Migration Statistics Quarterly Report: May 2017

• Net long-term international migration was estimated to be +248,000 in 2016, down 84,000 from 2015 (statistically significant); immigration was estimated to be 588,000 and emigration 339,000.

• The net migration change was driven by a statistically significant increase in emigration up 40,000 from 2015, mainly EU citizens (117,000, up 31,000 from 2015) and a decrease of 43,000 in immigration (not statistically significant).

• EU8 citizens have partly driven the changes with a fall in immigration (down 25,000) to 48,000 and a rise in emigration (up 16,000) to 43,000 in 2016 (both statistically significant changes); this resulted in the smallest net migration estimate (+5,000) for the EU8 since joining the EU in 2004.

• Work remains the most common reason for international migration with 275,000 people immigrating to work in 2016 (down 33,000 from 2015 (not statistically significant)); the majority (180,000) had a definite job (similar to 2015) but fewer people immigrated looking for work (95,000, a statistically significant decrease of 35,000 from 2015).

• Long-term immigration to study (136,000 in 2016) saw a statistically significant decrease of 32,000 from 2015, this largely reflects a decrease reported last quarter; however, the numbers of visas issued over the same period to non-EU students for 12 months or more was 141,248, a rise of 3%.

• In 2016 more people were emigrating with a definite job (116,000) than in 2015 (up 17,000, statistically significant); the estimated number of non-British citizens going home to live increased from 29,000 to 52,000 in 2016, this was largely driven by EU citizens, an increase of 21,000 to 43,000, of who around half were EU8 citizens (all statistically significant increases).

• A total of 9,634 people were granted asylum or an alternative form of protection in year ending (YE) March 2017, a grant rate of 32%. In addition, 5,453 people were granted humanitarian protection under the Syrian Vulnerable Persons Resettlement Scheme in YE March 2017.

The reporting period for the International Passenger Survey (IPS) data in this release is year ending (YE) December 2016 (referred to as 2016 in this report) and therefore includes 6 months of data following the EU referendum. Entry clearance visa, National Insurance number (NINo) and labour market data are available to the end of March 2017. ONS data covering a full year after the referendum (YE June 2017) will be available in November 2017. Details on when all ONS outputs will cover the referendum period are published on the Visual.ONS website.

ONS migration statistics use the UN recommended definition of a long-term international migrant: “A person who moves to a country other than that of his or her usual residence for a period of at least a year (12 months), so that the country of destination effectively becomes his or her new country of usual residence.” We include students in our international migration statistics in line with this internationally agreed definition. Immigration and emigration estimates are available broken down by reason for migration including “formal study”.

There is a difference between the long-term IPS estimates for immigration to study and the numbers of long-term student visas issued by the Home Office in 2016. There are a range of potential reasons why the data sources might differ, such as: timing, stated intentions and length of stay, sampling variability in the survey and if a visa was not used or not. See the section on study for more details.

Long-Term International Migration (LTIM) estimates are based on data from the IPS, with adjustments made for asylum seekers, non-asylum enforced removals, people resettled in the UK under resettlement schemes, visitor and migrant switchers and flows to and from Northern Ireland. LTIM estimates are used where available. Estimates of citizenship by reason for migration are based solely on IPS data. In these cases the IPS totals will not match LTIM totals, but will give a good measure of magnitude and direction of change. Figures relating to visas are produced by the Home Office and include long-term and short-term migrants and their dependants; the Department for Work and Pensions’ (DWP’s) National Insurance number (NINo) registrations to adult overseas nationals also include long-term and short-term migrants.

DWP statistics on NINo allocations to adult overseas nationals and nationality at point of NINo registration of DWP working age benefit recipients (NINo benefit claimants) have had their National Statistics status suspended, see the February Migration Statistics Quarterly Report (MSQR) for further details. ONS will continue to include statistics and commentary on NINo data currently published as Official Statistics.

Here is a quick overview of six of the main terms used throughout the report. More can be found in our First Time User Guide, Glossary and List of Products article.

Immigration

This refers to those people coming into the UK. Our estimates sometimes refer to EU or non-EU immigrants, but the estimates of total immigration include EU, non-EU AND British immigrants.

Emigration

This refers to those people leaving the UK and, as with immigration, totals include EU, non-EU AND British emigrants.

