UK immigration policy after leaving the EU: impacts on Scotland's economy, population and society

5. Demographic Effects

This chapter investigates how the proposed changes to immigration will affect the Scottish population. We start by providing a brief overview of population trends in Scotland over the past decades, and then analyse different scenarios of how changes to migration will affect Scotland over the coming two decades. This Scotland-level analysis is complemented by an examination of the spatial distribution of migrants and the effects on population in local areas. We analyse the demographic effects of migration in urban, rural and remote areas across Scotland, and consider how the proposed changes might affect different types of communities.

5.1 Population analysis

Population change

Over the last half-century Scotland's population has increased from 5.20 million in 1967 to 5.42 million in 2017 (Figure 5.1). However, the growth rate has varied significantly over these decades. While population growth was positive for short periods in the late 1960s and early 1970s and also during the early 1990s, Scotland's population declined for most of the 1970s, 1980s and 1990s (Figure 5.2). By 2000, Scotland's population had declined to 5.07 million. This decline launched a debate in the re-established Scottish Parliament on the long-run sustainability of Scotland's population. However, this trend began to change in the first decade of this century, and between 2000 and 2017 Scotland's population witnessed a significant increase in its size of 7%.

Figure 5.1. Scotland's population, 1967-2017

Source: NRS

Figure 5.2. Annual population growth in Scotland, 1967-2017

Source: NRS

Migration has been the major driver of population change over the past four decades. Substantial out-migration from Scotland accounted for population decline in the 1970s and 1980s, whereas recent population growth is largely attributed to significant in-migration flows to Scotland. Natural change (birth minus deaths) has been marginally positive over the past forty years, although it was negative in the first years of this century and has become negative again in the last few years (Figure 5.3). The comparison of Scotland to the rest of the UK shows that recent trends in population growth have been similar in the four nations of the UK. However, unlike in the other nations, Scotland has experienced population growth only in the last two decades and the growth has been smaller than the average growth of the UK's population (Figure 5.4).

Figure 5.3. Components of population change in Scotland, 1967-2017

Source: ONS, NRS

Figure 5.4 Annual population growth in Scotland and the UK, 1971-2017

Source: ONS

Mortality and fertility

Mortality has declined and life expectancy has increased in Scotland over the past fifty years. The life expectancy at birth was 70.65 in 1967, whereas the figure was 79.03 in 2016, an increase of 8.4 years over a half-century (or 1.7 years per decade). There are significant differences in life expectancy between sexes; on average, women live four to six years longer than men (Figure 5.5). The comparison of Scotland with the UK shows that life expectancy in Scotland is shorter both for males and females; in 2016, the differences were 1.8 and 2.3 years, respectively. The cause-specific analysis shows that higher cardiovascular and cancer mortality in Scotland explains most of the differences in life expectancy between Scotland and the rest of the UK.

Figure 5.5. Life expectancy at birth in Scotland, 1967-2017

Source: Human mortality database

Significant changes have also taken place in fertility levels. In 1967, at the heyday of the post-war baby boom, the total fertility rate (TFR) was 3.0 in Scotland. TFR measures the average number of children for a synthetic cohort of women in a particular year. Fertility declined significantly in the late 1960s and 1970s reaching the level of 1.7 in 1977 (Figure 5.6). Gradual decline continued during the last two decades of the twentieth century, and by 2002 the TFR had fallen to as low as 1.47. The first decade of this century witnessed a significant increase in fertility levels; in 2008, the TFR was 1.77. Thereafter, fertility declined again and the latest data (2017) report the figure of 1.47.

Figure 5.6. Total Fertility Rate in the UK by country, 1971-2017

Source: ONS

Again, fertility trends in Scotland have been similar to those in rUK in the past decades, except that fertility has remained at a lower level in Scotland compared to elsewhere in the UK. The average completed family size (or TCFR) has also declined over the past years, implying that the decline in the TFR is driven by reduced family size rather than the postponement of childbearing to later ages. While women born in 1951 had 2.03 children by age 44, those born in 1971 had 1.74 children by age 44. The implication of this is that fertility levels in Scotland are too low to replace generations in the long run.

