What UK Migration Statistics Include and Leave Out

A migration figure can look precise while answering a much narrower question than many readers realise. UK migration statistics usually measure people who enter or leave the country for at least 12 months, not every person who crosses the border or lives here without clear official records.

That distinction matters in political debate. Immigration, emigration, net migration, asylum applications and the foreign-born population are separate measures. They describe different parts of the picture, so comparing them without checking the definitions can produce a misleading result.

Key Takeaways

• Long-term migration statistics count people expected to change country for at least 12 months.
• Net migration is immigration minus emigration. It isn’t the total number of arrivals.
• Short-term visitors, tourists and many temporary workers don’t appear in the long-term headline.
• The figures don’t provide a complete live count of everyone without lawful status.
• Estimates can change when better administrative data, revised methods or new behaviour patterns become available.

What the headline migration number actually measures

The main ONS measure is long-term international migration. It follows the United Nations definition, counting someone as a long-term migrant when they move to another country for at least 12 months.

The calculation has three parts:

• Immigration: people arriving in the UK intending to stay for at least a year.
• Emigration: people leaving the UK intending to live abroad for at least a year.
• Net migration: immigration minus emigration.

For example, if 900,000 people move to the UK for at least 12 months and 400,000 leave for the same period, net migration is 500,000. The 500,000 figure isn’t a count of everyone who arrived. It is the balance after departures are deducted.

The latest available ONS data in July 2026 put provisional long-term immigration at 898,000 in the year ending June 2025. That figure is subject to revision. It describes a twelve-month period and covers people who meet the long-term definition, rather than giving a live total for the population.

The ONS publishes a useful guide to international migration statistics, including explanations of immigration, emigration and net migration. Checking the definition before quoting a number prevents one of the most common errors in migration reporting.

A yearly flow is also different from the number of migrants living in the UK. One person can appear in an immigration estimate after arrival, then appear in an emigration estimate years later. The figures are about movements over time, not a register of every foreign national currently present.

Net migration is a balance, not a tally of arrivals.

The distinction becomes even more important when a political claim uses a net figure to describe pressure on housing, schools or public services. Those effects may relate to arrivals, departures, the length of stay and where people live. A single national balance cannot show all of those details.

Who is included in UK migration statistics?

Nationality isn’t the main test. Length and purpose of residence matter more.

A British citizen returning after living overseas can count as an immigrant if they intend to stay in the UK for at least a year. Equally, a UK-born person who moves abroad for a year or longer can count as an emigrant. The statistics therefore don’t mean “foreign nationals arriving” when they use the word immigration.

The estimates can include people arriving for:

• Work under a visa or another lawful route
• Study at a university or other institution
• Family reasons
• Humanitarian protection or asylum
• Resettlement programmes
• Other long-term reasons

EU and non-EU nationals both appear in the data. The post-Brexit immigration system changed the routes used by many European citizens, but it didn’t remove EU nationals from migration statistics.

Recent methodology changes have also improved the treatment of asylum applicants and people arriving through resettlement schemes. These groups are now included explicitly in the international immigration estimate, alongside the separate Home Office figures that track asylum applications and decisions.

That doesn’t make the measures interchangeable. An asylum application is an event or case recorded by the Home Office. A long-term migration estimate is an estimate of a person’s residence for at least 12 months. One applicant may wait for a decision, appeal a refusal, receive permission to stay or leave the UK. The two systems answer different questions.

The figures can include someone who enters without a visa if the person later appears in relevant administrative records and meets the residence definition. However, the system doesn’t create a perfect list of every person in the country. Some people leave no reliable administrative trail, while others appear in one dataset but not another.

Short-term migrants are treated separately. A person who comes for less than 12 months may be counted in short-term migration estimates, but won’t normally appear in the long-term headline. Tourists, business visitors and ordinary holidaymakers also don’t count as long-term immigrants.

How the ONS builds the figures

Migration statistics once relied heavily on the International Passenger Survey, which asked travellers about their plans at selected ports and airports. That approach could provide useful information, but it had weaknesses. People stated their intentions, and intentions don’t always match what happens later.

Someone may say they plan to stay for two years, then leave after eight months. Another person may expect a short stay and later remain for several years. A survey based on interviews at the border can’t observe every change in plans.

