The labour market impact of refugee waves

CentrePiece Autumn 2017

Recent research has challenged the consensus that sudden inflows of refugees have little or no impact on natives’ wages and employment, claiming instead that there are uniformly large detrimental effects on natives without school qualifications. Michael Clemens and Jennifer Hunt demonstrate the flaws in this analysis: the labour market impact of immigration is small even on natives with low skill levels.

The recent surge in migration to Europe has brought renewed attention to research on the labour market effects of sudden major inflows of refugees and a reassessment of the findings of four particularly influential studies:

• Card (1990), who finds that a large inflow of Cubans to Miami in 1980 did not affect natives’ wages or unemployment.
• Hunt (1992), who finds that a large flow of refugees from post-independence Algeria to France caused a small increase in native unemployment.
• Friedberg (2001), who finds that a large inflow of post-Soviet Jews to Israel 1990-94 did not reduce natives’ wages.
• And Angrist and Kugler (2003), who find that a surge of Balkan refugees during the 1990s was associated with higher native unemployment across 18 European countries, but who do not interpret the association as causal because it is unstable and statistically insignificant.

Two recent studies have challenged these results by re-analysing all four of the refugee waves. The researchers claim that earlier work obscured uniformly large detrimental effects from all four waves, either by aggregating the affected workers with unaffected workers (Borjas, 2017), through inadequate identification of causality (Borjas and Monras, 2017) or both.

Continue reading (PDF 3pp)

---