Dynamic Refugee Matching

Asylum seekers are often assigned to a locality in their host country based on uninformed random mechanisms, which do not consider the characteristics of the asylum seekers in the matching process. Consequently, this approach may lead to an inefficient and unfair (“envious”) assignment. This paper proposes a dynamic mechanism for matching asylum seekers to localities, which considers the background of the asylum seeker and consequently the probability of successful integration at different localities. The proposed dynamic mechanism differs from traditional approaches because: (a) the sequential arrival of asylum-seekers introduces dynamics; and (b) the preferences of both the localities and the asylum seekers have to be estimated based on characteristics and historical data. The model is based on priority and rejection assumptions, which determines the assignment of highly demanded and “non-demanded” asylum seekers respectively. Using both theoretical and simulation methods, the authors demonstrate that the proposed mechanism leads to more efficient and less envious matchings.