In the wake of Brexit, it is widely recognised that the UK and Ireland, in the aftermath of the eastern enlargement, received a large wave of migrants (e.g. Portes 2018). It is less known that some other countries in Europe – Norway and Switzerland – received many more immigrants relative to their population sizes. The foreign-born population in Norway, relative to the total population, almost doubled between 2004 and 2013 from 8% to 14% (OECD 2019). The UK immigrant share rose by roughly 30% (from 9% to 12%). The immigration wave to Norway after the EU eastern enlargement constitutes one of the largest immigration shocks of the 21st century worldwide.
In a recent paper (Bratsberg et al. 2019), we ask what the impact is of such a large immigration-induced labour-supply shock on occupational wages, labour costs, and the industry mix of the economy. The impact of immigration on labour markets has received substantial attention over the last decades. However, most studies focus on the wage structure (e.g. Dustmann et al. 2016). Evidence on the general equilibrium adjustment of occupational wages, labour costs, and industry employment in response to immigration shocks is still relatively scant. We set out to close this gap using high-quality and detailed administrative Norwegian data.
The eastern enlargement in 2004 and 2007 extended the common European labour market to include roughly 100 million individuals from the EU accession countries. With real wages among the highest and unemployment among the lowest in Europe, Norway became a popular destination for labour migrants.
Over the ensuing decade, Norway stands out as one of the countries that received the largest inflows of migrants relative to country size. Figure 1 shows how the number of immigrant employees relative to total employment in the private sector in Norway developed over the period 2000 to 2015.1 Immigrant employees are split into refugees, family reunion from developing countries, education/work from developing countries, old EU/OECD countries, and new EU countries. The immigrant share rose from 7% to 17% between 2004 and 2013. About 60% of migrants came from EU accession countries. Before the Eastern enlargement of the EU, Norway had hardly received any migration from the accession countries.2
Figure 1 Immigrant employment shares, private sector, 2000 to 2015.
In addition to the sheer magnitude of the immigration shock, the Norwegian case is particularly useful to study because the policy change was exogenous. As a part of the single market, but not a member of the EU, Norway is bound to adopt EU legislation without representation in the European Parliament and Commission. The policy change was instant, comprehensive, and externally imposed, providing a unique setting to study the impact of immigration.
If immigrants sorted themselves into a similar distribution of occupations as the native population, we would not expect any change in relative wages or the industry mix. However, this was far from the case. We document that immigrants were highly concentrated in certain types of occupations and industries.
In line with standard trade and labour market theory, our hypothesis is that an immigrant supply shock lowers relative wages in those occupations into which immigrants enter. As a consequence, the industries that are most intensive in the use of occupations exposed to immigration will expand, while at the same time, relative wage costs in these industries will decline.
Proceeding to the empirical analysis, a major identification challenge is that the change in occupation-specific labour supply due to immigration is endogenous. For example, labour supply might be determined by demand factors. Since Altonji and Card (1991), many studies have relied on historical settlements to identify the effects of immigration on outcomes of residents in the destination country. However, due to the political divide of Europe after WWII, there was hardly any migration from Eastern Europe to Norway before the EU enlargement. Our analysis, therefore, requires a different approach.
To identify the causal effect of immigration, we propose a new instrument and methodology to overcome the identification problem. We exploit the fact that occupations differ in terms of language requirements, which represents a barrier to foreign workers since Norwegian is the typical workplace language. Our instrument is inspired by previous empirical evidence showing that immigrants sort into occupations that are less language intensive (Peri and Sparber 2009).
After opening the floodgates in 2004, we therefore expect more immigrants to sort into less-language-intensive occupations. For example, working as a daycare worker – a language-intensive occupation – is relatively less attractive for immigrants as it requires extensive local-language skills. On the other hand, working as a carpenter – a relatively low language intensity occupation – requires only rudimentary Norwegian skills.
Based on the Norwegian data, we observe that the relationship between the initial level of, and the change in, the immigrant share and language intensity is strong. According to our estimates, the change in the immigrant share is 11 percentage points lower in language-intensive versus non-intensive occupations (comparing the 90th versus 10th percentile) over the 2004-2013 period.
According to our results, labour immigration leads to large adjustments in relative industry employment and labour costs. These effects are particularly strong in industries that are initially intensive in the use of immigrant-heavy occupations. In line with our hypothesis, this can be traced back to movements in relative occupation wages: occupations with a large increase in labour supply faced 18% lower wage growth compared to occupations with a small increase (comparing the 90th versus 10th percentile) over the same 10-year period.
As is well known, a reduced-form approach can only identify relative effects – the common effect of immigration across all occupations and industries is not identified. To address the real wage and overall welfare effects of the migration shock, we therefore quantify the general equilibrium effects of immigration according to our calibrated model. The counterfactual analysis shows substantial real-wage losses in some occupations, whereas other occupations have real-wage gains. Although real wages in some occupations decline, the aggregate welfare effects of the immigration shock on natives are close to zero, as some natives switch to higher-wage occupations in response to the immigration shock. The welfare effect on the existing population of immigrants, on the other hand, is negative, as they have a comparative advantage in low-wage occupations.
According to the Gallup World Poll (Esipova et al. 2011), 630 million adults or around 14% of the world population would like to move. It goes without saying that understanding the impact of immigration on the receiving economies is essential. We develop a new methodology for estimating the impact of a large immigration shock on industry growth, labour costs, and occupational wages. We envision that this framework can be used in many different contexts to analyse the effect of migration following major disruptions—both for other time periods and other countries.
Altonji, J, and D Card (1991), “The effect of immigration on the labour market outcomes of less-skilled natives”, in J M Abowd and R B Freeman (eds.), Immigration, Trade and the Labour Market, Chicago: University of Chicago Press.
Bratsberg, B, A Moxnes, O Raaum and K H Ulltveit-Moe (2019), “Opening the floodgates: Industry and occupation adjustments to labour migration”, CEPR Discussion Paper 13670.
Dustmann, C, U Schoenberg and J Stuhler (2016), “The impact of immigration: Why do studies reach such different results?”, Journal of Economic Perspectives 30(4): 31–56.
Esipova, N, J Ray and A Pugliese (2011), “Gallup World Poll: The many faces of global migration”, IOM Migration Research Series no. 43.
OECD (2019), “Foreign-born population” (indicator), OECD iLibrary.
Portes, J (2018), “The economic impacts of immigration to the UK”, VoxEU.org, 6 April.
Peri, G, and C Sparber (2009), “Task specialization, immigration and wages”, American Economic Journal: Applied Economics 1(3): 135–169.
 Employees between the age of 20 and 61 are included.
 Before 2004, accession country citizens had very limited access to the Norwegian labour market. Work permits were provided via domestic employers in need of specialist competence, or on a temporary three-month seasonal basis, typically for agricultural work.