The ONS has been moving towards administrative data. These records can show interactions with immigration, tax, benefits and other public systems after a person arrives. The main sources include:

• Home Office Borders and Immigration data for many EU and non-EU nationals
• Department for Work and Pensions records
• The Registration and Population Interaction Database, known as RAPID
• Census information used to test and adjust population coverage
• Passenger and survey data where administrative information remains incomplete

Administrative data isn’t automatically perfect. RAPID was built to record economic activity and benefits interactions, not to function as a migration register. A person who doesn’t work, claim benefits or appear in another linked system may be harder to identify. The ONS therefore applies adjustments, including age-based comparisons with census data.

Irish citizens create a separate issue. The Common Travel Area allows movement between the UK and Ireland without routine border controls. Home Office border data doesn’t capture Irish nationals in the same way as other groups, so the ONS needs a separate adjustment using DWP information.

The data also arrives at different speeds. The latest estimates available in July 2026 still have a publication lag of about five months. That means a May release can cover the year ending in the previous June. A headline published today is therefore a recent estimate, not a live count.

Different records provide different pieces of the picture, and the ONS must join them without counting the same person twice. That process improves coverage, but it also means the published series can change when the method changes.

Immigration, net migration and the population are different measures

Many arguments about migration become confused because three separate statistics are treated as though they mean the same thing.

Immigration measures people arriving for at least 12 months. It shows the inflow.

Emigration measures people leaving for at least 12 months. It shows the outflow.

Net migration subtracts emigration from immigration. It shows the difference between the two.

The foreign-born population is a stock measure. It counts people living in the UK who were born in another country, regardless of when they arrived or what their current nationality is. A person who arrived 20 years ago can remain in the foreign-born population, even though they aren’t part of this year’s immigration flow.

The House of Commons Library reports a foreign-born UK population of 10.7 million people, or 16% of the total population, in the data covered by its migration statistics briefing. That figure cannot be compared directly with annual net migration. One is a population stock, while the other is a yearly balance.

Population change also includes births and deaths. A rise in the total population may result from natural change, migration or both. Migration statistics alone can’t show the full demand placed on public services.

The location of migration matters too. A national figure doesn’t show whether arrivals settle in London, a university city, a coastal town or a rural district. People may work in one council area, live in another and use services in a third.

The Migration Observatory’s analysis of long-term migration flows to and from the UK gives useful context on the difference between flows, stocks and net migration. It also shows why a single headline number can’t answer every question about population change.

What the official figures leave out

The most important omission is a complete, real-time count of everyone who is physically in the UK without lawful status.

The ONS can estimate long-term migration using available records and statistical models. However, it doesn’t publish a precise total of all people who have overstayed a visa, entered without permission or disappeared from official systems.

Current methods have known limits. They may not identify people who overstay a visa without claiming asylum. In some cases, the method assumes that a person has emigrated when the records no longer show them. That assumption can lead to undercounting if the person remains in the UK.

The data also has difficulty accounting for people arriving or leaving with Indefinite Leave to Remain. Their status doesn’t fit the same temporary visa patterns, so administrative records may not reveal their movement in a reliable way.

Short stays are another omission from the headline. A worker, student or visitor who remains for less than 12 months may still use housing, transport and local services. They may appear in a short-term migration measure, but not in long-term immigration.

The headline also leaves out detail about legal status unless another dataset supplies it. It doesn’t, by itself, show how many people have no right to work, how many are awaiting an asylum decision or how many have exhausted appeal rights. Those questions require Home Office enforcement and asylum data, court records or other sources.

Small-boat arrivals need careful handling. The number of people crossing the Channel is a count of detected arrivals or recorded events. It isn’t automatically the same as long-term immigration, net migration or the total number of people who remain in the country. Updated ONS methods can include asylum applicants in long-term estimates, but the border crossing data and migration data still measure different stages.

A further omission concerns impact. National migration statistics don’t show the cost of accommodation, the number of NHS appointments, school places or council services used by any particular group. Those questions require local and departmental data.

Why migration figures are revised

Official estimates are often provisional because the ONS needs time to confirm whether people stayed for 12 months. Early figures rely on patterns, assumptions and partial records. Later data can show that people stayed longer, left sooner or followed a different route than first expected.

The revision can be substantial. In May 2026, the estimate for 2024 net migration was revised down by 100,000, or 23%, compared with the previous estimate. That doesn’t mean the original number was deliberately false. It means the underlying evidence and method produced a better estimate later.

