National population projections, migration assumptions: 2018-based

• The migration assumptions are not forecasts or predictions of future migration levels; they should be interpreted as scenarios where net migration to the UK averages at a certain level.

• The principal migration assumption for the 2018-based national population projections (NPPs) assumes net migration averages at 190,000 per year from the year ending mid-2025 onwards.

• The method used to set the long-term international migration assumptions is a 25-year average. This is the same methodology used in the 2016-based NPPs.

• High- and low-variant migration assumptions assume net migration at 100,000 above or below the principal assumption respectively.

This article provides detailed information on the international migration assumptions used in the 2018-based national population projections (NPPs).

The principal NPPs assume net international migration to the UK remains constant at +190,000 per year from the year ending mid-2025 onwards; this is referred to as the “long-term assumption”. This assumption is 25,000 higher than the 2016-based principal migration assumption (+165,000).

Demographic behaviour is inherently uncertain, and it is not possible to predict future migration trends. The long-term assumptions should therefore not be seen as forecasts but rather as scenarios where future net migration averages at a certain level.

The methodology used to set the 2018-based long-term assumptions remains unchanged from the 2016-based NPPs. The difference in assumed long-term migration is purely the result of using new input data. The 2016-based NPPs were calculated using data from the years ending mid-1992 to mid-2016, while the 2018-based NPPs use data from the years ending mid-1994 to mid-2018.

Our decision-making process for setting migration assumptions is published in National population projections, how the assumptions are set: 2018-based.

Definition of international migration in the national population projections

The national population projections (NPPs) define migration using the UN definition of a long-term international migrant: someone who has moved to a country other than their usual country of residence for a period of 12 months or more.

The NPP long-term migration assumptions do not include short-term international migrants under the UN definition (that is, those who move to or leave the UK for less than 12 months).

Long- and short-term international migration assumptions

Migration assumptions in the NPPs are split into the long- and short-term assumptions.

“Long-term assumptions” take effect from seven years into the projection period, from the year ending mid-2025 onwards.

“Short-term assumptions” are effective from the start of the projections period up to the year ending mid-2025.

International migration

Long-term international migration assumptions are produced by extrapolating historical migration estimates to and from each UK country.

The principal long-term assumption is calculated using a 25-year average of historical long-term international migration estimates.

Short-term assumptions are produced using a linear interpolation to join the mid-2018 migration estimates to the long-term assumptions. Short-term assumptions also include additional flows such as refugees and home armed forces returning from Germany.

We use a number of different sources for historical migration estimates. The principal data source is the International Passenger Survey (IPS). We make adjustments to account for people who enter or leave the country initially for a short stay but subsequently decide to remain for a year or more (“visitor switchers”) and people who originally intended to move out of their country of origin for 12 months or more but in reality return in less than 12 months (“migrant switchers”).

We use the components of change data from the mid-year population estimates for the years ending mid-2001 to mid-2011. This is because the revised back series of IPS totals created after the 2011 Census only covers net international migration, not inflows and outflows. Conversely, the mid-year population estimates components of change data contain revisions to both international inflow and outflows.

Note:

We are aware of the recent designation of the international migration estimates based on the IPS as Experimental Statistics. We have not incorporated the preliminary adjustments because they are not available by UK country, or by age and sex. However, at national level, the overall revisions are small as a proportion of total migration, so the impact on the projections will be limited.

Asylum seekers

We obtain estimates of asylum seeker migration flows from Home Office data on the numbers of (i) applications for asylum and (ii) enforced removals. As with international migration estimates, we have used the asylum seeker estimates from the mid-year population estimates components of change data for the period from mid-2001 to mid-2011.

Long-term assumptions for asylum seekers are set by taking an average of 10 years of historical data. This is because the trend for asylum seeker flows saw a significant decline from 2002 onwards.

The Home Office states that the fall in the number of asylum applications after 2002 coincided with:

• the introduction of a process that prevented certain nationalities from appealing a decision while in the country in 2002

• the opening of UK border controls in France and Belgium in 2002 and 2004 respectively

• the introduction of fast-track facilities for asylum applications in 2003

The number of applications for asylum has remained relatively low in recent years compared with levels before 2004 (Figure 1). As such, it is deemed more reasonable to produce long-term assumptions for asylum seekers based on a historical period after the events described by the Home Office.

