Internal Migration, England and Wales: Year Ending June 2013

• There were an estimated 2.71 million residents moving between local authorities in England and Wales between July 2012 and June 2013. This compares with 463,000 international immigrants who arrived in England and Wales and 281,000 international emigrants who left England and Wales during that period

• 53,200 people moved from England and Wales to Northern Ireland and Scotland; 46,800 moved from Northern Ireland and Scotland to England and Wales. This means there was a net internal migration loss for England and Wales of 6,400 people

• The age with the highest proportion of movers was 19. Of all the 19 year olds resident in England and Wales as at mid-2013, approximately 1 in 5 had moved within the UK during the previous 12 months

• There were more female movers than males: 52% of movers were female and 48% were male

• Of the English regions, London had the highest net rate of outflow (6.6 per 1,000 residents as at mid-2012). The region with the highest net rate of inflow was the South West (3.9 per 1,000 residents as at mid-2012)

This is the first time that the Office for National Statistics (ONS) has published internal migration estimates for England and Wales for the year ending June 2013 (that is, the period from 1 July 2012 to 30 June 2013). As in previous years, internal migration is defined as residential moves between local authorities and regions in England and Wales, as well as moves to or from the rest of the UK (Scotland and Northern Ireland). Moves within a single local authority are excluded, as are international moves into or out of the UK.

These statistics are based on a combination of administrative data and represent the best available source of information on internal migration. They have a wide range of uses across central and local government, as well as academia and business. This is discussed in more detail in the ‘Users and Uses of Internal Migration Statistics’ section of this bulletin. Further information on the data sources and methods, and their limitations, is available via the internal migration methodology webpage.

This bulletin summarises the statistics in a range of ways including total moves, moves by age, moves by sex and moves by area. However, users wishing to do their own analysis are referred to the tables accompanying this release. These are described in more detail in the ‘Further Information’ section of this bulletin.

Moves between local authorities in England and Wales

In the year between July 2012 and June 2013 there were an estimated 2.71 million residential moves between local authorities in England and Wales. This compares with 463,000 international immigrants who arrived in England and Wales and 281,000 international emigrants who left England and Wales during that period.

As Figure 1 shows, the number of moves between local authorities in England and Wales has been broadly consistent since the year ending June 2002. The number of moves in the year ending June 2013 was just 40,000 lower than the number in the year ending June 2012.

Figure 1: Total moves between local authorities in England and Wales, years ending June 2002 to June 2013

Source: Office for National Statistics

Notes:

1. The black dotted lines indicate methods changes. The impact of these is discussed in the following paragraphs.

Download this image Figure 1: Total moves between local authorities in England and Wales, years ending June 2002 to June 2013

.png (13.6 kB) .xls (57.3 kB)

The year ending June 2002 was selected as the first year in the series because that is the earliest year for which an adjustment was made to take better account of study-related moves. However, the statistics have two further breaks in consistency:

1. In the year ending June 2009 a local government reorganisation reduced the total number of local authorities in England and Wales from 376 to 348. This meant that from that year onwards any moves between the former constituent parts of the merged authorities were excluded from the statistics. However, the estimated reduction in total moves (based on a comparison using the data for the year ending June 2012) is proportionally small, around 35,000 per year.

2. In the year ending June 2012 new methods were introduced to improve the adjustment for study-related moves. A review of the changes and their potential impact (194.6 Kb Pdf) was published in November 2013. However, although the new methods will have a more substantial impact at local level, especially in local authorities with universities, the impact at national level will be small.

Overall, even allowing for the geographic and methods changes, the difference between the lowest and highest values in the series is probably no more than around 10%. A likely cause of the slight decline in the late 2000s was the recession, which had an impact on people’s incomes, the employment market and, in consequence, the housing market (Campos et al., 2011 (395.6 Kb Pdf)). However, given the overall consistency of the series, it would appear that even the recession did not have a large impact on the number of moves across local authority boundaries.

Cross-border moves

A cross-border move is defined as a move, in either direction, between England and Wales and the rest of the UK (Northern Ireland and Scotland). Figure 2 displays cross-border moves since the year ending June 2002.

In the year ending June 2013 the net number of cross-border moves out of England and Wales was 6,400, which was the highest net outflow since the year ending June 2008. However, it is considerably lower than the typical net outflows of 10,000 or more which were occurring in the middle part of the last decade.

Figure 2 also shows that the overall number of cross-border moves has been lower in recent years than it was earlier in the series. The number of moves into England and Wales from Northern Ireland has varied comparably little, but there have been declines in the number of moves to Northern Ireland, as well as the number of moves both into and out of Scotland.

As with moves within England and Wales, it is likely that economic factors have had an impact on cross-border flows. Another factor likely to have had an effect is the differing (and evolving) policies on university tuition fees in the different parts of the UK, in particular that it has become financially more favourable for students from Scotland to remain in Scotland for study. It is not known whether the forthcoming (September 2014) referendum on Scottish independence, and the associated debate, has had any effect.

This section looks at internal migration in the year ending June 2013 by age, sex and area.