International migration

As we saw in Chapter 2, Scotland experienced a prolonged period of emigration in the last three decades of the twentieth century, which was reversed from around 2001 onwards. Net migration has been positive for both international and UK migration streams, averaging at a total of around 21,000 per year. Chapter 2 also showed that migrants to Scotland have a relatively young age profile. Migration between Scotland and overseas (including the EU) is dominated by those in their 20s (Figure 5.7 and 5.8).

Figure 5.7. In-migration rate in Scotland from overseas by age (1-year groups), 2001-2016

Source: NRS

Figure 5.8. Out-migration rate in Scotland to overseas by age (1-year groups), 2001-2016

Source: NRS

Population composition

Past demographic processes have also shaped the structure of Scotland's population. Those born in the 1960s form the largest cohort among the Scottish population, followed by their children or those who are currently in their mid-twenties (Figure 5.9). The number of individuals aged 65 to 69 is also large, indicating higher fertility levels shortly after World War II. The 'waves' in the age structure of Scotland's population are not surprising; however, the baby boomers' children cohort is smaller than their parents' cohort and their children's cohort is even smaller. This is most likely related to both the significant decline in fertility levels in the late 1960s and high emigration rates, which characterised Scotland in the last three decades of the twentieth century.

Figure 5.9. Scotland's population by age and sex, 2016.

Source: NRS

The comparison of Scotland's age structure to that of the UK shows that the share of baby boomers is larger in the Scottish population and the relative size of the youngest age groups is smaller than in the UK's total population (Figure 5.10). Population ageing is thus going to be more pronounced in Scotland than in the UK as a whole.

Figure 5.10. Share of age groups from total population in Scotland and the UK, 2017

Source: ONS, NRS

Increased immigration and positive net migration over the past two decades have led to a significant increase in the share of the foreign-born population. In 2018, there were 460,000 non-UK born individuals in Scotland (with a confidence interval of ±25,000); they formed 9% of the total population, compared to 14% of the population in the UK as a whole (Figure 5.11). Half of the foreign-born individuals in Scotland came from EU countries, whereas the share of EU-born individuals was two-fifths for the UK as a whole. Scotland has fewer immigrants than the rest of the UK, especially England, but a larger share of them come from other EU countries.

Figure 5.11. The Share of non-UK born population in Scotland and the UK, July 2017-June 2018

Source: ONS

5.2 Population projection

Methods and assumptions

In order to estimate the effects of the proposed policy changes on population trends, we conducted a population projection using the cohort component method as implemented in the Popgroup projection software (Edge Analytics 2018). We adopted the deterministic approach but conduct a series of projections to also deal with uncertainty (UNECE 2018). Our aim is not to predict the future, but rather to project changes in population size and structure conditional on different assumptions about future migration streams to Scotland. We used the 2016 ONS and NRS population projection as the baseline, which is the latest projection available. We prepared several projection variants, which are based on different assumptions of future migration to and from Scotland. We conducted a population projection until 2041.

The assumptions of the NRS 2016-based principal population projection are as follows:

• Fertility rates (the TFR) are expected to gradually increase from 1.49 in 2016 to 1.65 in 2041.
• Mortality rates are projected to decline, although the decline will be slower than in the past; the life expectancy at birth will increase from 77.0 for males and 81.1 for females in 2016 to 81.7 and 84.5 in 2041, respectively.
• Migration from overseas to Scotland is assumed to slightly decline over the next few years – 30,000 individuals are expected to arrive in Scotland annually (from 2022 onwards); out-migration will be 23,000 individuals annually; overseas net migration is projected to be +7,000 individuals a year. Long-term migration between Scotland and overseas is calculated using a 25-year average of migration trends. Additionally, annual net migration between Scotland and the rest of the UK is projected to be between +7,400 and +8,000 per year over the next 25 years.

We use the same assumptions about fertility, mortality and migration in the rest of the UK as the ONS/NRS principal population projection for Scotland. However, we use different assumptions about migration to and from overseas, based on the analysis set out in Chapter 2. Our baseline projection assumes that overseas migration flows remain the same as they have been in the past five years (using the five-year averages for in- and out-migration), so similar to the average migration flows between 2001 and 2017. These estimates are much higher than in the NRS principal projection; however, they allow us to measure the effect of changing migration flows on population size and structure, as a result of expected policy changes.

