This report is the third of three migration trends reports to be produced as part of the Migration Services in Wales project, providing quantitative analysis on migration in Wales. The report provides an overview of migration flows and population trends in Wales over time, with an emphasis on local area variations as well as national comparisons with England. The first and second reports in this series focused on migrants in the Welsh labour market and the characteristics of migrant households in Wales, respectively.
The first section of the report outlines key statistics on the characteristics of people born outside the United Kingdom who live in Wales using data from the 2011 Census. The discussion covers the shares of migrants from different countries of birth, years of arrival to the UK, intentions of staying as a resident for longer than a year and knowledge of the Welsh language. In the second section, the analysis continues with long-term international migration flows to and from Wales between 2004 and 2014. This discussion includes information on the number of inflows and outflows among non-British nationals arriving to the UK for a minimum of twelve months with Wales as destination (or prior residence for outflows). This section also describes trends in the share of non-British inflows to Wales from the European Union (EU) and the share of all non-British inflows to the UK with Wales as a destination.
The third section focuses on trends in migration flows to and from local areas in Wales using data from the Office for National Statistics (ONS) Local Area Migration Indicators. This data source provides information on internal (within-UK) and international inflows and outflows of people to and from local areas in Wales. The analysis includes a snapshot of international and internal net flows for each local area in 2004, 2010, 2014, and 2015. Net flows refer to the number of people who moved into a local area minus the number of people who left and can give a useful indication of overall trends in migration and local population change.
This briefing focuses on trends in migration flows to and from Wales. To provide this information using reliable statistics, this report draws on the 2011 Census for England and Wales, the ONS International Passenger Survey estimates of long-term international migration, and the ONS Local Area Migration Indicators suite. See the ‘Evidence Gaps and Limitations’ section of the briefing to understand challenges associated with these data sources.
There is no consensus on a single definition of 'migrant' and depending on the data source, this discussion defines migrants as those who are born outside the UK or those who do not hold British nationality/citizenship. Different analyses use a variety of definitions of 'migrant', including (1) country of birth, (2) nationality, (3) passports held, (4) length of stay, (5) reason for migration and (6) being subject to immigration control. For example, not all foreign-born people are also foreign nationals; some foreign nationals may have lived in the UK for decades while others may have resided in the UK for only a year. Perhaps most importantly, not all non-UK born UK residents are subject to immigration control, such as nationals of EEA countries and their dependants, or people born abroad to UK citizen parents. There is no information on place of first arrival for those born outside of the UK within the data sets used. Those born outside of Wales but within the UK are considered UK-born in this analysis.
In UK immigration debates, mobile EU citizens are considered as a distinct group because they can move freely within the EU. The UK or devolved governments cannot limit their rights to live and work in the United Kingdom in the same way as for migrants from outside the EU. The boundaries of the EU have changed significantly since 2004. A brief recent timeline of EU expansions is as follows:
Members pre-2004: EU15 (Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden, United Kingdom)
2004 expansion: A8 plus Malta and Cyprus (Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia, Slovenia)
2007 expansion: A2 (Bulgaria and Romania)
2013 expansion: Croatia (after the 2011 Census took place)
To facilitate the analysis and allow for more consistent classifications, breakdowns by country of birth include: a) EU14 (EU15 except UK), b) Accession (joined 2004-2013), and c) other non-EU countries grouped together.
The Census is the most complete source of information about the Welsh population. It is particularly useful for obtaining demographic information for small geographical areas. The census is based on a count of people and households, with efforts to include all residents, and supplemented by a survey to detect and estimate those who are missed at first. In England and Wales the most recent census took place on the 27th of March 2011 and was conducted by the ONS. Statistics based on the census refer to usual residents of England and Wales. A usual resident is defined as anyone in the UK on census day who had stayed or intended to stay in the UK for 12 months or more (or were outside the UK but had a permanent UK address and intended to stay outside the UK for less than 12 months). See the Migration Observatory video interview of Peter Stokes, 2011 Census Statistical Design Manager, for further discussion. For more information on the migrant population in Wales, please see the Wales 2011 Census Profile. The Maps section of the Migration Observatory website also provides a wide range of maps at local authority level using 2011 Census data for England and Wales.
