Institut für Arbeitsmarkt- und Berufsforschung

Navigation zu den wichtigsten Bereichen.


Zertifikat "audit berufundfamilie"

Inhaltsbereich: Institut für Arbeitsmarkt- und Berufsforschung

Impact of language learning courses on labor market participation of refugees

Project Number


Head of Project

Project Start

May 2017

End of Project

December 2020


The successful integration of refugees in the society and in the labour markets is one of core challenges for the civil society and policy makers. Since the refugee crisis in 2015, numerous labour market policy measures regarding language learning support and vocational training have been implemented in Germany. However, there exists barely hard evidence regarding the effectiveness of these labour market policy measures. Empirical evidence from other countries have in turn shown that good language proficiencies and attending language courses has a positive impact on employment and wages of migrants (cf. Bleakley/Chin 2004, Chiswick/Miller 2003, 2006, Schweigard 2008, Hangartner and Schmid 2017). Thus, language learning assistance as well as an immediate access to language courses have been seen as one of the important policy measures for a successful integration policy (see Hangartner und Sarvimäki 2017 for a review). 

The analysis focuses on refugees from countries with good prospects to remain in Germany (Eritrea, Iraq, Iran, and Syria). We decompose the group of refugees into a control and a treatment group. The treatment group consists of refugees from aforementioned countries of origin, who arrived in Germany before 31.12.2015. Thus, they had the opportunity to attend the exclusive one-off language program of the Federal Employment Office (BA) "Einstiegskurs zu Deutschförderung“ (nach § 421 SGB III), which was offered between October and December 2015. The control group consists of refugees from aforementioned countries of origin, who arrived in German after 31.12.2015. Thus, they did not have the opportunity to participate in the especial language learning program of  the BA. Assuming that both (control and treatment) groups do not on average differ substantially with respect to observable characteristics  (sex, age, education) around the threshold 31.123015, we can estimate precisely and unbiasedly the causal impact of language learning support measures on labor market integration. Furthermore, as a robustness check, we carry out a placebo analysis by considering the refugees from Afghanistan, who were not eligible for the language learning program.