Net migration

Net migration is the difference between immigration and emigration. If immigration is higher than emigration, meaning more people come to the UK than leave the UK, then net migration will be a positive (+) number. If more people leave the UK than come to the UK then net migration will be a negative (-) number.

Figure 1: EU groupings used within UK migration statistics, 2017

Source: Office for National Statistics

Notes:

1. There are 28 countries in the European Union (EU28) as at 1 January 2017. For the purposes of our statistics we do not include the UK within the EU15. EU15 joined before 2004, EU8 in 2004, EU2 in 2007 and for EU Other, Malta in 2004, Cyprus in 2004 and Croatia in 2013.

Download this image Figure 1: EU groupings used within UK migration statistics, 2017

.png (216.4 kB)

EU2 countries joined the EU on 1 January 2007. Between 2007 and 2013, in the first 12 months of stay, working EU2 nationals were subject to restrictions on the types of work they could undertake in the UK. These restrictions were lifted on 1 January 2014.

EU8 nationals previously had restrictions on their rights to work and were required to register under the worker registration scheme, but these restrictions were lifted from 1 May 2011.

EU15 statistics exclude British citizens (and as a result in some other publications the EU15 are referred to as the EU14).

The net migration estimate (the difference between immigration and emigration) for the year ending (YE) December 2016 was +248,000 and has a 95% confidence interval of +/- 41,000, compared with +332,000 (+/-38,000) in YE December 2015 (Table 1). This difference was statistically significant and is the lowest recorded net migration estimate since YE March 2014 (Figure 2).

In 2016, immigration was estimated to be 43,000 lower than the previous year whereas emigration was 40,000 higher. This resulted in net migration for 2016 being estimated to be 84,000 lower than the previous year. The annual changes in net migration and emigration were statistically significant, whereas the change in immigration was not statistically significant in this period (Table 1 – significance is denoted by an asterisk).

The ONS online interactive timeline provides annual totals and additional commentary from 1964 to 2015 to show the longer-term context, with the underlying data available for download.

The current level of net migration for EU citizens saw a statistically significant decrease compared with 2015 and is once again below the net migration estimate for non-EU citizens (though both are lower than 2015).

The majority of the fall in net migration is due to a statistically significant drop for EU citizens, down 51,000 from 2015 to +133,000 in 2016 (Table 2 – statistically significant changes are denoted with an asterisk). This decrease was driven by a statistically significant decrease of 41,000 for EU8 citizens to +5,000, the smallest net migration estimate for EU8 citizens since joining the EU in 2004. Net migration for British citizens was -60,000 in 2016 compared with -40,000 in 2015 (not statistically significant).

More detailed estimates over time can be found in Table 1 in the accompanying datasets.

The immigration estimate for the year ending (YE) December 2016 was 588,000, with a confidence interval of +/-34,000, compared with 631,000 (+/-33,000) in YE December 2015. This difference was not statistically significant. However, this is the lowest recorded estimate since YE June 2014.

In 2016 the total immigration estimate comprises of 264,000 non-EU citizens, 250,000 EU citizens and 74,000 British citizens. The immigration estimates dropped from 2015 for each of these citizenship groups but the differences were not statistically significant.

EU citizens

Long-term immigration of EU citizens rose from 2012, levelled off from 2015 and has seen a slight decline in the last two quarters. It is too early to tell if this trend will continue (Figure 4). Figure 5 shows how the trends in EU immigration vary between the three main EU groups: EU15, EU8 and EU2 immigration was estimated to be 132,000, 48,000 and 67,000 respectively.

EU8 immigration showed a statistically significant decrease of 25,000 compared with the previous year which is the lowest recorded estimate since the EU8 countries joined the EU in 2004 and is also driving the EU estimate decrease. There was little change for EU15 and EU2 immigration.

Non-EU citizens

Immigration of non-EU citizens was estimated to be 264,000, down from 279,000 the previous year. This difference was not statistically significant.