Unlike Census data that express the total number of people resident at a given time, population flows express the movement of people to and from an area or country. Inflows reflect the number of people moving into an area; outflows reflect the number of people moving out of that area; and balance/net is the difference between those who came minus those who left. Official estimates on international migration flows to and from the UK are based on the ONS International Passenger Survey (IPS). The IPS is a voluntary survey that collects information from 700,000-800,000 passengers each year at ports and airports (including 4,000-5,500 Long-term International Migrants). This briefing draws on this data source to discuss long-term international migration flows to and from Wales over time. In accordance with the UN definition, a long-term international migrant is a person who is travelling to another country to take up residence for a minimum of 12 months. The IPS estimates do not provide information on first port of entry among long-term international migrants, but ask travellers about their intended destination within the UK and their last region of residence for those who are emigrating. International migrants who stated their intended destination or prior residence as Wales are classified as flows to and from Wales in this report.
Trends in local area migration flows are based on data released in the annual ONS Local Area Migration Indicators Suite. The ONS and statistical offices in Scotland (NRS) and Northern Ireland (NISRA) produce long-term international and internal migration figures at local authority level for the purpose of calculating mid-year population estimates. Since internal flows of people are not recorded formally, internal estimates are based on a variety of administrative registers and other adjustments: the Patient Register (PR), the National Health Service Central Register (NHSCR), the Higher Education Statistics Agency (HESA) data, the Scottish NHSCR, and the Northern Irish medical cards register. For more information about the estimates and methods used see here.
Readers should exercise caution when comparing data points as some of the variation may not represent statistically significant differences or changes. In the first section of the report, explicit comparisons between two figures (e.g. noting that one data point is ‘larger’ or ‘smaller’ than another), have been tested for statistical significance.
The first section of this report reviews key characteristics of people who are born outside the UK and reside in Wales using data from the 2011 Census for England and Wales. This part of the report provides information on the shares of migrants across local areas in Wales, including breakdowns by different regions of birth (EU14, EU Accession, non-EU), years of arrival, intentions to stay in the UK and knowledge of the Welsh language.
According to the 2011 Census for England and Wales, 7% of the population in Wales was born outside the UK compared to 15% in England (see Table 1) 1. In both nations, the majority of foreign born were UK residents the year before the Census took place and intended to stay in the UK for longer than a year.
In Wales in 2011, 11% of the non-UK born were living abroad the previous year compared to 8% in England. Certain groups of migrants, such as students or temporary workers, may account for these differences.
% Population who are non-UK born (migrants)
% Non-UK born who were living in the UK a year ago
% Migrants born in EU14 countries
% EU14 born who arrived after 2004
% Migrants born in EU Accession countries (2004-2011)
% Accession born who arrived after 2004
% Migrants born in non-EU countries
% Non-EU born who arrived after 2004
2011 Census for England and Wales, author calculations using micro-data 5% subsample.
The majority of foreign-born residents in Wales are born outside the EU (60%). This is followed by EU14 born at 23% and Accession born at 17%. Compared with England, however, Wales has a larger share of migrants born in EU14 countries and EU Accession countries, but a smaller share of migrants born outside the EU.
Census statistics also indicate that different groups of migrants have settled in Wales at different periods of time, EU14 having settled earlier, followed by non-EU born and Accession born (see Table 1). In 2011, 27% of EU14 born had arrived after 2004, compared to 46% among those born outside the EU, and 84% of EU Accession born.
On Census day, time of arrival among migrants also varied by local area in Wales. Cardiff, the local authority with the largest overall share of migrants in the population was evenly split. In 2011, 51% of non-UK born residents in Cardiff had arrived after 2004. In Swansea and Wrexham this stood at 58% and 61% of non-UK born residents, respectively. On the other hand, in Conwy and Denbighshire and Neath Port Talbot, local areas with relatively small proportions of resident migrants, about 31% of foreign born had arrived after 2004.
When looking at the proportion of different groups of migrants within the overall migrant population in each local area, Swansea had the largest share of migrants born outside the EU at 71%, while Torfaen and Monmouthshire had the largest share of EU-14 born at 34% (see Figure 2). Cardiff is among the three local areas with the smallest shares of migrants from Accession countries at 11%, along with Isle of Anglesey and Gwynedd (11%) and Rhondda Cynon Taf (8%).