The new method of student adjustment introduced in the year ending June 2012 will have had more of an impact on moves by age, sex and area than on internal migration as a whole (ONS, 2013a (194.6 Kb Pdf)). Also since that year estimates have been based on age at mid-year rather than age at date of move, which will particularly affect statistics for age 0 and student ages. Because of these changes detailed time series comparisons are not provided. However, it is also assumed that the broad patterns identified here will, as a rule, have only changed slowly over time.

1. Age

Age of internal migrants is based on age at mid-2013 (specifically 30 June) rather than age at date of move – this is done so that the statistics integrate with ONS’s mid-2013 population estimates. So, for example, if someone born in May 1994 moved to another local authority to start university in September 2012 their age in the dataset would be 19, even though they were 18 when they actually moved.

Figure 3 shows the number of people who had moved local authority over the previous 12 months (including those arriving from Northern Ireland and Scotland), as a proportion of the population for each age in England and Wales as at mid-2013. This provides an approximation of what percentage of people moved at each age. However, the percentages will not be exact because:

1. Many people’s age at mid-2013 will have been one year older than when they moved. This will have had particular impact at age 0 (approximately half of people who moved aged 0 will have been aged 1 by mid-2013) and at student ages.
2. Some people will have moved more than once during the year.
3. Some people will have moved during the year but no longer lived in England and Wales by the end of the year, either because they had moved elsewhere or they had died. These people will appear in the internal migration data but not in the population estimates.

Figure 3 shows a comparatively high likelihood of moving for very young children. Part of this may be simply because their parents are at an age where moving is still common. However, the addition of children to a family may directly necessitate a move to more spacious accommodation, or to an area perceived as preferable for children to be raised. But it is also evident that once the children are at school moves are much less common, potentially because of the disruption it would cause the children as well as the parents.

However, it is in early adulthood where most moves occur, with the peak age for moves being 19, the main age at which people leave home for study. Even accounting for multiple moves and people who left England and Wales during the year, it is likely that approximately 20% of 19 year olds living in England and Wales at mid-2013 had moved during the previous 12 months.

There is another peak at age 22; in many cases this will reflect graduates moving for employment, further study, returning to their home address or potentially moving in with a partner (although Berrington (2009) (412.7 Kb Pdf) reports that since the 1980s there has been a decline in the number of people living with a partner in early adulthood).

Levels of movement remain comparatively high through the 20s and 30s but gradually decline with age. This may reflect people becoming more settled in their employment, in an area and in relationships, as well as because they have school-age children.

However, from the late 70s onwards the proportion of people moving rises. There are many reasons why people of this age may wish to move, including being closer to their family (potentially after a bereavement), downsizing into something more manageable, or specifically to access support and care (Myageingparent.com).

2. Sex and age

Of the 2.81 million internal moves affecting local authorities in England and Wales (including moves to or from Northern Ireland or Scotland) during the year ending June 2013, a total of 1.47 million (52%) were female and 1.35 million (48%) were male. This means that for every 100 female movers there were 92 male movers.

There are various factors which may contribute to this difference between the sexes. For example, differing likelihoods of moving for work, or as relationships are formed or break up. However, the most dominant factor is likely to be that more females attend higher education institutions: in the 2012/13 academic year 56% of all higher education students in the UK were female, and among full-time students, who are assumed to be more likely to move for study, 54% were female (HESA, 2014).

Another key factor is that these internal migration statistics exclude moves into and out of the armed forces, which are predominantly men.

In addition there is likely to be some impact of the fact that two of the key data sources for the estimates (GP Patient Register and the NHS Central Register) are derived from people’s re-registration with a new GP when they change address. On average young men wait longer than young women to register with a new GP after they have moved. This means that a higher proportion of men may not register with a GP at all while they are in a particular area, causing the respective moves to be missed from the statistics completely (Smallwood and Lynch, 2010 (145.3 Kb Pdf)).

Another way of considering the age / sex profile of internal migrants is to consider how it compares with the age / sex profile of the general population. For each age and sex group Figure 4 presents the number of movers (solid pyramid) as a percentage of all movers for that sex, and the mid-2013 population (outline pyramid) as a percentage of the total population for that sex. For example, 19 year olds constitute almost 6% of female movers, but not much over 1% of the overall population of females.

Figure 4: Population pyramids for (i) movers into local authorities in England and Wales (including moves from Scotland and Northern Ireland), year ending June 2013 and (ii) the total population of England and Wales, mid-2013

Source: Office for National Statistics

Notes:

1. Ages of 90 and above are combined into a single ‘90’ category.
2. Age is age at mid-2013.

Download this image Figure 4: Population pyramids for (i) movers into local authorities in England and Wales (including moves from Scotland and Northern Ireland), year ending June 2013 and (ii) the total population of England and Wales, mid-2013

.png (39.1 kB) .xls (174.1 kB)

The pyramids in Figure 4 show that despite the differences identified above, the internal migrant age distributions were broadly similar for both sexes, as were the general population age distributions. Young adults are disproportionately likely to have moved but at other ages apart from young children the number of movers is disproportionately low. However, the comparative increase in moves at much older ages is also visible, with the distance between the outlines of the two pyramids becoming smaller.

