Panel:

• Marie McAuliffe (MM) – Independent Consultant
• Arkadiusz Wisniowsk (AW) – University of Manchester
• Georgina Sturge (GS) – Migration Observatory
• Jakub Bijak (JB) – University of Southampton and Centre for Population Change

ONS presenters:

• Nick Taylor (NT) – ONS chair
• Victoria Chenery (VC) – ONS
• Oliver Gomersall (OG) – ONS
• Brendan Georgeson (BG) – ONS
• Annabelle Tyrrell (AT)- ONS

ABME – Admin-Based Migration Estimates

DPM – Dynamic Population model

DWP – Department for Work and Pensions

EUSS – EU Settlement Scheme

IPS – International Passenger Survey

HO – Home Office

HOBI – Home Office Borders and Immigration

LTIM – Long-Term International Migration

MARP – Methodological Assurance Review Panel

MSUG – migration Statistics User Group

NISRA – Northern Ireland Statistics and Research Agency

NRS – National Records Scotland

NS – National Statistician

ONS – Office for National Statistics

QA – Quality Assurance

RAPID – Registration and Population Interaction Database

UKSA – United Kingdom Statistics Authority

Introduction & purpose of the panel

1. NT welcomed the panel to the first meeting of the group and described the purpose and how it fits with the wider governance structure.
2. The panel will provide external, independent technical assurance, review and guidance on the statistical methodology used to estimate international migration. This panel will agree key messages to be included in a brief report which will be sent to the Advisory Panel on Migration Statistics.
3. As set out in the Terms of Reference for the Advisory Panel, two seats will be held by members of the MARP sub-group on international migration. It was agreed that the seats will not be for named individuals and the panel members will agree who will represent the MARP sub-group on international migration at each meeting.
4. The panel will be involved in the research from early stages where methods are being researched through to assuring more developed methods which are closer to being implemented.

Research on EU Overview – presented by Annabelle Tyrrell (AT)

1. AT provided an overview of the current approach to estimating long-term international migration of EU+ nationals using the DWP Registration and Population Interaction Database (RAPID). Previous research has set out limitations of using RAPID data to estimate migration including those under the age of 16, students and those who are economically inactive.
2. AT presented the newly developed methods using Home Office Borders and Immigration (HOBI) data, along with the Quality Assurance (QA) steps undertaken.
3. The ‘First Arrival, Last Departure’ method has been adapted for EU+ visa holders and the “Cumulative length of stay in the UK” method which has been developed to capture long-term migration of those with EU Settled Status. Using HOBI data will align methods for EU+ visa holders with the non-EU+ methods.
4. Remaining challenges were discussed, including the edge effect in provisional estimates and the need to clarify differences between EU+ estimates derived from RAPID and HOBI data.
5. AT confirmed that current methods include the assumption that those who arrive in the final year of data on the EU Settlement Scheme stay long-term. This assumption is being developed further.
6. On revisions analysis, the migration statistics team within ONS is actively examining historical behaviours to inform future estimates.
7. The following points were raised by panel members on EU
   • AW raised a query regarding whether edge effects impact the provisional estimates.
   • Action: AW to provide comments for further clarification.
   • GS sought clarification on the data captured in HOBI (Home Office Biometric Information), which led to a broader discussion.
8. The research will be brought to the next panel meeting for further discussion.

British Nationals – presented by Brendan Georgeson (BG)

