Home / publications / briefings /

Non-EU Family Migration to the UK

10 Apr 2019

This briefing examines the migration of non-EU family members to the UK. This includes both ‘family unification’, where migrants come to live in the UK with family members who are either British citizens or settled residents; and ‘dependants’, who are the family members of migrants who have permission to stay only temporarily.

  1. Key Points
    • Non-EU long-term family migration to the UK has declined since 2006, and stood at 49,000 or 17% of all non-EU immigration in 2017
    • Non-EU family migrants are more likely to settle in the UK long term, compared to workers and students
    • In 2017, 95% of non-EEA family unification migrants were marital partners or children
    • South Asia is the most common region of family migrants’ nationality
    • Dependants of temporary visa holders outnumber family unification migrants, and the dependants of workers contribute the largest share of dependants
  1. Understanding the Evidence

    Policy on family migration

    Unlike citizens of EEA countries or Switzerland, non-EEA and non-Swiss nationals require a visa to come to the UK for six months or more to live with a family member. (In this briefing, “non-EEA” should hereafter be taken to mean also non-Swiss.)

    In this briefing, we discuss two main kinds of family migration. ‘Family unification’ refers to migrants entering or remaining in the UK to live with family members who are British citizens or have settled status. This includes unmarried or same-sex partners; fiancé(e)s or proposed civil partners; spouses or civil partners; children; and (in a smaller number of cases) adult or elderly dependent relatives. Other people come to the UK as ‘dependants’ of migrants without British citizenship or settlement and who have permission to stay only temporarily, on any other visa, such as work, study, or family.

    Since July 2012, if a non-EEA citizen is applying as a partner of a British citizen or non-British citizen with indefinite leave to remain, the sponsor must have an annual income of at least £18,600, or £22,400 if the partner sponsored is bringing one child, plus £2,400 for each of any additional children. This is the ‘minimum income threshold’ (see the Migration Observatory report, The Minimum Income Requirement for Non-EEA Family Members in the UK). A partner may apply for permanent settlement after five years of residence in the UK.

    Family migration rules in the UK primarily admit spouses or children. Adult or elderly dependent relatives may be able to come under certain, restricted circumstances, and must, for example, demonstrate that they require long-term personal care to perform everyday tasks, and that such care can be provided only in the UK by their sponsor (and not abroad by someone else), without recourse to public funds.

    From November 2010, non-EEA nationals applying to enter or extend their stay as a partner of a British citizen or settled person were required to demonstrate a level of English language proficiency, with some exceptions.

    Data sources

    This briefing draws mainly upon three sources of data about non-EU/non-EEA family migration to the UK: long-term international migration (LTIM) estimates produced by the Office for National Statistics (ONS); Home Office administrative data on visas issued and asylum applications; and Home Office data on border admissions. These sources define and measure family migration differently, and hence produce different estimates of it.

    The ONS estimates are based primarily upon data produced by the International Passenger Survey (IPS). The IPS is a sample survey conducted at all of the UK’s major airports and sea routes, and the Channel Tunnel. The IPS defines a migrant as a respondent who states that they intend to come to the UK for one year or more. Migrant respondents are asked to provide their “main reason” for migrating. The question is an open one and the answers are coded by an interviewer into one of four categories, including to “accompany/join” a family member, which encompasses both family unification and dependant migration.

    Home Office administrative statistics on visas issued and asylum applications are the second main source of data. The Home Office classifies individuals by the type of visa they were granted, which includes visas to enter the UK, and any subsequent grants of extensions of stay in the UK, as well as grants of settlement (indefinite leave to remain) or settlement.

    Unlike ONS estimates, the Home Office breaks down family migration into three categories: “family” visas (i.e., visas for the purposes of family unification); “dependants joining or accompanying” (i.e., dependants); and “EEA family permits”, for non-EEA migrants who enter or remain in the UK as family members of EEA nationals. Home Office visa statistics include people entering or staying in the UK for any period of time, and not just a year or more. This is one reason that Home Office estimates of family migration are bigger than those based on IPS data (as seen in Figure 2). Other reasons for the difference include: some people issued visas do not arrive in the UK, which would prevent them from being included in IPS counts; the visa granted might be different from the main reason of stay stated in response to the IPS; and some people switch visas while in the UK, which might create a further inconsistency with their stated reason for migration in response to the IPS.

