The availability of the highest quality international migration estimates to support effective decision making is crucial. The Office for Statistics Regulation has continued to monitor the Office for National Statistics’ (ONS’s) progress as it works with others to improve migration statistics.

On 21 June, ONS published an update on its population and migration statistics transformation programme and the initial progress report on understanding different migration data sources. We welcome this update for users including the ambitions to be more responsive to user needs and provide more coherent statistics.

The work to improve migration statistics supports understanding of the quality and limitations of existing estimates of migration, as well as information on population by country of birth and labour market status by country of birth.

We note the progress report which begins to explore differences in estimates of EU and non-EU migration between the International Passenger Survey (IPS) and the Annual Population Survey (APS). The APS can be used to estimate the total number of people in the household population by country of birth (the “stock” of migrants), looking at changes over time in the stock provides another way of exploring changes in migration. The ONS Long-Term International Migration (LTIM) estimate (based largely on the IPS, an intentions-based survey) aims to measure the number of migrants entering and leaving the country in any given period (the “flow” of migrants) which can also be used to provide an estimate of net migration.

ONS analysis shows that between 2005 and 2018, the change in the APS “stock” gives a much higher estimate of migration from the EU than the LTIM net migration estimate of “flow” from the EU over the same period (2.1 million versus 1.4 million). However, by contrast, for non-EU migrants the change in the APS “stock” is lower than LTIM net migration estimate of “flow” (2.0 million versus 2.8 million). The scale of difference between the two sources varies by country of origin.

The report highlights that definitional differences between the sources contribute to the divergent patterns, but it is unlikely these differences are significant enough to provide the full explanation.

This research, while still in the early stages, raises concerns about the EU and non-EU splits across the different sources. ONS is undertaking further work to inform its understanding of the quality of these different estimates and plans a further update on this in August. We await ONS’s further work to clarify whether these differences have any relationship to other IPS measurement issues. ONS has separately identified an arrivals/departures imbalance for Chinese visitors, where ONS believes that it is under-reporting Chinese visitors’ departures in its IPS-based, overseas travel and tourism statistics.

We welcome ONS’s plans to provide an update to users in August and hope ONS will continue to work with urgency to resolve the considerable uncertainties around these estimates. We consider that ONS must:

  • complete its research into the differences between the migration estimates derived from the IPS and APS without delay;
  • be transparent and open regarding the outcomes of this research, and continue to inform users about the extent of any quality issues as it becomes aware of them; and
  • continue to clearly signpost quality issues within future statistical publications, providing guidance for users on appropriate interpretation.