We will develop six variants (including the baseline):

• Current migration: he annual number of migrants will be as follows: 34,500 immigrants, 21,500 emigrants, implying overseas net migration at +13,000. (15,500 immigrants come from the EU countries annually and 7,500 leave for the EU with net migration of +8,000).
• 50% EU: Migration to and from the EU countries will decline by 50% after 2020. This corresponds to scenario 2 set out in Chapter 2.
• 20% EU: Migration to and from the EU countries will decline by 80% after 2020. This corresponds to scenario 1 set out in Chapter 2.
• 0% EU: There will be no migration from and to the EU after 2020.
• 0% Overseas: There will be no overseas migration after 2020.
• No migration: There will be no overseas migration and migration to and from the rest of the UK after 2020. This variant is developed to illustrate the effect of (mostly) past demographic changes, accumulated in the population's age-sex structure, on the future development of Scotland's population.

Projection results

Five variants out of the six show an increase in Scotland's population (Figure 5.12). The size of the Scottish population is projected to increase 8% in the next 25 years for the current migration variant and 2% to 6% for the reduced migration scenarios. With no migration from overseas and the rest of the UK, Scotland's population will decline.

Figure 5.12. Projected Scotland's population, 2016-2041

Source: Expert Advisory Group on Migration and Population

Significant changes will take place in population composition by age. The share of people 65 and older is projected to increase from the current 18% to more than one‑fourth in 2041 (Figure 5.13). Interestingly, the differences in projected elderly population are not very large between variants, although the zero overseas migration (0% overseas) and zero migration scenario (no migration) will accelerate population ageing.

The dependency ratio (the ratio of individuals aged 0-15 and 65+ to that of 16-64) and the old age dependency ratio (the ratio of individuals aged 65+ to that of 16-64) provides another way of exploring population ageing (Figures 5.14 and 5.15). The share of people aged 65 and over is projected to increase from 29 per 100 to between 41 and 46 per 100 in the next two decades. The projection does not consider the planned change in pension age. Again, the variation is not large between the different variants, although the increase in the old age dependency ratio is largest for the zero migration scenario.

Figure 5.13. Projected share of individuals 65 and older in Scotland, 2016-2041.

Source: Expert Advisory Group on Migration and Population

Figure 5.14. Projected dependency ratio in Scotland, 2016-2041.

Source: Expert Advisory Group on Migration and Population

Figure 5.15. Projected old age dependency ratio in Scotland, 2016-2041.

Source: Expert Advisory Group on Migration and Population

Migration will have relatively little impact on the general trend of population ageing; however, it will significantly shape the size and the composition of the working age population (individuals currently aged 16-64). The analysis shows that with current migration levels the working age population would stay stable in the next 25 years (Figure 5.16). With reduced EU immigration it is projected to decline by between 3% and 5%, with no overseas migration by 8%, and with no migration to Scotland by 12%. The working age population is also expected to age, although the trends are more complex than anticipated. The analysis shows that the ratio of younger working age population (16-44) to older working population (45-64) would increase for all six variants, although the increase would be modest for the no migration variants (0% overseas and no migration in Figure 5.17). Interestingly, however, this ratio is projected to decline in the 2030s and the decline will be pronounced for the 'no migration' or the 'UK migration only' scenarios. This indicates the importance of overseas migration in shaping the composition of the working age population.

Figure 5.16. Projected relative change of individuals aged 16-64 in Scotland, 2016-2041

Source: Expert Advisory Group on Migration and Population

Figure 5.17. Projected ratio of population aged 16-44 to aged 45-64 in Scotland, 2016-2041

Source: Expert Advisory Group on Migration and Population

We can also compare projected changes in Scotland's population to those of the UK. We use the 2016 ONS principal projection and selected projection variants for Scotland. The ONS principal projection assumes that migration to the UK will decline by one-fourth post-2020. According to the principal variant, the UK's population is projected to increase by more than one-tenth in the next two decades, whereas Scotland's population will grow by 5% according to the ONS/NRS projections (with declining migration flows; Figure A7 in Annex A) and 8% if the current migration levels persist (Figure 5.18).

