Measuring Success: Will We Ever Know How Many Eligible EU Citizens Did Not Apply for Settled Status?

As the government prepares its scheme for granting EU nationals settled status after Brexit, its goal has been to develop a simple, easy-to-use application system in which the vast majority of applicants should be successful.

As the Migration Observatory has pointed out, how many EU citizens and their family members secure settled status will depend not just on how many applicants are rejected, but also on how many people know that they need to apply. A whole host of factors, from lack of awareness to fear of rejection to simple disorganisation mean that some eligible EU citizens will not apply.  Understanding whether people engage with the settled status process is an important policy question because EU citizens’ legal status in the UK depends on it.

As the scheme is implemented, there will be great interest in how well the UK is doing in getting EU nationals documentation of their status. But working this out will not necessarily be straightforward and could depend heavily on what data are made available. This commentary looks at what we will know about the progress of the scheme and points out that:

• Based on currently available statistics, it will not be easy to calculate exactly how many people miss out on settled status unless the numbers are very big. This is because we do not have precise figures on how many EU citizens are living in the UK and plan to stay. It is possible that tens of thousands EU citizens living in the UK could lose their legal status without this being clear from the data.
• Our understanding of how effectively the scheme is including different groups of applicants will depend on how much data the government decides to collect and release. With advance planning across government, it should be possible to produce a rich set of statistics that will help policymakers and the public understand how the settled status scheme is working.

What share of the EU citizen population will eventually gain settled status – and how will we know?

In principle it should be possible to assess whether the Home Office is on track to document the whole population of EU citizens living in the UK—and whether it has done so once the deadline has passed—by looking at how many of grants of settled status or pre-settled status have been made at any given point in time after the scheme opens. (Pre-settled status will be granted to people who arrive in the UK before the proposed cut-off of December 2020, but do not yet have 5 years of residence in the UK when they apply.)

In practice this will not be entirely straightforward, for several reasons.

First, before the deadline, the number of EU citizens applying will probably not be evenly distributed over time. Potential applicants will be motivated by factors like salience and deadlines, which means we might expect spikes in applicant numbers when the scheme first opens and as the deadline approaches.  A lot may ride on what happens in the scheme’s final months or weeks.

Second, even after the deadline, we will not necessarily have a precise figure for the number of EU citizens who are living in the UK and who plan to stay.

On one hand, not all EU citizens living in the UK at the point of the cut-off date, which is currently scheduled for December 2020, will stay permanently. Emigration figures have limitations but in the past they have suggested that emigration is approximately half of initial immigration, with most of those who leave doing so within 5 years (IPS table 3.15). If we measure the EU citizen population at the time of the cut-off date, this will include an unknown number people who have not applied because they are shortly planning to leave. It may also include people who are not expected to apply for settled status because they arrived after the cut-off date on a new residence or work permit implemented as part of any post-Brexit immigration system.

In addition, there is no ‘list’ of EU citizens living in the UK and estimates of their numbers are not precise. This is because annual population estimates—which are updated every 6 months—come with margins of error and do not count everyone (i.e. they exclude those living in communal establishments such as hostels or care homes). Some groups may also be undercounted due to the declining response rate in the Labour Force Survey (LFS), underlying ONS survey that is used to produce the estimates.

In 2017, for example, the ONS estimated that there were 3,813,000 EU citizens living in the UK, but this estimate came with a confidence interval of ±92,000 on either side (note that these figures include an estimated 350,000 Irish citizens who will not need to apply for settled status).This uncertainty is in addition to unknown levels of error in the estimates due to non-response.

Helpfully, there is a Census scheduled for March 2021, shortly after the cut-off date for EU citizens to arrive in the UK and a few months before the planned deadline for applying for settled status. This will give a more accurate count of EU citizens living in the UK, although here too there are limitations. The data will not become available for up to 2 years, and the Census is likely to ask for information about what passports people hold, rather than their nationality (as it did in 2011). This means that we may not know the nationality of people without passports. In the 2011 Census for England and Wales, there were approximately 100,000 people born in EU countries who reported not having a passport, in addition to UK-born EU citizens with no passport who cannot be separately identified in the Census.

