A new paper, co-authored by the Department’s Lecturer in Demography Francesco Rampazzo, suggests that the number of European migrants living in the UK decreased in the run up to Brexit, with the most significant decline occurring amongst those aged 20-29 and those with tertiary education.
The article, published in International Migration Review and co-authored by Jakub Bijak, Agnese Vitali, Ingmar Weber and Emilio Zagheni, used data from Facebook to estimate the number of European migrants living in the UK, in the year prior to formal withdrawal from the EU on 31 January 2020.
As early as 2016, the year of the Brexit Referendum, the Office for National Statistics (ONS) was reporting a positive yet declining net migration of EU nationals to the UK. Prior to this, the UK had seen high levels of net migration from the EU.
This paper aims to support ONS statistics using digital trace data, which refers to the digital footprints left behind by individuals as they interact with websites and online platforms.
This data provides a level of disaggregation not found in traditional data sources such as surveys. It is also useful for its timeliness – it can be used to promptly measure even minute shifts in populations.
This is particularly useful when researching the volatile processes of migration. As potential migrants weigh up the risks and opportunities of moving, this decision can be rapidly affected by real-time events which can either impede or accelerate migration.
Utilising the Facebook Advertising Platform, this paper used data collected each week on European migrants in the UK, between March 2019 and March 2020.
The collected data was disaggregated according to age, level of education, and country of origin. This paper focused on information on individuals over the age of 15, from France, Germany, Ireland, Italy, Latvia, Portugal, Poland, Romania, and Spain – the top nine European countries of origin of migrants residing in the UK.
The data indicates that overall numbers of EU migrants in the UK decreased during this period. The age groups showing the most significant decline were individuals aged 20-29 and 30-39, which also had the highest initial numbers of migrants.
Based on the presented trends, it appears that leading up to Brexit, the UK experienced a significant decrease in young and highly educated migrants from other European countries.
The paper also found that declines were present across all education levels. However, migration decreased especially amongst tertiary-educated individuals, despite new immigration criteria favouring this educational class.
Variations amongst countries were observed, with the largest and most recent groups of migrants (such as those from Poland and Romania) being most affected.
These declines appear to persist beyond the date of March 2020, although the paper notes that data after this time will have been affected by the coronavirus pandemic and subsequent lockdowns.
The trends indicated by this digital data are further confirmed by the results of the Labour Force Survey. Additionally, recent estimates of National Insurance Numbers indicate a declining trend in the number of new EU migrant social security registrations compared to previous years.
The paper concludes that digital trace data can provide timely information which may be unavailable from official statistics sources – in this case, the educational and age profile of migrants by country of origin.
However, it cautions that this digital data can be prone to instability, for example, when changes in algorithm prevent researchers from comparing data over longer periods of time.
While digital trace data cannot entirely replace traditional data sources, it nevertheless can be used as a complementary source providing valuable additional insights.
