Office for National Statistics follow-up written evidence to the Lords Justice and Home Affairs Committee’s inquiry into settlement, citizenship and integration

Dear Lord Foster,

Following my evidence session with the Committee on 27 January, I am happy to write with more detail on how the Office for National Statistics (ONS) estimates long-term international migration. I also committed to follow-up on data sharing arrangements with the Republic of Ireland.

Calculating long-term international migration statistics

Long-term migration is when people change the country where they usually live for 12 months or more. This is the United Nations (UN) definition of a long-term migrant. The method to calculate this, however, is different for each nationality group and whether the migrant has settled status, because the best data source is different for each of these groups: Non-EU+ and EU+ visa holders, EU Settlement Scheme (EUSS), Irish and British.

Non-EU+ and EU+ visa holders

For Non-EU+ and EU+ visa holder migration, the ONS uses Home Office Borders and Immigration (HOBI) data on long-term visas, which combines visa and travel information, to link an individual's travel movements into and out of the country.

To estimate long-term international immigration, we look for first arrival and last departure dates that are more than 365 days apart, allowing us to ignore short trips abroad. If either of these dates are missing, then visa start or end dates are used as a proxy.

To estimate emigration, we identify previous long-term immigrants with a last departure from the UK during the reference period. We record them as a long-term emigrant if they do not return to the UK within 12 months, or if they only return for a short-term stay.

To create timely information within 5 months of the reference period, we provide provisional estimates. For example, for individuals arriving or departing within the last 12 months, there is not enough information to confirm they meet the 12-month definition yet. We use an assumption based on the past behaviour of arrivals on long-term visas who only stay short-term (the 'early leaver adjustment') or those who leave and return within a year (the 'rearrival adjustment'). We apply this adjustment by reason for migration, nationality, and age, as different groups exhibit different behaviours. These provisional estimates are updated as we get further travel data to replace these assumptions with actual behaviour.

EU Settlement Scheme (EUSS)

For those with status on the EUSS, we use travel dates and calculate time spent in and out of the UK.

To identify a long-term immigrant on the EUSS, we identify periods within the UK of 45 days or more and add these periods together. This allows us to ignore short trips abroad. If the sum of these periods is 270 days or more, the person is classed as a long-term immigrant.

For a long-term emigrant, we do the same but looking for grouped periods of 45 days or more out of the country that amount to 270 days within a year.

Irish nationals

The movement allowed within the Common Travel Area means that HOBI data exclude Irish nationals. To estimate them, we use the Department for Work and Pensions' (DWP) Registration and Population Interactions Database (RAPID) data, which is data from the tax and benefits system.

For immigration we use the number of long-term Irish national migrants coming into the UK captured in RAPID as a proportion of total EU+ immigration in RAPID in that tax year. These proportions are then applied to our total EU+ (EU+ visa holders and EUSS) HOBI immigration estimates.

For emigration, we use the proportion of Irish nationals leaving the UK in RAPID in that tax year and apply that to our total EU+ (EU+ visa holders and EUSS) HOBI immigration estimates.

British Nationals

For British nationals' migration we also use DWP's RAPID data. We derive a RAPID British nationals dataset covering everyone who has a National Insurance Number (NINo) and who is not included in the Migrant Worker Scan (MWS). The MWS identifies all non-UK nationals registering for a NINo from 1975 onwards. We account for those who have naturalised and become British.

If a person has activity (interactions with the tax and benefit system) and has no evidence of a foreign address in RAPID, then the person is generally regarded as being "resident" in that year.

We look for changes in activity to see migration. An immigration event is defined as a change in residency status from "not resident" to "resident", and an emigration event is defined as a change in residency status from "resident" to "not resident".

A number of adjustments are made to account for situations where people may become inactive in the data for reasons other than migration, including those who become economically inactive. The method also applies a population level adjustment to correct for those who are not covered by the RAPID dataset.

These are simplified descriptions of methods that make further detailed assumptions and adjustments, which can be found in our technical user guide.

Republic of Ireland data sharing

You asked whether data is shared between the Republic of Ireland and the UK on migration. There is no data sharing between the ONS and the Republic of Ireland presently.

The ONS has previously discussed the possibility of using our data sources to understand more about travel across the border with the Northern Ireland Statistics and Research Agency (NISRA) and the Central Statistics Office (CSO); however, there was no clear source of data that could assist at that time. This is noted as one of our areas for future research.

I hope this is helpful and please do let me know if I can assist the Committee further.

Yours sincerely,

Mary Gregory