Figure 5.18. Projected relative population change in the UK and Scotland, 2016-2041.

Source: Expert Advisory Group on Migration and Population

Interestingly, while the working age population is projected to decline in Scotland unless overseas migration stays at the current level, the number of individuals aged 16-64 in the UK will increase slightly according to the ONS/NRS principal projection variant, which assumes declining migration trends. Moreover, the ratio of younger working age population (16-44) to older working population (45-64), which is higher in the UK than in Scotland, is projected to remain stable in the UK in the next decades, whereas this ratio is expected to decline in Scotland in the 2030s for reduced migration scenarios. Clearly, migration has been and will continue to be an important factor to ensure the stable size and age composition of Scotland's working age population in the future. The share of people aged 65 and older is projected to increase in similar rates in the UK, Scotland and also in other European countries, although the figures will be slightly lower for the UK (Figure 5.19).

Figure 5.19. Percentage of individuals 65 and older in selected European countries, 1950-2050

Source: UN Population Division, 2008

5.3 Population and migration in local areas of Scotland

Thus far, we have set out aggregate-level population projections for Scotland as a whole. However, current migration flows have a differential spatial impact. Moreover, as we saw in Chapter 3, the proposed new rules on immigration would have differentiated effects on migration to local areas. It is therefore important to consider how the changes would affect different types of areas in Scotland, especially those that are already in a relatively fragile situation in terms of their demographic and economic future.

Although regional demographic trends do not feature among the objectives of UK migration policy, they are very important in relation to the Scottish Parliament's devolved responsibilities for regional economic and social development. Indeed, in recent years Scottish economic policy has increasingly emphasised the twin goals of economic growth and reductions in inequality (both between individuals, and between places). This ethos is summed up by the term “inclusive growth”, and it forms one of the four key priorities of Scotland's Economic Strategy.^[25] More specifically a concern to ensure that prosperity is shared by all parts of the country is reflected in the establishment of Regional Economic Partnerships,^[26] the local plan requirements of the most recent National Planning Guidance Framework,^[27] and the upcoming National Islands Plan.^[28]

Available evidence on the geography of migration in general, and EU migration in particular, is sparse, and requires careful interpretation. Almost all the data on which this section relies is available for Scotland's 32 local authorities, but not at any finer geographical resolution which could facilitate an accurate analysis of urban-rural differences. However, broad contrasts can be explored on the basis of the recently published RESAS classification of Council Areas, which groups the Scottish council areas into the following four types:

• Larger Cities
• Urban with Substantial Rural
• Mainly Rural
• Islands and Remote Rural

In addition, the mapping of selected variables at local authority level is helpful in highlighting geographic patterns which are to a degree independent of the broad categories of this council area classification.

Key features of demographic change in different parts of Scotland 2007-17

Over the past decade total population change is estimated to have been positive in three of the four categories of Council Areas. The exception is the islands and remote rural category, which have (overall) shown a decline of 1.3% in their total population. Taking the two components of population (natural change and migration) separately, it is clear that the two more urban groups of council area have a relatively positive situation and dynamic, benefitting from both positive natural change and positive migration.

The two rural categories show a more mixed situation. In the mainly rural areas, net migration over the 2007-17 period equates to more than 4% of the 2007 population, more than counterbalancing a small negative natural change. In the remote rural areas net migration of +1.5% (of the 2007 population) was insufficient to replace the -2.8% contraction of the population due to the surplus of deaths over births.

Figure 5.20: Components of Population Change 2007-17 by Council Area Type

Source: NRS Annual Population Estimates mid-2017

However, this classification masks some important variation between individual areas (Figure 5.21). Edinburgh, East and West Lothian and Glasgow showed some of the largest proportions of change due to migration. Of the more rural areas, Orkney, Highland, Moray, Aberdeenshire, Perth and Kinross and Stirling also stand out. In all of these net migration accounts for more than 4% of the population change experienced between 2007 and 2017. Argyll and Bute, Shetland, Lanarkshire, Ayrshire and Dumfries and Galloway were at the opposite extreme.

