Interview with Giacomo Solano
Interview to our researcher Giacomo Solano (MPG) on the Pilot Report “Targeted technical support to implementation of Action ‘Facilitating Evidence-based Integration in Cities’”, coordinated by MPG and the KU Leuven University for the Partnership on the Inclusion of Migrants and Refugees of the Urban Agenda for the EU.
The report explores how integration outcomes differ on a regional level in EU, and what regional characteristics influence these outcomes.
1) What is the main objective of the research?
Scholars and policy makers have long focused predominantly on the national level when analysing migration and integration policy. National-scale outcomes reflect and inform the development of national integration policies and enable cross-national comparisons, but they fail to adequately capture spatial differences within countries.
The analyses and results presented in this paper are a first step to showcasing the newly available comparative data on infranational level from Eurostat, in making meaningful comparisons in education and labour market integration outcomes across cities and regions. The overall aim of this exercise was to understand how EU regions (NUTS-2) differ concerning integration outcomes of migrants, and how national-level integration policies and certain regional characteristics (such as GDP, net migration, population or competitiveness) influence the migrants’ integration outcomes.
2) You say that there is “newly available comparative data”. Can you be more specific?
EUROSTAT has been testing the possibility of publication, to the widest possible extent, of the existing EU integration indicators on a regional level (NUTS-2 level) and by ‘degree of urbanisation’ (cities, towns and suburbs, rural areas). The feasibility testing has resulted in the recent publication of new indicators for most classic and robust indicators as part of the Eurostat migrant integration database (employment regional series). Statistics on activity rate, employment rate, unemployment rate and educational attainment and young people neither in employment nor in education or training (NEET) are now available to be disaggregated by country of birth and country of citizenship at regional level and by degree of urbanisation (cities, towns and suburbs, rural areas).
This data allowed us to group regions based on (1) heterogeneity of integration and (2) on characteristics of NUTS-2 in the EU. Clustering regions provides optimal groups of similar units allowing NUTS2 regions within clusters to learn more easily from those that are more similar.
3) What indicators of integration did you take into consideration?
At the beginning we included Zaragoza’s five indicators, but because of lack of data we decided to include just the three most relevant:
- Activity rate is defined as the percentage of the population in a given age group who are economically active. According to the definitions of the International Labour Organisation (ILO) people are classified as employed, unemployed and economically inactive for the purposes of labour market statistics. The economically active population (also called labour force) is the sum of employed and unemployed persons. Inactive persons are those who, during the reference week, were neither employed nor unemployed.
- The employment rate is calculated by dividing the number of persons aged 20 to 64 in employment by the total population of the same age group. Employed population consists of those persons who during the reference week did any work for pay or profit for at least one hour or were not working but had jobs from which they were temporarily absent.
- Finally, the share of tertiary educated is defined as the percentage of the population aged 30-34 who have successfully completed tertiary studies (e.g. university, higher technical institution, etc.).
The regions on the other side have been classified based on their foreign born population, net migration, GDP, and their competitiveness (‘the ability of a region to offer an attractive and sustainable environment for firms and residents to live and work’) which takes into account both business success and personal well-being.
4) How was the study conducted?
We ran cluster analysis in order to find specific correlations between integration indicators and regions’ characteristics. In relation to the variables, we explored the gap between EU-28 and non-EU-28 migrants, between EU-28 migrants and natives, and between non-EU-28 migrants and natives – each time for activity rate, employment rate, and share of tertiary educated.
The ambition was to investigate whether integration outcomes reveal a clear pattern of differences between regions.
So, first, we grouped regions based on integration gaps, secondly on regional characteristics, and finally we combined the results of the two grouping processes.
When we grouped regions based on integration gaps, two clusters emerged.
Cluster 1 is characterised by more favourable integration outcomes for non-EU-28 migrants (versus both EU-28 migrants and natives) in all three integration outcomes under consideration. Non-EU-28 migrants in this cluster region have favourable or less unfavourable (compared to Cluster 2) integration outcomes (versus both EU-28 migrants and natives) in all three integration outcomes. Furthermore, this cluster is characterised by more favourable outcomes for EU-28 migrants as opposed to natives (compared to the situation of regions in Cluster 2).
Cluster 1 includes Porto, Braga, London, Napoli, Rome, Dublin Cluster 2 presents a different picture: non-EU-28 migrants have far worse integration outcomes than natives and EU-28 migrants in this cluster, in all three integration outcomes under consideration. It is interesting to notice that EU-28 migrants have also more favourable outcomes in terms of education gap as opposed to natives, while they fall behind natives on activity and employment rate.
Cluster 2 includes Brussels, Antwerp, Copenhagen, Berlin, Catalonia, Madrid, South Holland, Stockholm, South Sweden.
We then grouped regions based on regional characteristics, in order to observe whether they would reveal a pattern between NUTS-2 regions. Based on characteristics such as regional typology, GDP, Population, net migration, foreign-bord population and regional competitiveness index (RCI) e obtained other two clusters.
Cluster 1 is made up of mostly urban regions, with high GDP and RCI, diverse population. It includes Vienna, Brussels, Berlin, Catalonia, Stockholm. On the other side Cluster 2 emerged as mostly rural, with low GDP and RCI, and low diversity. It includes Canary Islands, Southern Ireland, West Midlands.
5) What happened when you combined the two groups?
We could categorise four different situations:
In high-competitive and diverse mostly urban regions we can have two scenarios: one where non-EU migrants tend to be less educated and active in the labour market than natives (which applies to Prague, Budapest, northern Italy, all NUTS- regions in Austria, Denmark, Netherlands), and one where non-EU migrants tend to be more educated and active in the labour market (or less disadvantaged than in the other group) than natives (for example in Dublin, Malta, Luxembourg, London, Nuremberg, Cyprus).
Similarly, also low-competitive and non-diverse mostly rural regions can open to different scenarios: in regions such as southern Czech Republic, all NUTS2-regions in Estonia, northern Spain, southern Italy non-EU migrants tend to be more educated and active in the labour market than natives (or less disadvantaged compared to regions in other situations), whereas in northern Greece, central and north-east Spain, northern Croatia, southern Croatia, eastern France non-EU migrants tend to be less educated and active in the labour market than natives.
Most of the studied regions fall into the situation in which high-competitive and diverse mostly urban regions where non-EU migrants tend to be less educated and active in the labour market than natives.
6) How can we explain these results and do policies have a role in this outcome?
When we look at the association between integration policy indicators and integration outcomes, several trends emerge.The results show that migrant integration policies do not influence integration outcomes in low-competitive, rather homogeneous and mostly rural regions. By contrast, integration policies affect labour market integration of migrants in high-competitive and diverse urban regions. First, there was a negative association between policies and outcomes: better policies are associated with more negative outcomes for migrants as opposed to country nationals. Second, a shift to more inclusive policies over the years led to a reduction of the employment gap between migrants and nationals. This suggests that more inclusive policies may be a reaction to a wider gap between country nationals and migrants.