Nationality information on individuals, who are subject to immigration controls, coming from non-EU countries, is provided by visa data published by the Home Office. Recent trends in visas granted have provided a good leading indicator for trends in long-term non-EU immigration. The total number of visas (excluding visitor and transit visas) granted in the YE March 2017 (565,180) increased from YE March 2016 (up 34,112 or 6%). There were increases for Chinese nationals (up 11,912 or 13%) and falls for Nigerian nationals (down 2,291 or 16%) and Australian nationals (down 1,502 or 7%), with the highest numbers of visas being granted to Chinese and Indian nationals.

There are a range of potential reasons why the long-term immigration estimates from the International Passenger Survey (IPS) may differ from figures for visas granted, including:

• visa data will include shorter visas (for less than 12 months)

• differences in coverage, for example differences between stated intentions and length of stay

• sampling variation in the IPS

• timing differences between when visas are granted and when an individual travels

• visa and admissions data can include dependants recorded in different subcategories from the IPS

For more information see the Home Office Immigration Statistics, January to March 2017 bulletin.

British citizens

Long-Term International Migration (LTIM) estimates show that the estimated number of British citizens immigrating to the UK was 74,000 in 2016 compared with 84,000 in 2015. This difference was not statistically significant. These numbers include British citizens returning to the UK after living abroad for a year or more; and British citizens who were born abroad and may be coming to the UK for the first time.

More detailed estimates over time can be found in Table 1 in the accompanying datasets.

International Passenger Survey (IPS) data show that, of those whose main reason for long-term migration was work-related, the majority (61%) were EU citizens, whereas of those whose main reason for migration was study, the majority (70%) were non-EU citizens. The difference in reason for immigration is likely to reflect the differing rights of EU and non-EU citizens to migrate to the UK as well as the impact of government policies and other factors (such as economic conditions in origin countries).

Work remains the most commonly stated reason for immigration to the UK

In 2016, 275,000 (47%) people immigrated for work, compared with 308,000 in 2015 (Figure 6). This difference was not statistically significant. However, this is the lowest recorded estimate since YE September 2014.

The majority of those arriving to work, 180,000 (65%) had a definite job to go to, similar to the 178,000 in 2015. The number arriving looking for work had a statistically significant decrease of 35,000 to 95,000 compared with 130,000 in 2015.

IPS data showed that the decrease in the number of people arriving looking for work was driven by EU citizens. Estimates showed a statistically significant decrease from 77,000 in 2015 to 55,000 in 2016. Half of this decrease was due to the numbers of EU8 citizens falling from 27,000 to 16,000 in 2016 (a statistically significant decrease).

Of EU immigrants arriving for work-related reasons (160,000), 34% arrived looking for work in 2016 compared with 44% in 2015; in 2016 within each EU citizenship group the percentages were: EU15: 28%, EU8: 44%, EU2: 36%. The number and proportion arriving looking for work for all main citizenship groups fell in 2016, the number for those with a definite job largely remained steady (Figure 7).

IPS estimates show around 61% of immigrants arriving for work were EU citizens, 25% were non-EU citizens and 14% were British citizens.

Of the 66,000 non-EU immigrants who arrived for work-related reasons, 18,000 (27%) arrived looking for work and these are likely to be accounted for by people granted visas for the Youth Mobility Scheme.

An estimated 36,000 British citizens arrived for work reasons. Of these almost half (17,000) arrived looking for work.

Figure 7: IPS estimates of immigration to the UK for work-related reasons by citizenship, year ending December 2013 to year ending December 2016

Source: Office for National Statistics, International Passenger Survey

Notes:

1. Figures for 2016 are provisional. In this chart provisional estimates are represented by a break in the series. Rolling year data are produced quarterly and are for year ending (YE) March (q1), YE June (q2), YE September (q3) and YE December (q4).
2. Peaks are shown with the appropriate YE data.
3. Note that the EU total given in published tables and the text of this report includes Malta, Cyprus and Croatia, so will be slightly different from the total of the three EU charts shown in Figure 7.
4. Figures may not sum due to rounding.

Download this image Figure 7: IPS estimates of immigration to the UK for work-related reasons by citizenship, year ending December 2013 to year ending December 2016

.png (33.1 kB) .xls (30.2 kB)

More detailed estimates over time can be found for LTIM in Table 2 and for IPS in Table 3 in the accompanying datasets.