Figure 5.21: Migration as a percentage of Estimated Total Population Change by Council Area, 2007-17

Source: Annual Population Estimates mid-2017

Although the above analysis does not distinguish domestic migration from overseas, it serves to underline three key points about the importance of migration to population sustainability in Scotland:

• Across the first three of the categories of Council Area migration has, over the last ten years, been by far the most important driver of population change, natural change being very much constrained (even here) by the ageing of the population.
• In the Mainly Rural and Islands and Remote Rural Council Areas in-migration is the only option to sustain (or grow) the population in the short to medium term. The age structure of these areas is too “damaged” by past out-migration for natural change to play a role in recovery or repopulation.
• With certain notable exceptions, migration seems to play a more significant role in local or regional demographic sustainability in rural areas in the north of Scotland than in the south of Scotland.

Origins of migrants

When the focus shifts to areas within Scotland, the two streams of migrants discussed in section 4.1 (overseas and rest of UK) are joined by a third component – migration between different areas of Scotland.

Data exists for 2017 to allow us to begin to explore the relative importance of these three components of migration as part of current demographic processes across the Scottish council areas. Figure 5.22 shows the composition of in-migration assumed by NRS in their 2017 population estimates. Not unexpectedly, within-Scotland migration accounts for the largest share in all four categories of council area, ranging from a little less than half to three-quarters of all in-migration. The rest of the UK accounts for between 17% and 37% of in-migrants, the highest proportion being in the islands and remote council areas. The share of foreign migrants varies considerably between the four groups, rising to 28% in the large cities, and reaching only 7% to 8% in the more rural and remote areas.

Figure 5.22: Composition of Migrant flows by Council Area type, 2017

Source: Annual Population Estimates mid-2017

However, Figure 5.23 again suggests that the council area classification hides a significant North-South pattern. Of the more rural local authorities (where demographic sustainability is more finely balanced) Perth and Kinross and Stirling stand out, with more than 11% of their population change due to overseas migration. International migration was also relatively important in Shetland, Highland, Aberdeenshire and Angus.

Figure 5.23: In-Migration from Overseas as a percentage of total In-Migration 2017

Source: Annual Population Estimates mid-2017

The significance of the geographic variation in the role of the three components of migration lies in their potential impact on the age structure of the recipient population. Overseas migration is particularly conducive to future demographic sustainability because, whereas a substantial proportion of domestic migration (both within Scotland and from rUK) is associated with retirement (or pre-retirement adjustments to working patterns), the majority of overseas migrants (68%) are within the younger part of the working age population (WAP), and therefore have a benign impact upon the age structure and fertility rates (Figure 5.24).

Figure 5.24: Age structure of Migrants from Rest of UK and Overseas 2017

Source: Annual Population Estimates mid-2017 (Table 8)

5.4 Learning from the most recent (sub-national) projections

The production of sub-national projections is a complex process, not least because of the need to ensure that flows between constituent areas, and projected population trends for those areas, are consistent with, and sum to, those of the country as a whole. As a consequence, the following account relies upon the latest principal projections made by NRS. These look ahead over the 25 years from 2016 to 2041. As explained above, these projections differ from the baseline projection produced for this report (section 4.1). Nevertheless, these NRS sub-national projections provide a good starting point for a discussion of the implications of the proposed migration policy after leaving the EU on geographic patterns of demographic sustainability.

The NRS projections indicate that even with ongoing free movement of people the demographic situation of the Islands and Remote Rural areas will be a continuing cause for concern. Natural change over the 25-year period is predicted to equate to a more than 10% reduction in the population. Net migration is predicted to offset this by less than 3%.

Figure 5.25: Components of Population Change 2016-41 by Council Area Type

Source: NRS 2016-based Projections

By contrast, net migration into Mainly Rural areas is forecast to reach the equivalent of almost 10% of the current population, more than offsetting the natural decrease of a little over 5%.

In the Urban with Substantial Rural areas, natural increase is predicted to shift into negative territory, but still to be more than offset by net migration, whilst in the Larger Cities both natural change and net migration are projected to remain positive, giving these areas the most optimistic future in terms of population growth, which approaches 10% over the next 25 years.

