Assigning Refugees to Landlords in Sweden: Efficient, Stable, and Maximum Matchings*

• Date: 01 July 2020
• Published date: 01 July 2020
• DOI: http://doi.org/10.1111/sjoe.12362

Scand. J. of Economics 122(3), 937–965, 2020

DOI: 10.1111/sjoe.12362

Assigning Refugees to Landlords in Sweden:

Efficient, Stable, and Maximum Matchings*

TommyAndersson

Lund University,SE-220 07 Lund, Sweden

tommy.andersson@nek.lu.se

Lars Ehlers

Universit´e de Montr´eal, Montr´eal, QC H3C 3J7, Canada

lars.ehlers@umontreal.ca

Abstract

In this paper, we investigate the problem of finding housing for refugees once they have been

granted asylum. In particular, we demonstrate that market design can play an important role in

a partial solution to the problem. More specifically, we investigate a specific matching system,

and we propose an easy-to-implement mechanism that finds an efficient, stable, and maximum

matching. Such a matching guarantees that housing is efficiently providedto a maximum number

of refugees, and that no refugee prefers another specific landlord to their current match when, at

the same time, that specific landlord prefers that refugee to their own current match.

Keywords: forced migration; housing markets; market design; refugees

JEL classification:C71; C78; D71; D78; F22

I. Introduction

The European refugee crisis began in 2015 when a rising number of

refugees made the journey to Europe to seek asylum. The member

states of the European Union received 1.2 million first-time asylum

applications (more than double compared with 2014).1Apart from the

Dublin Regulation, which dictates that the member state in which an asylum

seeker enters first is obliged to render asylum, there has been no systematic

way to divide refugees between the member states. Obviously, this puts

*We are grateful to Christian Basteck, P´eter Bir´o, Francis Bloch, Estelle Cantillion, David

Delacr´etaz, Jens Gudmundsson, Guillaume Haeringer, Bettina Klaus, Flip Klijn, Fuhito Kojima,

Scott Duke Kominers, Vikram Manjunath, Jordi Mass´o,Michael Ostrovsky, Parag Pathak,Marek

Pycia,Alvin Roth,Alexander Teytelboym,William Thomson, Utku ¨

Unver,and the three referees.

Financial support from the Jan Wallander and Tom Hedelius Foundation (P2016–0126:1), the

Ragnar S¨oderbergs Stiftelse (E8/13), SSHRC (Canada), and FRQSC (Qu´ebec) is gratefully

acknowledged.

1Eurostat News, Release 44/2016, 4 March 2016 (https://ec.europa.eu/eurostat/documents/

2995521/7203832/3-04032016-AP- EN.pdf/790eba01-381c-4163- bcd2-a54959b99ed6).

C

The editors of The Scandinavian Journal of Economics 2019.

938 Assigning refugees to landlords

great pressure on member states located at the external border of the

European Union and, more specifically, on Greece, Hungary, and Italy.

In an attempt to reduce pressure on these three member states,

the European Commission decided in September 2015 on a temporary

European relocation scheme for 120,000 refugees who were in need of

international protection.2However, the relocation scheme did not specify

which refugees should be relocated to which member states. This specific

problem has attracted interest among researchers, and more systematic ways

to relocate refugees between European Union member states have been

proposed. In two early papers, Fern´andez-Huertas Moraga and Rapoport

(2014, 2015) approached the problem as a system of tradable quotas – such

as, for example, emissions control. They demonstrated that these quotas can

be designed, based on matching techniques, to solve some specific refugee

resettlement problems. A different matching problem has been proposed by

Jones and Teytelboym (2017). In their system, member states and refugees

submit their preferences – about which refugees they most wish to host and

which state they most wish to be hosted by, respectively – to a centralized

clearing house that matches member states and refugees according to these

preferences.

Even if membership quotas are settled and a centralized matching

relocation system is in place, several obstacles for successful integration

remain. We focus on one of these obstacles: that is, the problem of

finding housing for refugees once they have been relocated to a European

Union membership state, and, in particular, how market design can play

an important role in the solution to the problem. The background to the

housing problem will be described from the perspective of the situation in

Sweden during 2015–2016.

In 2015, the population of Sweden was 9.9 million, which accounted

for around 1.4 percent of the population in Europe. Yet, 12.4 percent of the

asylum seekers in the European Union in 2015 were registered in Sweden,

which made Sweden the state in the European Union with most asylum

seekers per capita.3A refugee who enters Sweden is temporarily placed

at a Migration Board accommodation facility in anticipation of either a

deportation order or a permanent residence permit. The average waiting time

for this decision was 15 months in May 2016.4Refugees who are granted

permanent residence permits are, under Swedish law, entitled to a number

of establishment measures (e.g., accommodation and a monthly allowance),

2European Commission, Statement 15/5697, 22 September 2015 (https://europa.eu/rapid/

press-release STATEMENT-15- 5697 en.htm).

3See the Eurostat News Release cited in footnote1.

4See the web site of the Swedish Migration Board, www.migrationsverket.se/Kontakta-oss/

Tid-till-beslut.html.

C

The editors of The Scandinavian Journal of Economics 2019.