In 2011, about 27% of non-UK born people in Wales lived in a household where no one reported English or Welsh as their main language. At the same time, about 6% of people born outside the UK resident in Wales said they understood Welsh. It is worth noting that this question is likely to capture a general familiarity with the language rather than fluency and everyday usage, since it asks people solely about their ability to understand spoken Welsh, rather than speak, write or read. These people may have learned Welsh recently or may have settled in Wales as children.
Local areas in Wales that have the highest rates in knowledge of spoken Welsh among UK-born residents also have the highest rates among non-UK born residents (see Figure 3). In 2011, 20% of non-UK born people in the Isle of Anglesey and Gwynedd understood Welsh, while the rate was at 13% in Carmarthenshire and 12% in Ceredigion and Pembrokeshire.
The second section of this report focuses on long-term international migration flows to and from Wales among non-British nationals, using data from the International Passenger Survey (IPS) for the years 2000-2014. Long-term international migrants are those who are moving from one country to another to take up residence for more than one year.
In 2014, an estimated 21,000 non-British nationals arrived to the UK for a minimum of one year and reported Wales as their destination within the UK (see Figure 4). Emigration among non-British nationals from Wales to other countries stood at about 3,000 people, leaving a positive migration net/balance of about 19,000 foreign nationals. The latest available calendar year (2014) had the highest positive net estimate between 2000 and 2014.
Notes: ONS IPS estimates by destination within the UK. Non-British nationals. Totals may not match due to rounding. Immigrants arriving in the United Kingdom are not always sure as to where in the United Kingdom they will eventually settle. From 1999 onwards, IPS data for long-term immigrants has been calibrated against migrant geographical distribution data obtained from the Labour Force Survey.
After the 2008 recession there was a small upward trend in outflows among non-British nationals from Wales and a corresponding downward trend in net flows. From 2012 onwards, however, inflows increased gradually while outflows returned to average levels for that decade (fewer than 5,000 emigrants). For each of the years 2012, 2013 and 2014, net flows were above an estimated 10,000 non-British nationals.
Notes: ONS IPS estimates by destination within the UK. Non-British nationals. Immigrants arriving in the United Kingdom are not always sure as to where in the United Kingdom they will eventually settle. From 1999 onwards, IPS data for long-term immigrants has been calibrated against migrant geographical distribution data obtained from the Labour Force Survey.
The majority of long-term international inflows to Wales are among people who do not hold British nationality. In 2006, 85% of all long-term international inflows to Wales were foreign nationals, similar to the 2008 and 2014 estimates of 84%. Up until 2013, non-EU citizens made up a larger share of non-British inflows to Wales than EU citizens. In 2014, EU citizens represented about half of non-British inflows with Wales as a destination. When looking at the total number of inflows among non-British nationals to the UK, a small proportion reported Wales as their first intended destination, at 4% in 2014.
The third section of this report continues with information on local area migration to and from Wales. Migration flows, along with fertility and mortality are the three main contributors to population change. The ONS releases national and local area estimates of long-term internal and international migration inflows and outflows as part of its series on Local Area Migration Indicators. These estimates are calibrated using various sources of data to improve reliability of statistics at the local level, but remain subject to margins of error. Therefore caution should be taken when comparing differences between local areas or years.
In 2015, long-term international inflows to Wales (all nationalities) represented about 5 people per thousand residents, compared to a figure in England of 10 people per thousand residents (see Figure 6). Cardiff had the highest rate of long-term international inflows in 2015 at 15.5 people per thousand residents. Local areas with higher inflow rates tend to have higher outflow rates as well.
Figure 7 shows the long-term international migration net flows to and from local areas in Wales for the years 2004, 2010 and 2015. For these years, annual long-term international net flows to and from Wales represented less than one per cent of the population. In 2015 Cardiff and Swansea had the highest long-term international migration net flows, at about 1,900 and 1,100 respectively. Monmouthshire (-80), Conwy (-35), and Powys (-24) had a marginally negative international migration balance.
Notes: ONS Local Area Migration Indicators, August 2016. Estimates include all nationalities and correspond to the mid-year equivalent (e.g. 2004 reflects mid-2003 to mid-2004). International migration figures are taken from the components of change used in the mid-year population estimates and will differ from the official estimates of long-term international migration (LTIM) – see here.