Work-related visas (non-European Economic Area nationals)

In the YE March 2017, the number of sponsored skilled work (Tier 2) visas, including dependants, rose by 2% to 94,000. Tier 5 youth mobility and temporary visas were 4% lower at 42,000. In all work categories there were 164,168 work-related visas (including dependants) granted in the YE March 2017, broadly the same level as previously.

Indian nationals accounted for 58% of total skilled work visas granted (53,863 of 93,566), with USA nationals the next largest nationality (9,255 or 10% of the total). The information technology sector sponsored 42% of skilled work visa applications, followed by professional, scientific and technical activities (18%) and financial and insurance activities (12%).

Home Office Immigration Statistics show that IPS long-term immigration estimates for work and formal study among non-EU nationals have broadly followed the same long-term trends as visas granted for work and study. IPS estimates will be lower than the visa figures because the IPS estimates exclude those individuals who state that they intend to stay for less than 1 year and there will be other differences in categorisation. Furthermore, the dependants of those granted a visa to work or study are included in the work and study visa figures, whereas the reason for migration for such individuals, in most cases, are likely to be recorded as accompanying or joining others by the IPS. Other differences may be due to sampling variation in the IPS and other potential sources of bias in IPS data. For more information see the Home Office Immigration Statistics: user guide.

How many adult overseas nationals register for a National Insurance number (NINo)?

National Insurance number (NINo) allocations to adult overseas nationals and nationality at point of NINo registration of DWP working age benefit recipients (NINo benefit claimants) have had their National Statistics status suspended until work to improve their limited supporting guidance and overall public value is undertaken. The full assessment can be found on the UK Statistics Authority website: report on the reassessment of the NINo statistics.

NINo registrations and IPS estimates should not be directly compared. NINo figures include short-term migrants and the figures are based on the recorded registration date on the National Insurance Recording and Pay As You Earn System (NPS) (after the NINo application process has been completed) and so should not be used as a direct measure of when a person migrated to the UK. As such, there will be differences between the NINo registration data shown in Table 4, and the estimates of long-term international migration. More detailed information on these differences is discussed in the publication from 12 May 2016.

In YE March 2017, there were 593,000 National Insurance number (NINo) registrations by EU nationals (down 6% on the previous year) and 191,000 by non-EU nationals (down 2% on the previous year).

As with the LTIM estimates, NINo registration numbers reflect the rise in EU immigration between 2012 and 2015. The sharp rise in registrations to EU2 nationals in 2014 to some extent mirrors the long-term migration trend over an extended period. However, in the last year, the growth in EU2 registrations has been offset by a fall in registrations from other EU member states.

Table 5 shows the top registrations for individual EU and non-EU nationalities.

How many overseas nationals are working in the UK?

Labour market statistics are a measure of the stock of people working in the UK and are not designed to provide a measure of migration flows. As a result, changes in the labour market stock cannot be directly compared with long-term net migration.

Of the total in employment in January to March 2017, 88.9% were British nationals, 7.3% were EU nationals and 3.9% non-EU nationals. These compare with 89.4%, 6.8% and 3.8% respectively, in January to March 2016, showing a small proportional difference.

There have been increases for all nationality groupings in employment in January to March 2017, compared with the same quarter for the previous year (Table 6).

The growth in overall employment over the last year was 385,000. Of this, 46% can be accounted for by growth in employment for British nationals, 45% by growth in employment for EU nationals with the remaining 9% accounted for by non-EU nationals.

Note that these growth figures relate to net changes in the number of people in employment. They do not indicate the proportion of new jobs that have been filled by UK and non-UK workers.

Immigration for study

The Long-Term International Migration (LTIM) estimate for immigration to study was 136,000 in 2016 (23% of total LTIM) and saw a statistically significant decrease of 32,000 from 2015, this largely reflects a decrease reported last quarter.

Home Office visa data provide additional detail for non-EU migrant students¹. In YE December 2016 the number of visas issued to non-EU students for 12 months or more was 141,248, a rise of 3%.

There is a notable difference between the long-term International Passenger Survey (IPS) figures for immigration to study and long-term student visa data for 2016. There are a range of potential reasons why the data sources differ such as: timing, stated intentions and length of stay, sampling variability in the survey and if a visa was used or not. Further comparison of the IPS and Visa data is described in section 6 ”Comparability of data sources on student migration” of the ONS publication International student migration what do the statistics tell us? and the Home Office Immigration Statistics: user guide.