1. BG explained that ONS currently produces estimates of long-term migration of British Nationals using the International Passenger Survey (IPS) and a roll forward method. This method has become unviable to continue producing a British nationals migration estimate due to changes in the IPS.
2. The ONS is exploring Department of Work and Pensions (DWP) Registration and Population Interactions Database (RAPID) to improve estimates of long-term migration among British nationals.
3. Long-term migration is defined as being in or out of the UK for 12 months or more, such that the destination country becomes the country of residence.
4. ONS assesses activity or interactions within RAPID, (e.g. receipt of a benefit.) If an individual shows no activity in a given tax year, this may suggest they have emigrated. If an individual who was previously inactive shows new activity in a subsequent tax year, this may suggest they have returned to the UK or newly arrived. Additionally, deaths are recorded in RAPID, which helps distinguish between emigration and other reasons for inactivity.
5. However, inactivity in the data does not always equate to migration, as individuals may be inactive for other reasons. To address this, BG explained that a set of rules has been developed to flag individuals as residents. A person may receive multiple flags, but only the presence of a flag is counted, not the number.
6. The flagging rules categorise residency status based on set criteria e.g. their address is in the UK, and demographic or characteristic based assumptions for inactive individuals to decide if this is due to migration or not. This allows us to generate estimates of British long-term international migrants (both immigration and emigration) from RAPID.
7. One limitation is that RAPID is not updated when foreign nationals on the Migrant Workers Scan (MWS) obtain British citizenship. These individuals may be erroneously excluded from British migration estimates if they later emigrate or return. To address this, a naturalisation adjustment is required, as no current data source provides these figures.
8. The panel discussed the proposed British nationals’ method along with exploring other data sources, including possible linkage to HOBI data and academic research on naturalisation emigration rates compared to native born people’s emigration rates.
9. The following points were raised by panel members on British Nationals and EU.
   • Inclusion of BN(O) Individuals
     • Panel members asked whether BN(O) individuals (e.g. from Hong Kong) are included in the data. It was clarified that they are not included as British nationals but some who naturalise will be included in the naturalisation adjustment.
   • IPS vs RAPID Data Source
     • Question raised on whether the IPS data used in RAPID is modelled or raw. Response: BG explained:
     • Modelled data was used for the COVID years.
     • Raw IPS data is used post-COVID.
     • For the most recent quarter, data is 50% modelled (as IPS only covered half the time period).
     • For this new proposed RAPID method the IPS will not be used.
   • Capturing Pensioners Abroad
     • Question raised on whether pensioners living abroad are captured in the data. BG noted that a Census comparison is used, and adjustments are made to count them as emigrated.
     • Action: BG to ensure the MARP paper clearly states whether pensioners abroad and BN(O) individuals are included.
   • Migration Patterns of Naturalised Citizens
     • Comment made that the assumption naturalised citizens migrate in the same way as British nationals is not valid.
     • Action: MM to circulate academic papers to support this point.
   • Comments Raised for MARP and 14/11 Publication:
     • MM asked about current clustering and analytical capability work. Action: AT/BG to confirm if this is ongoing and include in MARP if applicable. If not, consider seeking clarification.
     • JB discussed the value of highlighting incentives for dataset interaction, particularly in relation to data quality and conceptual mapping. He also proposed including conceptual maps for HOBI and RAPID datasets to support user understanding—an idea that aligns with GS’s earlier comment regarding dataset content (e.g., BN(0), passport swipes). Action: AT/BG to assess and, where feasible, incorporate these elements into the MARP and 14/11 publication.

Actions:

1. Arkadiusz Wisniowski raised a query about the potential impact of edge effects on the provisional estimates. He will provide further comments to clarify.
2. Brendan Georgeson to ensure the MARP paper clearly states whether pensioners abroad and BN(O) individuals are included. Consider referencing the definition published in the research update, which will also be needed for the 14 November paper.
3. Marie McAuliffe to forward academic references for Brendan and team and for them to then research a naturalisation adjustment more aligned to the available research.
4. Query was raised regarding current clustering and analytical capability work. Annabelle/Brendan to confirm if such work is ongoing. If so, include relevant details in the MARP paper. If not, consider seeking clarification from the original requester.
5. Jakub Bijak highlighted the importance of including incentives for dataset interaction and proposed adding conceptual maps for HOBI and RAPID to support user understanding—aligning with GS’s earlier comments on dataset content (e.g., BN(0), passport swipes). Annabelle/Brendan to assess and, where appropriate, incorporate these elements into the MARP and 14/11 publication.

Next meeting: Thursday 28 August 2025