Internal migration estimates from the ONS Local Area Migration Indicators express the number of people who move into (or out of) a Welsh local area from another area within the UK. This information does not allow for a distinction between those who move within Wales and those migrating to/from Wales from other areas of the UK (ONS). For example, inflows to Swansea will include people who relocate from Newport, as well as those who relocate from Birmingham. The estimates also refer to people of all nationalities, including British nationals.
As shown in Figure 8 the difference between internal inflows minus internal outflows varies from year to year, especially at the local area level. Overall, in both 2014 and 2015 the internal migration net/balance tended to be negative in larger cities in Wales and positive in some smaller local areas. Ceredigion (-884), Cardiff (-719) and Wrexham (-442) had the highest negative net internal migration in 2015. Bridgend, Neath Port Talbot, and Carmarthenshire were the top three local areas with a positive internal migration balance that ranged from 550 in Carmarthenshire to 666 in Bridgend. It is worth noting that internal long-term inflows and net flows represented less than one per cent of the population of each respective local area.
For Wales as a whole, the net number of internal inflows and outflows between the years 2004 and 2015 fluctuates between positive and negative, but remains below 10,000 people or one per cent of the population of Wales. This is similar to estimates for England on long-term internal migration flows as a share of the population.
Notes: ONS Local Area Migration Indicators, August 2016. Estimates include all nationalities and correspond to the mid-year equivalent (e.g. 2015 reflects mid-2014 to mid-2015). The internal migration figures are derived from those used in ONS's annual mid-year population estimates. Due to changes in the estimation and calibration methods of internal migration flows, comparisons to estimates prior to 2012 are avoided.
The Census is the most reliable source of information about the population. Nonetheless, it remains subject to certain limitations. Firstly, it is only conducted every 10 years. Secondly, there are only a limited number of questions included in the census, limiting the breadth of information available. While the census aims to include the entire population, it does have a certain margin of error. For England and Wales as a whole, the relative confidence interval at the 95% confidence level published by the ONS was 0.15% (83,000 more or less people than the estimate). To give an example at a more local level, for Cardiff’s usually resident population, the relative confidence interval at the 95% level was 1.76%, suggesting the true value of the population is somewhere between 339,999 and 352,181 persons (ONS, 2012). Specific confidence intervals are not currently available for census data relating to country of birth. For further discussion see the data sources and limitations section of the Migration Observatory website.
The information shown in the first section of the briefing is based on a 5% subsample of the 2011 census in micro-data format (safeguarded sample at grouped local area level). It includes more than 2 million observations selected to be representative of census totals. Therefore estimated statistics based on the Census subsample have smaller margins of error compared to other survey based population estimates.
The second section of the report uses the International Passenger Survey (IPS). It is a sample survey, which is voluntary and relies on people outlining their intentions. Migration estimates are based on respondents’ initial intentions, which may not accord with what they do in practice, e.g. in terms of duration of stay, reason for migrating or destination. The coverage of the IPS is not comprehensive, it excludes migration over the land border between the UK (Northern Ireland) and Ireland due to the existence of the Common Travel Area (CTA). It also excludes most asylum seekers and their dependants. The sample is small, estimates for subsets of the sample such as individual nationalities are subject to substantial uncertainty. For more details see the Migration Observatory report on this data source here.
Long-term internal and international migration flows at the local level are part of the ONS annual mid-year population estimates and are produced using a variety of sources and methods. There are different factors that impact the reliability and accuracy of each component (for more information see ONS).
The ONS uses a combination of three administrative data sources as a proxy for internal migration: the National Health Service Central Register, the Patient Register Data Service and Higher Education Statistics Agency (HESA) data. The “ONS considers the annual mid-year estimates consisting of a combination of administrative data to be more complete and it is these estimates which are used in the calculation of mid-year population estimates” (Internal Migration Estimates Methodology Notes, ONS). The data, however, exclude moves into and out of prison and the armed forces. Groups of the population who are less likely to register promptly with a GP when they move may also be missed. Lastly, certain adjustments made using the HESA data are not applied to cross-border flows between constituent countries (England and Wales/Scotland/ Northern Ireland).