Long-term international student migration has a marked seasonal pattern with a large proportion of people immigrating to study in the July to September quarter of the year, particularly for those who will start the academic year in September. Changes in immigration for study are more likely to be first seen in the YE September reporting year as reported in the February Migration Statistics Quarterly Report (MSQR). It is too early to tell if the 2016 IPS student estimate is an indication of an emerging trend as it is based on only two quarters’ of the latest academic year data and we are aware of the increase seen in the number of study visas issued to non-EU citizens. We will continue to monitor the trends and compare against other sources such as university admissions data. The current analysis being undertaken by ONS and the Home Office will also help to understand such differences – International migration data and analysis: Improving the evidence.

IPS estimates show 70% of those immigrating to study for more than 12 months were non-EU citizens, 24% were EU citizens and the remaining 5% were British citizens. In 2016 the estimate of non-British citizens immigrating to the UK to study for 12 months or more was 124,000, a statistically significant decrease of 24,000 from 148,000 the previous year. In the IPS this was driven by citizens of Asian countries who made up 52% (64,000) of non-British citizens immigrating to study, a statistically significant decrease of 16,000 from 80,000 in 2015.

In the YE March 2017, total sponsored visa applications to study from non-EU nationals (main applicants only) rose by 2% to 202,042². This included a rise in visa applications to study for higher education (to 168,591, up 3%); within this total, applications from Russell Group universities rose by 7%. There was also a rise for English language schools (to 2,955; up 11%) but falls for:

• further education sector and other educational institutions (to 14,534, down 6%)

• independent schools (to 13,379, down 1%)

Table 7 shows that more than one-third of non-EU study visas were granted to Chinese students (up by 7,000 or 10% compared to a year earlier), and together the 3 largest nationalities made up around a half of the non-EU foreign students granted visas in the year ending March 2017.

More detailed estimates over time can be found for LTIM in Table 2 and for IPS in Table 3 in the accompanying datasets.

For more information on immigration to the UK for study visas, see the Home Office Immigration Statistics.

Immigration to accompany or join others

The third most common reason for migrating to the UK is to accompany or join others. This category can include people who are arriving to marry a UK resident as well as people coming as a family member of another migrant coming for other reasons, such as to work³. In 2016, of all long-term immigrants, 14% (85,000) arrived in the UK to accompany or join others (Figure 6), the highest figure since 2014.

IPS data show a statistically significant increase in the estimates to accompany or join others for non-British citizens from 61,000 in 2015 to 77,000 in 2016.

Home Office entry clearance visa statistics show that the combined total number of family-related visas granted, including European Economic Area (EEA) family permits granted, and visas granted to dependants of other visa holders (excluding visitors) increased by 2,267 (2%) to 136,787 in the year ending March 2017. Note these figures include both long-term and short-term migrants and are therefore not directly comparable to the IPS.

Further information on visas granted for family reasons is in the Home Office Immigration Statistics.

Immigration for other reasons

The number of immigrants arriving in the UK for other reasons, increased from 53,000 to 59,000 in 2016, this was not a statistically significant change. This includes asylum seekers (as measured by Home Office data), volunteers, religious pilgrims and those coming for medical treatment. From 2015 the LTIM adjustments additionally include the number of people resettled in the UK under the various resettlement schemes, as reported by the Home Office.

Asylum and resettlement

There were 36,846 asylum applications (including dependants) in YE March 2017, a decrease from YE March 2016 (down 5,472 or 13%).

The number of applications is considerably below the peak in 2002 (103,081) and the UK received the fifth-highest number of asylum applications of the 28 EU member states in YE March 2017.

A total of 9,634 people were granted asylum or an alternative form of protection in YE March 2017, an overall grant rate of 32% at initial decision. In addition, 5,453 people were granted humanitarian protection under the Syrian Vulnerable Persons Resettlement Scheme in YE March 2017 (7,307 since the scheme began in January 2014).

Table 8 shows the five nationalities with the highest number of applications for asylum to the UK in YE March 2017.

