Abstract
Learning the host country language is an important factor behind refugees' integration outcomes. At the same time, learning a new language can be very difficult and often refugees do not achieve fluency. Many factors enter the decision about whether and how much to invest in learning the language. In this project, we focus on beliefs about the labor market return to host country language proficiency. We, first, elicit the extent of misperceptions about these returns. Then, we provide an information treatment to half of our sample about the actual correlation between Danish proficiency and earnings among refugees in the Danish labor market. Lastly, we ask how many hours refugees would like to invest in learning Danish. This information experiment is embedded in an online survey that we run among Ukrainian refugees in Denmark (DARECO, wave 3). The survey is sent out and administered by Statistics Denmark. Once we close the data collection, Statistics Denmark spends 1-2 weeks preparing a pseudoanonymized version that we will get access to and be able to merge to population registers, including actual enrollment in language classes and labor market participation.