Additional flows

The national population projections (NPPs) contain separate assumptions for people granted humanitarian protection arriving into the UK through the Vulnerable Persons Resettlement Scheme. These data are obtained from the Home Office.

We also make an assumption for home armed forces returning from Germany. These data are obtained from British Forces Germany.

Data for Northern Ireland

The data for Northern Ireland international migration are based on medical card data supplied by the Northern Ireland Statistics and Research Agency (NISRA).

This data source captures all migration flows to and from the country, and we therefore do not produce separate assumptions for asylum seekers, refugees and armed forces for Northern Ireland.

Please see Section 6 of the Long-term International Migration Statistics for Northern Ireland (2018) for the limitations of the IPS for Northern Ireland.

The long-term migration assumptions for Northern Ireland are calculated by taking a 25-year average of the historical data. No additional adjustments are made for data for the years ending mid-2001 to mid-2011.

Cross-border migration

Cross-border migration is the movement of people between the four countries of the UK. Cross-border migration assumptions are set as rates calculated using five years of historical data and are applied to the population of each UK country to calculate how many people are moving to a different UK country for each year in the projection.

The main advantage of applying rates for cross-border migration is that migrant flows are linked to the changing underlying population size and age structure. This means that the projection cannot produce implausible values such as negative population stocks.

The methodology for producing cross-border migration rates in the NPPs is explained in greater detail in the cross-border methodology (399KB PDF) report. Cross-border migration is calculated using data from the mid-year population estimates, the NHS Central Register (NHSCR) and the Personal Demographics Service (PDS).

Age and sex distributions

We assume a fixed age and sex distribution for each UK country based on an average derived from the latest five years of historical international migration data from the mid-year population estimates components of change for the corresponding country.

The national population projections (NPPs) are not forecasts, and our migration assumptions are not predictions of future migration.

The assumptions should instead be interpreted as plausible scenarios based on what has happened in the past.

We do not produce measures of uncertainty in our population projections; however, demographic behaviour is inherently uncertain and the actual levels of future migration to and from the UK may deviate from our projected flows.

We produce variant migration assumptions to provide users with a wider range of migration scenarios and illustrate uncertainty.

Please see National population projections, how the assumptions are set: 2018-based for details on the decisions made in producing our migration assumptions.

International migration assumptions are held constant from year ending mid-2025 onwards. This “long-term assumption” of net international migration to the UK is +190,000 per year in our principal projection.

Our migration assumptions are not forecasts. Our principal migration assumption is set based on average migration flows over the past 25 years for international migration and 10 years for asylum seekers (see Section 2). Assumptions for the years before the long-term assumption applies are created using a linear interpolation and adjusted to account for short-term migration flows such as refugees and home armed forces returning from Germany.

We produce additional migration assumptions (“variant” assumptions) in order to provide users with a range of alternative, plausible migration scenarios.

These variant migration assumptions are also not forecasts, nor are they measures of uncertainty around our principal projection. Instead, they offer plausible alternative future outcomes but with net migration at different average levels.

The long-term high migration variant assumption for the UK is +290,000 and the low migration variant assumption is +90,000, giving a range of +/- 100,000 around our principal migration assumption. This difference of 100,000 migrants is distributed to the four UK countries based on each country’s share of net international migration from the principal assumptions.

We have increased this range compared with our previous 2016-based national population projections (NPPs) (+/- 80,000). This is because of our discussions with our independent expert advisory group, where it was recommended that the Office for National Statistics (ONS) broaden the range between the high and low migration variants to highlight the considerable level of uncertainty facing the future of migration to and from the UK.

In addition to our standard high and low migration variants, we have also produced the 0% and 50% future EU migration variant projections in response to confirmed user needs following consultation. These are user-requested statistics we first produced for the 2016-based NPPs and are not classified as National Statistics because they have not been created using a standard method. They were created purely to fulfil specific stakeholder requirements.

The EU variant assumptions are produced by applying percentage changes by single year of age and sex to the principal international migration assumption. The percentage changes are calculated using the last three years of the long-term international migration estimate. These changes take effect from the year ending mid-2021 onwards.