into the last estimates can be found in our Methods to produce provisional long-term international migration estimates methodology.

There is a difference between what each of the data sources, such as Home Office visa data, data on the number of National Insurance numbers (NINos) allocated, and our published migration data, aim to estimate. We use Home Office visa and travel data to estimate the number of non-EU arrivals but our estimates do not estimate visas issued, but instead consider people classed as long-term international migrants. These are people who changed their usual country of residence for 12 months or more.

Some people who are granted a visa choose not to come to the UK and some who do come will not stay in the UK for 12 months, so the number of visas issued will be higher than long-term immigration estimates. The net migration figures that we publish also consider the number of people that are leaving, which Home Office visa data cannot.

Overseas nationals can apply for a NINo as soon as they arrive in the UK, or at any point during their stay so NINo registrations do not necessarily mean a recent move to the UK. For example, a student might migrate to the UK, complete their studies, and only then apply for a NINo if they choose to stay and work through the graduate scheme or by transitioning to a work visa.

We estimate that:

• study-related visas are the most common reason for immigration for non-EU nationals

• "work" is the most common reason for immigration for EU nationals

• "other" (not work, family or study) is the most common reason for immigration for British nationals

For non-EU nationals we can infer their reason for migration based on the type of visa they enter the UK on. However, for EU migration and returning British nationals, we do not have this information. We plan to use Home Office data for EU nationals applying for visas (excluding EU Settlement Scheme visas) in the future, so these will be comparable with non-EU nationals. More details are available in our International migration research progress update article.

Therefore, to produce statistics on the reason for EU and British migration, we still rely on International Passenger Survey (IPS) data. The data should not be compared to non-EU data. For EU migration we use the IPS data to estimate the proportion of immigration for each reason and apply this to the admin data-based estimates. For more information see research update.

In line with the current UN definition of a long-term migrant (PDF, 5.18MB), international students who are studying in the UK for a period of 12 months or more are included in our estimates of long-term immigration.

Students are often thought of as a temporary population who arrive in the UK to study and leave on completion of their studies. However, our previous research on visa journeys and student outcomes has shown that 35% of students who finish their studies go on to obtain further visas or become British citizens.

We are looking to explore this further in the future (see Reason for international migration, international students update: November 2023 - Office for National Statistics.

Because of these patterns of movement between different reasons for migration, and their contribution to population change, we need to develop our methods to add students who transition to other visa types into our estimates.

Every 10 years, the census provides the opportunity to measure the number of international migrants in the country at that given point in time. The long-term international migration (LTIM) estimates are produced from admin data sources (see previously) in addition to other adjustments.

We continually quality-assure our estimates against many data sources, including the 2021 Census. We are confident in these estimates, but it is important to acknowledge that our LTIM estimates relate to migration flow, whereas the census data provide a glimpse into the stock at the point in which the census was taken.

In our Estimating UK International Migration: 2012 to 2021 article, we have used new methods and admin sources to revise EU, non-EU and British long-term international migration to and from the UK. This is in line with evidence provided by Census 2021 and other sources. These revised estimates for the full data time series date back to 2012 and are also now as consistent as possible with our latest methods.

We are confident that these new methods provide us with a more accurate picture of migration patterns over the last decade than it was possible to produce at the time using the intentions-based International Passenger Survey (IPS).

Our best estimates for international migration (and other population statistics) during the period of the Brexit vote, exiting the EU, and the coronavirus (COVID-19) pandemic have been updated. However, it is difficult to separate the individual impact of each of these.

For context, while the UK left the EU on 31 January 2020, the country continued to be in a "transition period" until the end of 2020. During this time, existing arrangements applied, meaning that people could migrate freely between the UK and the EU without needing a visa. It will also be difficult to disentangle the effects of Brexit and the pandemic on migration patterns, given that both the UK exit and impact of pandemic restrictions occurred over the same time period.

As part of our iterative transformation process, we are continuing to develop our methods to include all migrant groups. From the estimates for year ending December 2022 (released in May 2023), we have included applications of asylum for the first time in our admin data-based estimates of immigration and revised these back to previous estimates. Someone is included in our estimates of asylum from the point they have applied and under our current system, they would all be expected to remain more than 12 months, thus would be considered a long-term migrant. We use the available return data supplied by the Home Office to remove known emigrants.

From May 2023, we also included resettlement scheme arrivals in our immigration estimates. There are no returns for resettled refugees on the basis that all resettled persons are long-term migrants based on declared intent to stay in-country.

Those who arrived on small boats and proceeded to claim asylum are included in our estimates following their claim. We did not separately identify small boat arrivals in our estimates, but analysis published by the Home Office showed that, for the year ending March 2023, 90% of small boat arrivals claimed asylum or were recorded as a dependant on an asylum application.

The way that estimates are produced for international migration differs from country to country. Therefore it is difficult to compare directly. Eurostat provide breakdowns of immigration, emigration and acquisition, and loss of citizenship for each EU country (including the UK up until 2021).

We are currently working to identify the best possible data source for estimating migration of British nationals. Because of the complexity associated with identifying British national migrants in the admin data sources that we use to estimate net migration, we cannot currently use these data. As a result of this, the International Passenger Survey (IPS) is currently our source of information in measuring these migrants and this helps to form our net migration estimates. However, we do not produce separate emigration estimates of British nationals living abroad. The last release looking at British nationals living in the EU was released in April 2018.

We currently publish net migration at a UK level. As our transformation work continues, our ambition is to give further breakdowns of international migration for users, including by geographies.

The Home Office immigration data for 2023 suggest that:

• for EU countries, people with German, French and Spanish nationality were issued the most visas

• for non-EU countries, people with Indian, Nigerian and Chinese nationality were issued the most visas

This includes all reasons for migration. It is important to consider that not all visas issued are used, and not all visas are for long-term migration, so these numbers do not equate to the number of long-term international migrants arriving.

It is possible to request more detailed information and data on international migration by emailing pop.info@ons.gov.uk.