    Dependant visas are granted by the Home Office to a “main applicant” and their dependant(s), with the dependant recorded under the immigration route of the main applicant (i.e., as a dependant of a worker or student). There are also data on the dependants of asylum seekers. Asylum seekers and their dependants are not issued visas and are recorded separately by the Home Office. These figures do not include people who join someone who has already been granted refugee status or humanitarian protection.

    A third source of information on family migration are Home Office administrative data on non-EEA border admissions, which are compiled from samples of landing cards completed by passengers arriving at UK ports of entry. Home Office data on border admissions break down family unification migration into ten sub-categories, depending on the entrant’s relationship with their sponsor: common law spouse, same-sex partner, fiancé, fiancée, proposed civil partner, husband, wife, civil partner, child, and ‘other’, which includes adult or elderly dependent relative).

    Note that the ONS data are for non-EU citizens while the Home Office visa data are for non-EEA citizens (i.e., citizens not from EU countries, or Iceland, Lichtenstein, or Norway). Because immigration from Iceland, Lichtenstein and Norway is small, non-EU and non-EEA migrant numbers are very similar.

Non-EU long-term family migration to the UK has declined since 2006 and stood at 49,000 or 17% of all non-EU immigration in 2017

The number of family migrants to the UK has been relatively stable over the last 25 years, in contrast to other types of migration that have risen and fallen more dramatically.

During the 1990s, the number of non-EU citizens moving to the UK for at least a year to be with family members averaged around 36,000 per year. Non-EU family migration averaged around 60,000 per year from 2000–2009, and 48,000 per year from 2010–2017.

Despite family migration levels being relatively stable, family makes up a smaller share of all migration than it did in the 1990s. This is because total immigration also increased over this period, and more sharply than family migration (Figure 1). In 2017, family migration made up 17% of all non-EU long-term immigration.

Recent declines in non-EU family migration began in 2006, reaching a low point of 39,000 in 2012 – the lowest level since 1998. This coincided with a fall in overall migration from 2006, and may also have been driven in part by the introduction of the English language requirement in 2010 and minimum income threshold in 2012 (see the Migration Observatory commentary, How many people have been prevented from bringing a partner to the UK due to the £18,600 minimum income requirement?).

Figure 1

Data on the number of family visas issued confirm the downward trend in the late 2000s (Figure 2). (Comparable visa data are not available before 2005.) The number of family visas granted is higher than the ONS estimate of long-term family migration, primarily because visa data include people intending to move to the UK for less than one year (see Understanding the Evidence for further causes of the difference).

Figure 2

Return to top

Family migrants are more likely to settle long term in the UK than workers and students

Of just under 50,000 people granted a family visa in 2011, 69% (just over 34,000) had permanent settlement or British citizenship five years later (Home Office, 2018). This included 79% of people with visas for family unification and 44% of dependants (Figure 3).

One reason dependants are less likely to settle than family unification migrants is that their stay in the UK depends on the length of stay of the “main applicant”. Among main applicants, rates of settlement are lower for those on work or study visas than for those on family unification visas. Further information on settlement, including data on parents, grandparents, and other family members, can be found in the Migration Observatory briefing, Settlement in the UK.

Figure 3

Whilst family migration makes up a small share of all immigration to the UK (Figure 1), because family migrants are more likely to settle, they make up a large share of the non-EU migrant population living in the UK. In 2017, 50% (2,815,000) of the UK’s non-EU born population said that they had come to the UK for family reasons, according to the Labour Force Survey (Table 1).

Table 1

Great Britain non-EU born population by main reason for migration, 2017

Main reason for migrationNumber of foreign bornPercentage of non-EU foreign born
Source: Migration Observatory analysis of Labour Force Survey, taken from Migration Observatory briefing "Where do migrants live in the UK?" Note: excludes Northern Ireland.


Return to top

In 2017, 95% of non-EEA family unification migrants were partners or children

A large majority of people admitted to the UK for family unification are partners or children: 95% in 2017, according to data on admissions at the border. The most common category of family unification migrant is wives. In 2017, wives constituted 55% of all family unification border admissions (11,300), while husbands made up 20%, civil partners 9%, and children 10%.