Mapping the percentage of 2016-41 population change consequent on net migration, Figure 5.26 suggests that in addition to the expected areas of Glasgow, Edinburgh and the Lothians, the areas most affected by migration over the next 25 years are Perth and Kinross, Stirling, Moray, Aberdeenshire and Scottish Borders. The explanation of this change from the north-south contrast evident in the Annual Population Estimates is not immediately apparent.

Figure 5.26: Net Migration as a % of Total Population Change 2016-41

Source: National Records Scotland 2016-based Projections

The above patterns of change assume only the continuation of existing trends, taking no account of the implications of UK immigration policy after leaving the EU. There are two arguments to suggest that the changes proposed by the White Paper would exacerbate the challenges facing the remote rural parts of northern Scotland:

• The above projections suggest that these areas are already dependent on overseas migration as a 'corrective' for age structures which are incapable of generating positive natural change.
• As Chapter 3 has shown, the replacement of free movement with an adjusted Tier 2 framework is highly likely to result in a substantial reduction of overseas migration to precisely these areas, because of their dependence upon economic activities (tourism, farming, food processing) which do not pay sufficiently high wages to meet the Tier 2 salary threshold.

Summary and Implications

Immigration to Scotland has been the major driver of population change since the early 2000s. Our analysis suggests that if migration stays at its current level (the average in- and out-migration flows over the past five years), the Scottish population is projected to increase by 8% over the next 25 years. With reduced international migration, population growth will be slower. Migration can also help reduce the speed of population ageing, although its impact is smaller than expected. The proposed increase in pension age may be a more important factor than migration in reducing the increase in the dependency ratio.

While migration may not have a considerable effect on population ageing, it will significantly shape the size and composition of the working age population. In other words, migration will have an important impact on the absolute size of the working age population, as well as the age distribution within the working age population. Reduced international migration to Scotland would lead to a gradual decline and rapid ageing of the working age population. This is in contrast to the UK as a whole, whose population aged 16-64 would still grow in the scenario of reduced international migration.

The ageing of the baby boom generation will drive population ageing in Scotland in the next two decades. Whether the working age population will also decline in absolute terms will depend to a large degree on the size of future migration to Scotland. With reduced migration from the EU, Scotland's working age population is projected to shrink and become older because relatively small cohorts will enter the labour market over the next two decades. Overall, our projection results are very similar to those prepared by ONS and NRS, except that we project a larger growth of Scotland's population over the next two decades (the ONS/NRS projections are presented at Figures A1 to A7 in Annex A).

Importantly, the impact of these population trends would not be spread evenly across different areas of Scotland. Between 2007 and 2017, urban and mixed council areas have benefited from positive natural change (more births than deaths) and substantial in-migration (mostly from rUK but also from overseas). Mainly rural areas have seen a small natural increase, but this has been more than compensated by in-migration. Remote rural and island areas, by contrast, have seen negative natural change, which has not been compensated by in-migration. The demographic challenges for these areas will therefore be exacerbated by the proposed changes to EU immigration.

This challenge is particularly acute given that for many remote rural and island areas, in-migration is the only means of sustaining the population in the short to medium term. The age structure has been too 'damaged' by past out-migration for natural change to contribute to repopulation. With some exceptions, migration plays a far more important role in demographic sustainability in rural areas north of Scotland than it does in the south of Scotland.

Overseas migration is especially conducive to future demographic stability because of its relatively younger age structure. However, in rural and remote areas a much smaller share of in-migrants are from overseas than is the case in urban areas. Just 8% of migrants to rural areas are from outside of the UK, compared to 28% for cities. Moreover, rural areas are far less likely to have jobs that meet the proposed Tier 2 threshold. As a consequence, the changes proposed in the White Paper would largely eliminate opportunities for encouraging the longer-term stay and settlement of non-UK nationals in rural areas.

For remoter rural areas and islands, attracting working-age migrants (including from overseas, and EU countries in particular) is the only realistic option to avert a downward demographic spiral driven by the age structure legacy of selective out‑migration during the last decades of the twentieth century. These areas of Scotland seem to be facing a demographic 'double whammy' in leaving the EU and ending free movement which is likely to have far-reaching implications for economic activity, the provision of services, and levels of general well-being.