Over the course of the INTEGRATE project, we have performed work on a range of national asylum and integration policies. We have published several peer-reviewed papers in leading peer-reviewed journals and obtained results on a range of asylum and integration policies in Europe.
In a first paper published in PNAS, we show that recently arrived refugees in Switzerland were more likely to be employed if they lived close to a larger group of people who share their nationality, ethnicity or language. The publication was covered by several news outlets in Switzerland, the US and in Sweden as well as cited in a World Bank report (Schuettler, Kirsten; Caron, Laura. 2020. Jobs Interventions for Refugees and Internally Displaced Persons. Jobs Working Paper; No. 47. World Bank, Washington, DC).
Also in Switzerland, we studied welfare benefit payments and tested whether high welfare benefit municipalities serve as “welfare magnets”. We find very limited evidence that this is the case. The findings were published in the American Journal of Political Science and were presented at academic conferences, in a podcast, and via a press release.
We have further looked into the effects of labour market restrictions and find that restricting refugees' employment opportunities reduces their labor market integration and earnings in the long run. The findings were covered in Swiss news outlets and will be shared further once the paper, which is currently under review, is published.
In France, we have studied the effects of integration contracts and find the integration contract facilitated employment in the short term without backlash but did not translate into long-lasting integration gains. The paper is forthcoming in the American Journal of Political Science. The publication will be accompanied by various dissemination efforts, incl. a press release, and are shared with the Office Français de Protection des Réfugiés et Apatrides (OFPRA), the French government office responsible for processing asylum applications.
In Germany, we have examined the impact of two publicly funded language programs in response to the large increase in the number of asylum seekers in 2015: a rapidly developed, ad hoc program that offered basic language training to over 230,000 newly arrived refugees and a smaller, preexisting program that offered refugees comprehensive language training. We find that the more comprehensive, preexisting program increased refugee employment but document no discernable benefits for the ad hoc program. This paper is currently under review.
In another project conducted in Germany, prompted by the crisis in Ukraine, we have investigated whether private hosting of Ukrainian refugees improved their integration outcomes. We find that private accommodation, in comparison to public asylum centers, improves refugees’ social, psychological and navigational integration outcomes. As soon as the two papers are published, we will disseminate the findings to policy makers in Germany and Europe as well as to media outlets internationally.
To understand how public sentiment about asylum seekers has changed through Europe’s two major recent waves of refugee migration, we have conducted a large-scale public opinion survey among approximately 15,000 vote-eligible citizens in 15 European countries. Our findings suggest that European support for asylum seekers remained remarkably stable across the Syrian (2015–16) and Ukrainian (2022) refugee protection crises. The paper was published in Nature and was picked up by several international news outlets.
Finally, we have been running a large-scale randomized control trial to evaluate the benefits of data-driven, algorithmic geographic placement to optimize employment outcomes. We have presented the results at various occasions, engaging stakeholders such as the State Secretariate for Migration in Switzerland and the World Bank. Additionally, the Dutch asylum administration (COA) has sought our expertise, requesting a series of backtests in order to gauge the potential of algorithmic matching within the Dutch context.