Statistics are not regularly published on the number of adult or elderly dependent relatives (ADRs) admitted. However, the Home Office (2018b) has said that the number of ADR visas issued has been small since July 2012, when new rules came into effect restricting their entry: 2,782 ADR visa applications were made from 9 July 2012 to 31 December 2015, of which 606 were successful – contributing an average share of around 0.5% of all family unification visas per year.

South Asia is the most common region of nationality for family migrants

South Asia has been the most common region of nationality for family migrants since at least 2004, according to long-term immigration estimates. In 2017, citizens of South Asian countries (Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, Sri Lanka) were particularly likely to have moved to the UK for family reasons, compared with other regions (Figure 4). Over a third (36%) of all long-term migrants from South Asia were family migrants.

Figure 4

Return to top

Dependants of temporary visa holders outnumber family unification migrants, and the dependants of workers contribute the largest share of dependants

From the year ending December 2005 to the year ending December 2018, dependants of people on temporary visas contributed 50% or more of total family migrants in all years except 2018, outnumbering family unification migrants (Figure 5).

Figure 5

The post-2012 increase in the issuance of EEA family permits coincided with an increase in the population of EU migrants during this period (see the Migration Observatory briefing, EU Migration to and from the UK).

Among dependants of people on temporary visas, the largest category is dependants of work visa holders (Figure 6). From 2005 to 2018, the dependants of workers averaged around 45,000 per year. The annual average number of student dependants for this period was 20,000 per year, compared with 6,000 per year for asylum dependants.

Figure 6

Return to top

Evidence gaps and limitations

Because there is no single definition of family migration used across all datasets, evidence produced by different datasets are not comparable. ONS long-term migration estimates are based on migrants’ self-reported intention to reside in the UK for one year or more, and on migrants’ self-reported reason for migrating, though motivations for migration are complex, with people often moving for multiple reasons.

By contrast, Home Office visa data are based on administrative counts and reflect a person’s legal permission rather than their intention. They also include people coming for periods of less than a year. This explains in part why estimates of family migration differ across data sources, sometimes substantially, and why caution must be exercised in comparisons across them.

In addition to this lack of commensurability between data sources, each source also has its own particular limitations. ONS estimates of long-term international migration are based on the International Passenger Survey, which has several well-known limitations. These have led the ONS to state that the IPS has been “stretched beyond its original purpose”. Of these limitations, two are especially relevant to ONS estimates of non-EU family migration.

First, according to a review of the IPS by the UK Statistics Authority (2013), estimates based on the survey are “subject to relatively wide margins of uncertainty”. These margins become proportionally larger when examining subsets of long-term migrants, such as family migrants.

Second, in a review (2014) of the quality of its long-term international migration estimates from 2001 to 2011, the ONS revised its net migration estimates for that decade, based on evidence that they had substantially underestimated migration from the EU8  countries. However, no revised estimates have been provided for the more detailed breakdown of inflows and outflows of long-term international migrants by variables such as reason for migration and citizenship, variables critical to the data on family migration that have been drawn upon in this briefing.

A limitation of Home Office data on visas issued and border admissions is that they do not show how many of the British sponsors of family unification migrants were themselves foreign-born. A Home Office (2011) investigation into this question was limited by small sample sizes, and not all cases were selected randomly. The Home Office (2011) examined a selection of 531 records of marriage-related migration through marriage in 2009, concerning applicants from selected countries. For those that were randomly selected (according to Migration Observatory correspondence with the Home Office), there was notable variation in the percentages of sponsors who were themselves foreign-born, from 11% of people sponsoring American spouses, to 67% of those sponsoring Bangladeshi spouses.


Related material

Research for this briefing was supported by the Joseph Rowntree Charitable Trust

Next Update

April 9, 2020


Peter William Walsh

Download Briefing

Press Contact

If you would like to make a press enquiry, please contact:

Rob McNeil

+ 44 (0)7500 970081

 Contact Us 


This Migration Observatory is kindly supported by the following organisations.

  • University of Oxford logo
  • COMPAS logo
  • Esmee Fairbairn logo
  • Barrow Cadbury logo
  • Paul Hamlyn Foundation logo