If the transition period is extended as many have suggested it will be, the cut-off date for EU citizens to arrive in the UK and apply for status will presumably be extended too. In that scenario we would again need to rely on population estimates of the size of the EU citizen population, rather than the Census.

In summary, if the number of successful applications falls well short of the estimated population of EU citizens and family members, it will be clear—for example, if there were only 2 million people granted some form of status by the deadline. But if the gap is ‘only’ in the tens of thousands, it may not be possible to estimate with any precision how many eligible people who are planning to remain in the UK have not secured the documentation they need to do so.

What can be done to get good statistics in place?

Getting a full, accurate picture of who is and isn’t receiving settled status will inevitably be a tricky process. It will not be possibly to fully resolve the data problems described above, and all migration statistics will have limitations.

However, there are some relatively straightforward options for improving the data that could make a big difference to our understanding of the settled status scheme. These involve (1) developing and releasing detailed administrative (i.e. operational) data from the application process itself; and (2) collecting new data from EU citizens on whether they have applied and secured settled status.

First, if the government releases detailed demographic information on the characteristics of applicants, it should be possible to identify groups that appear to be significantly underrepresented. Currently, most immigration statistics are provided by nationality only. However, much more demographic information should in principle be available that, if processed for statistical purposes and regularly published, would provide a valuable demographic profile of who is and isn’t applying, such as:

• Age;
• Gender;
• Location of residence within the UK;
• Whether they previously held permanent residence documents;
• Year of arrival or years of residence documented for the application.

Other data that would facilitate understanding of how the scheme is functioning include:

• How many people receive settled status vs. pre-settled status, and how many transition from the former to the latter;
• Numbers of refusals and reasons for refusal;
• Reasons for cases being classified as ‘complex’ or requiring additional information (e.g. due to limited evidence of residence or criminal records);
• Reviews or appeals and their outcomes.

Second, the government could arrange for new data to be collected to estimate the share of EU citizens who have and haven’t received settled status. One way to do this would be to add an ad hoc question to a large-scale survey that is already being collected, like the LFS.

The LFS, which is the main survey used to understand the characteristics of the UK population, contains information on respondents’ nationality and country of birth, but does not currently ask any questions about non-UK nationals’ residence status. Some other countries do collect such data. Canada, for example, asks LFS respondents whether they have permanent residence rights, and Australia asks a series of questions once every three years about whether people are temporary or permanent residents and what kind of visa they hold.

While LFS data have limitations, including some simple questions on whether respondents have settled status (and/or other kinds of residence status), either periodically or every quarter, would provide a rich source of information that could be quite important for policy discussions.

Getting new questions into the LFS takes time. Because of pressure to keep the survey short and manageable for respondents, the process of adding questions is tightly managed through an annual LFS Steering Group meeting which consists of a range of government bodies and the ONS. If a strong case is made for a new variable, the Steering Group can recommend it in principle. The new questions then need to be tested (e.g. for wording and position in the survey) and presented for a final decision at the next annual meeting. This means that the process can take at least 18 months.

Together with the operational data discussed above, new data collection along these lines could be used to understand whether there are particular groups of people who are less likely to have sought settled status and what their characteristics are—such as women, the elderly, long-term residents, or particular nationalities. If available while the scheme is still open for applications, this could inform efforts to target information campaigns and assistance to those who are yet to apply. In addition to the government’s own communications campaign, a wide array of other organisations including charities, unions and local government are likely to be involved in this effort and will want access to the data.

Conclusion

New data sources take time to develop (e.g. due to the need to agree, test and schedule any new survey questions, or ensure that the IT system handling applications will generate detailed and usable figures). But if planning starts well in advance, it should in principle be possible to put in place a carefully designed set of statistics to ensure that policymakers can evaluate one of the biggest programmes the immigration system has had to handle.

Funding for this analysis comes from the UK in a Changing Europe’s Brexit research project which is funded by the Economic and Social Research Council. The Migration Observatory’s work on post-Brexit immigration policy is also made possible with the support of Trust for London and the Paul Hamlyn Foundation. This commentary also draws on analysis of survey data supported by Oxford University’s John Fell Fund.

Thanks to Jonathan Portes, Ruth Gripper, Carlos Vargas-Silva and two anonymous reviewers for comments on a previous draft.