Population and migration statistics system transformation – recent updates

Migration Statistics Quarterly Report (MSQR): August 2020, including long-term international migration estimates for the year ending March 2020

The latest long-term international migration estimates are based on International Passenger Survey (IPS) data for the period up to March 2020. The period in which data were collected from the IPS covers up to 16 March 2020. The IPS was suspended on this date because of the coronavirus (COVID-19). The weighting methodology has been adapted to compensate for these missing data. This compensation means that the reporting period for this release remains unchanged.

As a result of the Migration Statistics Transformation Journey, the IPS being stretched beyond its original purpose and the impact of the coronavirus, this will be the last MSQR based on IPS estimates. Administrative data will be used to deliver new measures of migration from November 2020 onwards.

Processing and quality assurance of the IPS data for the first quarter of 2020 identified a possible issue in the data, namely an unexpected large rise in the number of non-EU students interviewed in the IPS.

A detailed review of the data and associated processes suggests this may be because of the IPS being susceptible to sampling variability. This observation aligns with previous analysis, which identified that large numbers of students may travel at the same time because of course start dates, meaning that some IPS shifts may include clusters of student contacts. If this was to occur the IPS data may overestimate the number of long-term student migrants. This is a limitation of using a sample survey.

Uncertainty of this type in the data is represented by the confidence intervals around the estimates, which can be viewed in the associated data tables. Because of this oversampling, an adjustment, in line with previous adjustments, has been applied to non-EU student immigration for year ending Quarter 1 (Jan to Mar) 2020 estimates using Home Office (HO) visa allocations, this ensures we are providing the best possible estimates of migration based on all available sources. See Section 6 of the statistical bulletin for more detail.

As it is recognised that international travel patterns have changed significantly in 2020 because of the coronavirus pandemic, we have also included in this release additional data and insights on recent travel patterns in the period up to the end of June 2020. See Section 8 of the statistical bulletin for more detail.

This release presents different migration-related data sources at local authority level including migration flows, non-UK-born and non-British populations, National Insurance number registrations, GP registrations, and births to non-UK-born mothers.

We have reduced the content of our Short-Term International Migration (STIM) publication to focus on our migration transformation work and the development of these statistics. Further information can be found in Short-Term International Migration for England and Wales: year ending June 2018. As a result, the STIM table normally included in Local area migration indicators (LAMIS) is not currently being produced. This means that the STIM data included in the LAMIS 2020 release is the mid-2016 to mid-2017 provisional data (previously published in the LAMIS 2019 release).

We welcome any thoughts you have on the development of STIM estimates in the future. If you would like to get in touch, please contact us by email at pop.info@ons.gov.uk.

It was announced alongside the May 2020 MSQR that the August 2020 MSQR is the last set of long-term international migration estimates based upon IPS data.

As part of our transformation journey to make use of all available data sources to provide a richer and deeper understanding of migration, November 2020 will see the first admin-based migration estimates (ABME) report published. Whilst the ABME report will replace the MSQR publication, it is important to recognise that the move to ABME reports marks a change in how we will report migration estimates.

As we progress our ambition to maximise the value and benefits of admin data, the initial ABME reports will not report migration data in the same way as the MSQR. The current availability of admin data means that the time periods reported upon in the MSQR, and some of the underlying detail such as reason for migration, will not be included in our initial ABME reports. However, it is important to recognise that this short-term change in content is an important step towards the delivery of a transformed migration system, one which makes best use of all available data sources and underpins a new robust, sustainable and coherent population and migration statistics system.

Our ABME reports will demonstrate iterative progress towards a transformed migration system and will provide users with clear messages on our next steps and challenges. As such the ABME reports will, like the MSQR, be classified as Experimental Statistics. As our work progresses we look forward to being able to produce greater insights into migration patterns and statistics at an ever more granular level.

Ahead of November 2020, we plan to publish:

• analysis of non-UK employees in the UK labour market, including those who could be considered as potential "key workers" in the response to the coronavirus pandemic

• a synthesis report bringing together a range of insights to highlight the impact the coronavirus pandemic has had on international migration patterns so far

The November 2020 ABME report will not provide a net migration figure for the year ending June 2020. However, given the suspension of the IPS because of the coronavirus, it would not have been possible to provide a net migration figure for this period under the MSQR approach either. The impact of the coronavirus, complexities of sampling and the suspension of the IPS, has highlighted the challenges of the current migration system, and accelerated our plans to harness the power of admin data and progress our transformation journey.

The net migration figure for the year ending June 2020, usually published in November, would normally feed into the calculation of our mid-year population estimates. We are currently working with our colleagues in Population Statistics to agree a suitable method for providing data for this period.

The November ABME report will focus on providing an overview of main data sources using an admin data quality framework. Alongside this, the report will also provide our first iteration of admin-based migration estimates, alongside analysis that compares our new approach with the existing back series of Long-Term International Migration (LTIM) estimates.

We welcome your feedback on our latest updates and on our transformation journey. If you would like to get in touch, please contact us by email at pop.info@ons.gov.uk.