Publications, reports, and articles.

Improving refugee integration through data-driven algorithmic assignment (2018)

Posted on:
August 20, 2021

Abstract

"Developed democracies are settling an increased number of refugees, many of whom face challenges integrating into host societies.We developed a flexible data-driven algorithm that assigns refugees across resettlement locations to improve integration outcomes.The algorithm uses a combination of supervised machine learning and optimal matching to discover and leverage  synergies between refugee characteristics and resettlement sites.The algorithm was tested on historical registry data from two countries with different assignment regimes and refugee  populations, the United States and Switzerland.Our approach led to gains of roughly 40 to 70%, on average, in refugees’ employment outcomes relative to current assignment practices.This  approach can provide governments with a practical and cost-efficient policy tool that can be immediately implemented within existing institutional structures."

Key points

  • Refugees frequently remain economically marginalized, with low levels of employment in the years following their arrival
  • The assignment of refugees to different resettlement locations within a host country is one of the first policy decisions made during the resettlement process. It is also one of the most
    consequential in maximizing refugees’ economic integration and self-sufficiency as a first step toward a more comprehensive integration into society.
  • Three sets of factors affect refugee integration: geographical context, personal characteristics, and synergies between geography and personal characteristics
  • Host countries’ current procedures for determining how to allocate refugees across domestic resettlement sites do not fully leverage synergies between refugees and geographic locations
  • The authors have developed a data-driven approach that, in contrast, can be immediately implemented by using existing data to optimize integration outcomes. Their algorithm has three  stages:modeling, mapping, and matching.
  • Their analysis demonstrated large potential improvements, but they did not test the algorithm prospectively.
  • In contrast to more expensive interventions (such as language or job training programs) that are sometimes implemented long after refugees’ arrival, the authors assert that their approach is cost-efficient and implemented before refugees’ arrival, giving them the strongest foundation possible from which to integrate into host societies
Improving refugee integration through data-driven algorithmic assignment (2018)

Summary

The authors developed a flexible data-driven algorithm that assigns refugees across resettlement locations to improve integration outcomes.The algorithm uses a combination of supervised machine learning and optimal matching to discover and leverage  synergies between refugee characteristics and resettlement sites.
arrow-circle-upenter-downmagnifier