Existing studies – largely based on the analyses of large immigration episodes (e.g. Friedberg 2001, Card 1991), and the experience of the US and the UK (Ottaviano and Peri 2012, Manacorda et al 2012) – show that immigration can be absorbed with small changes in employment and wages of natives.
Theories of native-immigrant complementarities (as described in Lewis 2013) and of efficient task specialisation (as proposed in Peri and Sparber 2009) have been articulated in order to explain those findings. They suggest that:
- The inflow of low-skilled immigrants may encourage natives to upgrade and adjust their jobs taking advantage of immigrant-native complementarity as those two groups specialise in different occupations.
Critics of those studies, however, argue that lacking a genuine random supply shock to the distribution of immigrants and without the ability of following native workers over time one cannot establish the causality of that relationship. Moreover, some argue that European labour markets are different and those results cannot be extended to immigration in Europe.
Our new research provides a cleaner and more convincing test of the causal effect of low-skilled immigrants on labour market outcomes of natives (Foged and Peri 2015). In it we use a panel of all residents of Denmark between 1991 and 2008 and exploit an exogenous dispersion of refugees across Danish municipalities and a later surge in immigrants to track how such exogenous shock affected native workers. We focus especially on the effects on workers at the low end of the wage and income spectrum, specifically the less educated and those who were young and with low job-tenure.
An ideal setting: Dispersal policy and immigration surge in Denmark
Immigrants represented a limited share (three percent or less) of total employment in Denmark until 1994, equally divided between those from EU countries and those from other countries. Refugees were distributed across municipalities between 1986 and 1998 following the Spatial Dispersal Policy (Damm 2009). This policy implied that the Danish Refugee Council, independently of the economic characteristics and preferences of the refugees, distributed them across municipalities based only on information on their nationality and family size. The goal of the dispersal policy was to distribute the total of the refugees uniformly across municipalities and to provide them with housing for a year (hence most of them accepted the offer). Clusters of refugees from specific countries in specific municipalities were generated by the dispersal due to the timing of their arrival and house availability when they arrived. These clusters were completely uncorrelated to the labour market conditions of the municipalities and to the economic characteristics of the immigrants.
Then, beginning in 1995, the presence of non-EU immigrants had a rapid surge. In particular, as shown in Figure 1, immigrant inflows from specific refugee-sending countries experienced a strong and sudden increase due to a sequence of international conflicts (Yugoslavia, Somalia, Afghanistan, and Iraq).
Figure 1. Refugee-country immigrants in Denmark
Notes: Growth in immigrant populations since 1 January 1995 from major source countries for refugee inflows between 1986-1998 and from Eastern Europe.
When the dispersal policy was phased out and family reunification became the main channel of entry between 1995 and 1998, the location of those new immigrants from refugee countries was driven by their inclination to locate near the communities of co-nationals formed earlier by the dispersal policy. This unique combination of events provides us with an ideal random supply shock to immigrants. This shock is represented by the increase of refugee-country immigrants after 1995, distributed to the municipalities that experienced dispersal-driven clustering of these refugee-country immigrants between 1986 and 1995.
Labour market effects on the less educated
The refugee-country immigrants were quite representative of non-European immigrants in terms of their education and skills. Forty to fifty percent of them did not have post-secondary education (only 32% of natives did not). Similarly, the more basic occupations (‘sales and elementary service occupations’) employed 13% of these immigrants and only 4% of the natives. In particular, by measuring the ‘manual skill’ content of occupations, we establish that refugee-country immigrants, as typical of non-European immigrants, were in large part concentrated in manual-intensive occupations.
In our analysis, we first test how non-college educated native workers responded to an increase of refugee-country immigrants.
- We find that, especially for native workers who moved across establishments, refugee-country immigrants spurred significant occupational mobility and increased specialisation into complex jobs, using more intensively analytical and communication skills and less intensively manual skills.
This upgrade to less manual intensive and more complex jobs was accompanied by a significant wage increase. Certainly the high job mobility, facilitated by the flexibility and competitiveness of the Danish labour market, were key catalysts for the observed native workers’ response.
A clean way to visualise the effect on native outcomes is Figure 2 that shows a difference-in-difference representation of the effect of exposure to refugee-country immigration. The figure plots the difference in job complexity and hourly wage between less educated natives in municipalities with large and small (top and bottom quartile) increases in refugee-country immigrants as determined by the dispersal policy and the post-1994 surge (1994 is year 0). We see that the differences in occupation complexity (panel A) and wages (panel E) for low-skilled natives increased significantly after 1994 in favour of the municipalities that received the surge of refugee-country immigrants.
Figure 2. The short- and long-run differences in native outcomes in a high-versus-low immigration municipality
Notes: Parameter estimates and 95% confidence limits on the difference in outcomes between the upper and lower quartile of immigrant exposure.
The figures (and regression results in the study) suggest that the complexity index increased with 3% more and wages increased up to 2% more for low-skilled natives in high exposed municipalities (with a 1.6 percentage point increase in the refugee immigrant share) relative to the less exposed municipalities. This took place over 13 years and it appears to be a permanent positive change.
We then focus on the groups with lower wage and higher probability of unemployment, namely those who were young and had a low-tenure job as of 1994 and we compare their performance between high and low refugee-immigration municipalities. These groups are also those with larger potential lifetime gains from changing occupation and upgrading. And in fact, for those groups, job complexity and wages increased the most. Older (over 45 years of age) and long-tenured workers did not take advantage of immigration complementarity, did not experience a wage growth, and some may have retired earlier.
Summary and concluding remarks
Overall, our study finds that a labour market that encourages occupational mobility and allows low-skilled immigrants can generate an effective mechanism to produce upward wage and skill mobility of less educated natives, especially the young and low-tenure ones.
Card, D (1990) “The impact of the Mariel boatlift on the Miami labor market”, Industrial and Labor Relations Review, ILR Review, Cornell University, vol. 43(2), pages 245-257, January.
Damm, A P (2009), “Determinants of recent immigrants' location choices: quasi-experimental evidence”, Journal of Population Economics 22 (1):145-174.
Foged, M and G Peri (2015), “Immigrants’ Effect on Native Workers: New Analysis on Longitudinal Data”, IZA Discussion Paper No. 8961
Rachel M F (2001), “The Impact Of Mass Migration On The Israeli Labor Market”, The Quarterly Journal of Economics, MIT Press, vol. 116(4), pages 1373-1408, November.
Lewis, E (2013), “Immigration and Production Technology”, Annual Review of Economics 5 (1):165-191.
Manacorda M, A Manning and J Wadsworth (2012), “The Impact Of Immigration On The Structure Of Wages: Theory And Evidence From Britain”, Journal of the European Economic Association, European Economic Association, vol. 10(1), pages 120-151, 02.
Ottaviano, G I P and G Peri (2012), “Rethinking the Effect of Immigration on Wages”, Journal of the European Economic Association 10 (1):152-197.
Peri, G and C Sparber (2009), “Task Specialization, Immigration and Wages”, American Economic Journal: Applied Economics 1 (3):135-169.