There were 1,269 asylum applications from Syrian nationals (including dependants) in YE March 2017, the tenth most common nationality amongst asylum applicants and a decrease of 54% compared with the previous year (2,731). There were 1,507 grants of asylum or an alternative form of protection (including dependants) to those giving Syrian as their nationality in YE March 2017, a grant rate of 85% at initial decision, in addition to the 5,453 granted humanitarian protection under the Syrian Vulnerable Persons Resettlement Scheme.

Further detail is provided in Home Office Immigration Statistics.

Settlement in the UK

There were 57,111 people granted permission to stay permanently in YE March 2017, 28% fewer than the previous year. The majority of the decrease was accounted for by falls in work grants (-17,634). These decreases are likely in part to reflect changes in the Immigration Rules.

For more information on settlement in the UK, see the Home Office Immigration Statistics.

Notes for: Why do people move to the UK?

1. There are a range of reasons why LTIM and IPS figures for long-term migrants for study may be different from Home Office figures for study visas granted. For example, the Home Office visa statistics include short-term students migrating to the UK for less than 1 year, but only record students who are non-EEA nationals. LTIM estimates are based on a sample survey and therefore may fluctuate due to sampling variability.

2. Note that this figure relates to Certificates of Acceptance to Study (CAS) data used in the application process for a visa, which will differ from the number of study visas granted, although the number is likely to be similar as sponsors are expected to only issue CAS to legitimate students.

3. More detail can be found in the Home Office Immigration Statistics: user guide.

The emigration estimate for the year ending (YE) December 2016 was 339,000 with a confidence interval of +/-23,000, compared with 299,000 (+/-20,000) in YE December 2015. This difference was statistically significant and is the highest recorded estimate since YE September 2012 but still below the previous highest recorded estimate of 427,000 in 2008.

Emigration from the UK for 2016 comprises of (Figure 9):

• 134,000 British citizens – 40% of all emigrants and the highest recorded estimate since YE December 2014

• 117,000 EU citizens – a statistically significant increase of 31,000 and the highest recorded estimate since YE June 2009

• 88,000 non-EU citizens – steady over the last 2 years

The increase in EU citizens emigrating was partly driven by a statistically significant increase of 16,000 for EU8 citizens compared with the previous year to 43,000. More detailed estimates over time can be found in Table 1 in the accompanying datasets.

Why are people leaving the UK?

In 2016, work-related reasons remain by far the most common reason given for emigration for all citizen groups, accounting for just over half (179,000) of emigrants. Of these emigrants 65% had a definite job to go to, a statistically significant increase of 17,000 to 116,000 compared with 2015; the estimates for those emigrating looking for work saw little change.

International Passenger Survey (IPS) data show that non-EU citizens remain the most likely group to emigrate for work reasons, 70% (58,000) in 2016; compared to 43% (48,000) of EU citizens and 54% (65,000) of British citizens.

Those British citizens emigrating for work reasons are more likely to have a definite job to go to, 80%, compared with 58% of EU and half of non-EU citizens.

IPS data shows that there was a statistically significant increase in the estimated number of non-British citizens going home to live, from 29,000 to 52,000 in 2016. This increase was largely accounted for by EU citizens going home to live, an increase of 21,000 to 43,000; nearly half of these EU emigrants were EU8 citizens whose numbers rose by 11,000 to 20,000 (both statistically significant increases).

Long-Term International Migration (LTIM) estimates report there were 26,000 people emigrating to accompany or join others and 24,000 for formal study in 2016, similar to the last three years.

More detailed estimates over time can be found for LTIM in Table 2 and for IPS in Table 3 in the accompanying datasets.

Emigration from the UK by previous main reason for immigration

The IPS asks emigrants about their main reason for originally coming to the UK. Detailed estimates derived from this IPS question can be found in Table 4 in the accompanying datasets.

Of the 315,000 individuals who emigrated from the UK, 89,000 (28%) were “new” long-term emigrants (individuals who had not previously lived away from the UK for 12 months or more). The remaining 226,000 (72%) were long-term emigrants who had immigrated to the UK in a previous year. This was a statistically significant increase of 32,000 compared with the previous year (194,000).

This included 105,000 who originally came to work in the UK and of these, 74,000 (70%) were EU citizens, a significant increase of 28,000 compared with the previous year (46,000). Within the EU grouping, 33,000 were EU8 citizens, a statistically significant increase of 16,000 compared with the previous year (17,000 in 2015). Of the 74,000 EU citizens who emigrated having originally arrived to work, 59,000 (80%) had arrived with a definite job to go to, a statistically significant increase of 28,000 compared with the previous year.

