This article estimates the local economic cost of hosting humanitarian migrants (refugees, asylum seekers and subsidiary protection recipients) and the impact of refugee dispersal policies in France.
Refugee dispersal policies—spreading the settlement of refugees across the municipalities—aim to mitigate the costs to natives due to competition for scarce resources such as jobs, housings, and public services (“refugee rivalry”). The authors hypothesize that, if anti-refugee sentiment is driven by cultural insecurity or xenophobia, refugee dispersal policies are unlikely to affect attitudes to refugees. Additionally, natives may respond to the presence of refugees by segregating themselves from refugees, either by avoiding moving into or migrating out of refugee-hosting municipalities.
The authors exploit the opening of new refugee accommodation centers in 98 municipalities in France between 2004 and 2012. They compare local population trends and other economic outcomes in hosting and non-hosting neighboring municipalities up to 2 years before and after the opening of a refugee center. To investigate the longer-term effects, the authors focus on the opening of 76 refugee centers between 2004 and 2010, which enables the analysis to be extended to 48 months.
The analysis is based on: (i) data on the location of refugee accommodation centers from the French Ministry of the Interior; (ii) municipal-level employment and wage data from the Declaration Annuelle des Donnees Sociales (DADS); (iii) municipal-level data on taxable revenues, number of fiscal households, number of retiree households, and tax-paying households from the Impot sur le Revenu par COMmune (IRCOM) database; and (iv) municipal-level data on number of firms, value-added, and value of sales from the Fichier complet unifie de Suse (FICUS) and Fichier approchee des resultats (FARE) datasets.
- The local population declined when a refugee center was opened in a municipality. Two years after the opening of the refugee center, the number of employed residents in hosting municipalities decreases by up to 2 percent. Since the median size of a refugee center relative to the number of employed residents is 0.3 percent, this implies that there are three to six fewer people in the municipality for one refugee place in the center.
- The decline in population is due to fewer people moving into hosting municipalities (‘avoidance behavior’) rather than out-migration of natives from refugee-hosting municipalities. Native avoidance, rather than native flight, appears to drive the decline in population following the opening of a refugee center.
- Wealthier households that pay income tax react more strongly to the opening of refugee centers, possibly because they have more location options.
- Avoidance behavior is not caused by competition in the labor market, as the refugee presence is negligible (0.3 percent of the population). Other results substantiate this inference: (i) employment and wages do not fall significantly following an opening of a refugee center; (ii) there is no effect on the number of non-resident workers in the municipality; (iii) there is no increase in migrants’ hiring after the opening of a refugee center; (iv) there is no heterogeneity of the effect with respect to the job type or wages of the resident and worker populations; and (v) retiree households are affected in the same way as the general population.
- Native avoidance has a negative effect on local economic outcomes. Tax collected by hosting municipalities falls by around 3 percent after opening a refugee center. The economic activity of firms located in the hosting municipalities declined by 3 percent and the value of their sales declined by 5 percent. Overall, the opening of a refugee center reduces aggregate welfare growth by 10-4 percentage points over two years.
Small inflows of refugees at the local level are unlikely to be a credible threat to natives. Nevertheless, natives avoid municipalities where a refugee center has opened, which leads to a decline in the local population, and consequently fewer customers and taxpayers. The authors conclude that refugee dispersal policies can have adverse economic impacts at the local level (reducing the local tax base and economic activity) because of native avoidance. They suggest that policymakers might reduce the cost of housing refugees by addressing the root cause of native avoidance—natives prejudices against refugees—rather than focusing on potential refugee rivalry.