3. Area

The following section considers internal migration at regional and local authority level. However, it should be remembered that these comparisons consider internal migration only and that several other factors will influence total population change in an area, including births, deaths and international migration. This means that total population change will not necessarily be in the same direction as net internal migration. Information on total population change in the year ending June 2013 is available in the mid-2013 population estimates publication (ONS, 2014).

Table 1 shows the number of internal migration moves into and out of the regions of England and the country of Wales (including moves to or from Northern Ireland or Scotland) in the year ending June 2013. The table is presented in order of net moves per thousand mid-2012 population. The mid-2012 population is used because it is the starting point of the year ending June 2013.

The two regions with the highest numbers of moves in and moves out are London and the South East; these are also the regions with the largest net change. However, they are also the regions with the largest population.

By comparing net moves per thousand population, however, we can take account of the effect of population size. Some key messages are:

• regions in the north of England, plus West Midlands and the country of Wales, all had small net internal migration outflows during the year ending June 2013

• London was the region with by far the largest rate of net internal migration outflow, and also the largest net change in relation to population size (net of 6.6 moves out per 1,000 mid-2012 population)

• regions in the east and south of England had net internal migration inflows. The region with the highest rate of net migration inflow was the South West (net of 3.9 moves in per 1,000 population)

Previous years’ internal migration estimates shows that this general pattern of a large net outflow from London and a large net inflow to other parts of southern and eastern England has existed for a number of years.

Map 1 shows net flows per 1,000 mid-2012 population for each of the 348 local authorities in England and Wales.

Map 1: Net flows of internal migrants by local authority, year ending June 2013, per 1,000 mid-2012 population, local authorities in England and Wales

Source: Office for National Statistics

Download this image Map 1: Net flows of internal migrants by local authority, year ending June 2013, per 1,000 mid-2012 population, local authorities in England and Wales

.png (431.7 kB) .xls (70.1 kB)

In total there were 212 local authorities with more people moving in than out, of which 93 had a net inflow of more than 5 per 1,000 population (mid-2012). Most of those are in regions that also had a net inflow: South West, South East, East and East Midlands. However, some local authorities in all of these regions experienced a net outflow, demonstrating that there is considerable within-region variation.

Conversely there were 136 local authorities with more people moving out than in, of which 47 had a net outflow of more than 5 per 1,000 population (mid-2012). A particular concentration of local authorities with high net outflow was in London, driving the high net outflow for the London region. This will partly reflect the fact that London is the most common region of first residence for international migrants to the UK (ONS, 2013b), many of whom may later move on to other regions.

Making international comparisons of internal migration statistics is not straightforward. Complications include:

• different data sources for creating internal migration statistics, which may also have different levels of quality

• different frequencies of production: within the UK statistics are available on an annual basis, but this is not true for all countries

• different definitions of what constitutes internal migration

• different geographic structures and sizes of geographic areas. In theory two countries could have the same number of people changing address but have differing internal migration totals because one country has larger regions or local authorities than the other

However, a recent comparison was made by the United Nations (UN, 2013). The UN’s analysis did not include the UK and focussed on migration of individuals over a longer time period (5 years, or a whole lifetime). It did, however, use a technique to take account of differing geographic structures and area sizes. The conclusion was that internal migration levels were highest in Australia, Canada, New Zealand and the United States of America, and lowest in Asia. Levels in Europe, Latin America and Africa were intermediate, but also showed considerable differences between individual countries.

Internal migration statistics have a range of uses across the public sector and beyond. For example:

• the Office for National Statistics uses them in population estimates and projections

• central government uses them to inform policy-making and resource allocation to local authorities, who in turn use the statistics to help estimate the demands on their services – for example, the number of school places required

• clinical commissioning groups use them to help forecast requirements for health services, particularly for maternity and geriatric services

• various universities do research into internal migration: for example the University of Leeds, the University of Manchester and Newcastle University

• providers of service across the public and private sectors: for example, housing associations, developers and utility companies may use internal migration statistics to predict demand for their respective services in their area. - The media may use them to inform articles and debate on migration and related topics

The statistics in this release are the best available source of information on internal migration moves affecting England and Wales. However, over time data sources change and new data sources become available. ONS is therefore part way through a project researching how internal migration statistics can be improved. The earliest date at which any resulting improvements may be included in official estimates will be in 2016, at the time of the release of the estimates for the year ending June 2015.

Another development is that later in 2014 ONS intends to publish internal migration statistics for the years ending June 2012 and June 2013 in a new ‘Data Explorer’. This will contain all the records in the detailed flows file, but will make it much easier for users to select and export just those records they want.

This release is intended to meet the needs of users and provide an interesting summary of the key messages in the data. But are we achieving this? If you have any comments on what you like, what you don’t like, or what we could do differently, please let us know. Contact details are available in the ‘Statistical Contacts’ section of this bulletin.