There were also an estimated 63,000 who emigrated from the UK and had originally immigrated to study. This estimate provides information on how many former international students emigrate at the end of their studies but does not provide the complete picture on what all international students do when they finish their studies, since not all students will emigrate. An ONS report on International student migration – what do the statistics tell us, was published in January 2016. It identified a number of potential reasons for the difference between students immigrating and those who were emigrating having previously arrived for the purpose of study. These were:

• students staying longer than initially expected and obtaining extensions of stay in the UK, whether as a student or in other categories such as skilled work

• students finishing their courses and overstaying their visas

• the IPS not completely recording student flows, either due to sampling or non-sampling errors (such as not responding to the survey or responding incorrectly)

• when student migration is in a period of growth, as it generally has been in the UK since the 1990s, then student numbers will make a positive contribution to net migration during that period because the numbers arriving in any year will tend to be larger than the numbers leaving (reflecting the lower number of previous years’ arrivals); if student immigration were to decline, the opposite will be true

ONS is working closely with other government departments to better understand a number of data sources related to student migration and what insight they can provide on what international students do after their studies. The most recent International student migration research update was published in April 2017.

There is a lot more detail within the charts and tables found in the accompanying datasets alongside this report, including breakdowns of the statistics by main reason and by citizenship grouping. Definitions of the main terms used in the report can be found in section 2, towards the end of the report you can find links to further information, and a note on statistical uncertainty.

If you are new to migration statistics, you might find it helpful to read our “International Migration Statistics First Time User Guide, Glossary and List of Products”. This contains information on other measures of international migration as well as source and topic specific reports, for example:

• Short-Term International Migration estimates for England and Wales, year ending June 2015 were published in May 2017 and cover migrants coming to or leaving the UK for 1 to 12 months and 3 to 12 months

• Population of the UK by Country of Birth and Nationality, 2015 was published in August 2016 and covers the number of people resident in households in the UK by country of birth and nationality in 2015

• International immigration and the labour market, UK: 2016 was published in April 2016 and studies the labour market characteristics of UK, EU and non-EU nationals in the UK labour market in 2016

• International migration and the changing nature of housing in England – what does the available evidence show? was published in May 2017 and provides information about international migration, population change and changes in housing trends in England.

There are many sources of official statistics that measure the number and characteristics of international migration into and out of the UK (flows) as well as the migrants who have settled in the UK (stocks). Taken together they provide a rich picture of migration in the UK. It is important to understand that these sources measure different things: some measure flows, some measure stocks, some measure workers, some students and some only measure the characteristics of those migrating from outside the EU. Each source is valuable in its own right in measuring particular aspects of international migration.

We have produced several notes looking at how some of these sources differ such as: the note on International Student Migration published 22 January 2016, the most recent International student migration research update was published in April 2017; IPS and National Insurance number reconciliation note published 12 May 2016; and the IPS and Annual Population Survey (APS) comparison paper published 1 December 2016. A separate summary note takes some of the main themes from the notes to help users understand that the differences between the sources are driven by differences in definitions and coverage.

There is now much more interest in the characteristics of migrants who reside in the UK and the impact they have on the economy and society. There are sources of data that shed some light on this but these only go so far in meeting the needs of the public, media and policymakers. We are collaborating and data sharing across government to improve the information that is currently available to understand migration as described in our future work programme – International migration data and analysis: Improving the evidence.

There is considerable interest in migration statistics both nationally and internationally, and they are a fundamental component of our mid-year population estimates. These are used for planning and service delivery, resource allocation and managing the economy. Additionally, these statistics are reported regularly by the media to assess the government’s progress against its ambition to reduce levels of net migration to sustainable levels.

If you would like to subscribe to our newsletter, please send an email to pop.info@ons.gov.uk with the subject title “Subscribe to ONS Population Statistics Newsletter”, or you can also follow our statistician @PaulVickers_ONS on Twitter for the latest population statistics news and updates and join in the conversation.

We have updated the structure and some of the headings of this bulletin as we move towards the new ONS Statistical bulletin guidance. Content required to support interpretation of the data continues to be included.