Changes in data sources can also affect historic years. The ONS has revised population estimates for 2022, 2023 and 2024 while incorporating improved migration estimates. It has also improved its treatment of people with Section 3C leave, which can extend a person’s immigration permission while an application or appeal is being decided.

Family-related emigration was another area where earlier data had limitations. Improved methods can raise the number of people identified as leaving, which reduces the resulting net migration figure even if immigration is unchanged.

Student movements have required attention too. Analysis indicated that non-EU student emigration may previously have been underestimated. An uplift to estimated non-EU departures was applied, lowering non-EU net migration.

These changes make comparisons harder. A newspaper may place today’s revised estimate beside an older number that was published before revision. That creates an apparent change that partly reflects better measurement rather than a sudden movement in the real world.

The figures also depend on future travel patterns. The ONS must estimate what visa holders will do after arrival and how many will remain for the full period. Behaviour can shift when visa rules, labour demand, university recruitment or economic conditions change.

A revision doesn’t erase the earlier number. It shows that migration measurement is updated as evidence improves.

What the figures can and cannot show locally

National statistics are useful for understanding broad movements, but they become less certain below the national level. Small samples and incomplete records create wider margins of error for individual regions and council areas.

The ONS has warned that some local estimates aren’t reliable enough for fine-grained analysis. A national total may support a strong estimate, while dividing it across hundreds of local authorities can produce figures too uncertain for confident conclusions.

That matters because migration’s practical effects are local. A national immigration estimate doesn’t tell a Durham resident how many new households entered their area, how many homes they occupied or how many children joined local schools. It also doesn’t show the number of people who moved elsewhere after arriving in Britain.

The same caution applies to claims about costs. A figure described as a bill per household normally involves extra calculations about public spending, tax receipts, housing demand or future liabilities. It isn’t contained inside the ONS migration headline.

Political parties use migration statistics alongside policy arguments. Reform UK’s 2026 border plan argues for tougher enforcement, faster removals and tighter routes to long-term residence. The Reform UK official website sets out the party’s wider position on immigration and border control.

Those are policy positions, not statistical findings. A reader should keep the two separate. The data can establish how a number was calculated, while a party manifesto or policy paper argues what government should do about it.

The distinction doesn’t weaken the political case for stricter control. It improves the argument by showing which claims the evidence supports, which require extra assumptions and which remain unanswered by the headline.

How to read a migration statistic before sharing it

A few checks can reveal what a figure really says:

1. Check the period. Is it a calendar year, a year ending in June or a shorter quarter?
2. Identify the measure. Is it immigration, emigration, net migration, asylum applications or the foreign-born population?
3. Read the definition. Does the number cover stays of at least 12 months, or does it include short-term movement?
4. Check its status. Is the estimate provisional, revised or final?
5. Find the source. ONS, the Home Office, the Census and the Labour Force Survey don’t measure identical things.
6. Look for the denominator. A percentage of the population, a rate per 1,000 people and a total count can tell very different stories.

Then ask what the statistic cannot answer. If a chart shows 898,000 immigrants, it doesn’t reveal how many were still in the UK at the end of the period. It doesn’t show their nationality alone, their future contribution or the number who later left.

If a claim cites net migration, check whether it also discusses emigration. If it cites asylum applications, check whether it treats applications as successful claims or permanent settlement. If it cites the foreign-born population, remember that the figure includes people who may have lived here for decades and may now hold British citizenship.

A strong source will state its coverage, date and uncertainty. It will distinguish recorded counts from estimates. It will also explain revisions rather than presenting every change as a new arrival or departure.

For campaigners and voters, that discipline matters. Clear statistics are harder to dismiss, easier to test and more useful when Parliament debates border rules, visas, removals and integration.

Conclusion

UK migration statistics provide important evidence, but each number has a defined boundary. Long-term immigration counts expected stays of at least 12 months, net migration balances arrivals against departures, and the foreign-born population measures people living here who were born overseas.

The figures leave gaps around short stays, visa overstays, people with Indefinite Leave to Remain and anyone who never appears reliably in administrative records. They also change when the ONS improves its methods or receives better evidence.

Before accepting a headline, check the measure, date, source and assumptions. Accuracy begins with asking what the number includes, then asking what it leaves out.