Data tables including “old country groupings” which described old and new commonwealth country groups have been removed from the published accompanying datasets. These “old country groupings” data will continue to be included in the annual data tables next published in November 2017.

In line with the new ONS Statistical bulletin guidance and following feedback from the Consultation on International Migration Statistics Outputs ONS will move to shorter more streamlined Migration Statistics Quarterly Report (MSQR) publications in 2017. We will also work to redesign the MSQR and migration products to meet user needs of developing a more coherent story of international migration in the UK. Regardless of future changes all the data previously published in the MSQR will still be made available for users to download and explore.

ONS will work towards including further labour market data by country of birth and nationality in the August 2017 MSQR release.

The Long-Term International Migration Quality and Methodology Information document contains important information on:

• the strengths and limitations of the data and how it compares with related data

• users and uses of the data

• how the output was created

• the quality of the output including the accuracy of the data

For more detailed information on how our migration statistics are used, along with information on their strengths and limitations, please see the articles Quality and Methodology Information for Long-Term International Migration (LTIM) releases and Long-Term International Migration Estimates Methodology, as well as the note on sampling and uncertainty in this bulletin.

A note on sampling and uncertainty

Surveys gather information from a sample of people from a population. Using the International Passenger Survey (IPS) as an example, the population is passengers travelling through the main entry and exit points from the UK including airports, seaports and the Channel Tunnel. The estimates produced are based on only one of a number of possible samples that could have been drawn at a given point in time. Each of these possible samples would produce an estimated number of migrants. These may be different from the true value that would have been obtained if it were possible to ask everyone passing through about their migration intentions. This is known as sampling variability.

The published estimate is based upon the single sample that was taken and is the best estimate of the true value based on the data collected. However, to account for sampling variability, the estimates we publish include a “95% confidence interval”.

The confidence interval is a measure of the uncertainty around the estimate. Confidence intervals become larger (meaning there is more uncertainty) for more detailed estimates (such as citizenship by reason for migration). This is because the number of people in the sample who have these specific characteristics (for example, EU8 citizens arriving to study) is smaller than the number of people sampled in higher level categories (such as the total number of EU citizens arriving to study). Where possible, it is better to use the highest level breakdown of data available.

We use the widely accepted 95% confidence interval, meaning that over many sampling repetitions under the same conditions, we would expect the confidence interval to contain the true value 95 times out of 100. Equivalently, we can say that there would be a 1 in 20 chance that the true value would lie outside of the range of the 95% confidence interval.

Confidence intervals are reported in the accompanying datasets. Users are advised to be cautious when making inferences from estimates with relatively large confidence intervals. For immigration and emigration estimates where the lower confidence interval is below zero users should assume the estimate is above zero. Estimates from a survey could change from one period to the next simply due to sampling variability. In other words, the change may be due to which individuals were selected to answer the survey, and may not represent any real-world change in migration patterns.

Statistical tests can be used to determine whether any increases or decreases that we see in the estimates could be due to chance, or whether they are likely to represent a real change in migration patterns. If the tests show that the changes are unlikely to have occurred through chance alone, and are likely to reflect a real change, then the change is described as being “statistically significant”. The usual standard is to carry out these tests at the 5% level of statistical significance. This means that in 1 out of 20 differences identified as statistically significant, the difference may not be the result of a real change.

For information on the accuracy of these statistics, comparing different data sources, and the difference between provisional and final figures, please see the MSQR - information for users.

Revisions to net migration estimates in light of the 2011 Census

In April 2014, we published a report examining the quality of international migration statistics between 2001 and 2011, using the results of the 2011 Census. A main finding of the report was that, over the 10-year period, annual net migration estimates were a total of 346,000 lower than total net migration implied by the 2011 Census. However, the report also showed that the quality of international migration estimates improved following changes made to the IPS in 2009.

Within the report, we published a revised series of net migration estimates for the UK. Published tables have been updated on our website to include the revised estimates. The report, a summary and guidance on how to use these revised figures are available on our website. You should be aware that no revisions were made to separate immigration and emigration estimates at the time the net migration estimates were revised – ONS analysis of the differences between the 2011 Census and population estimates suggest immigration figures could have been higher in the mid-2000s as a result of migration from eastern Europe.