Immigration and Support for the Welfare State: Local and Institutional Responses

In this project?s four different sub-projects, we have studied how increased immigration is related to attitudes towards immigration, support for anti-immigration parties, and support for features of the welfare state such as trade unions. The project has resulted in six scientific articles, several of which have been published in leading international journals. In the first sub-project, we have studied how immigration has affected election results. In one article, we have looked at how the government's policy spreading the settlement of newly arrived refugees affected election results and political polarization. The policy, which was introduced in November 2013, meant that municipalities that had not previously been asked to settle newly arrived refugees were asked to do so. We find a certain tendency for increased political polarization in these municipalities in the subsequent election. In another article, we have conducted a meta-analysis of the research literature studying how immigration at the local level affects voting for anti-immigrant parties. The analysis reveals both a very small average effect of immigration on anti-immigrant parties? vote shares, and a large variation in this effect across countries and time periods. Furthermore, the study indicates that research that actually finds a positive and statistically significant effect of immigration on such voting is more likely to result in a finished article (so-called publication bias). In the second sub-project, we have studied how labor immigration has affected the tendency to become a union member. The unions have been historically important for the development of the Nordic model, and many have feared that labor immigration will weaken their support. The expansion of the EU area in 2004 affected different types of construction workers differently, in that some - those in occupations without special licensing requirements, such as painters - were exposed to a clear increase in competition with foreign labor, while others - those in occupations with such requirements, such as electricians - did not experience the same increase in competition for jobs. By comparing these groups, we can study how labor market competition from immigrants affected the tendency to organize. We find that the workers in the occupations that were exposed to increased competition had a weaker wage development than with construction workers in non-exposed occupations, but the increased competition had no effect on trade union organization. This study is published in European Union Politics. In the third sub-project, we study how attitudes towards immigration are affected by education and technological change. The idea is to investigate hypotheses related to how the level of education and access to information affect anti-immigrant attitudes. In both cases, we have made use of local responses to changes that were decided by central political authorities, which can thus be regarded as a form of natural experiments. In one article, we studies how the gradual introduction of nine years of compulsory schooling (as opposed to seven years) in Norwegian municipalities in the years 1959-1972 affected attitudes to immigration in adulthood. If education provides a more positive attitude towards immigration, we should expect to find that this expansion of compulsory schooling led to more positive attitudes and support for more immigration-friendly parties among the cohorts who, unlike their predecessors, suddenly had to complete nine years of schooling. However, we do not find that attitudes towards immigration or voting for different parties were affected by this extension in compulsory schooling. This article was published in Electoral Studies. In the last sub-project, we study the political and social integration of immigrants, by looking at processes related to trade union organization and participation in elections. In one article, we show how the low degree of organization among immigrants relative to the majority population is related to the number of years since arrival in Norway and the lower rates of union membership in firms where immigrants typically work. In another article, we study the relationship between the residence municipality of newly arrived quota refugees and their propensity to vote in political elections. We find that the tendency to vote is strongly linked to their first residence municipality. The results show that it is particularly important whether neighbors of the same sex and age vote, which may indicate that the quota refugees are directly affected by people in their social network. This article is published in the American Journal of Political Science.

Prosjektet har gitt viktige bidrag til den akademiske forskningen på samspillet mellom innvandring, integrering og velferdsstatens institusjoner. Det meste av arbeidet i prosjektet er grunnforskningsorientert, men eksempelvis kan resultatene fra studiene av konsekvenser av bosettingspolitikk (delprosjekt 1 og 4) påvirke utformingen av bosettingspolitikken. Alle studiene har en nytteverdi i form av å opplyse den offentlige debatten om sammenhengen mellom innvandring og velferdsstaten.

The main object of the project is to study the impact of immigration on welfare state support. The project is divided into four subprojects. The first subproject studies the relationship between refugee immigration and welfare state support. We do so by exploiting a shift in government placement policies, which implied that the number of municipalities receiving refugees increased. We study how welfare state support were affected in these municipalities, compared to a control group of municipalities. The second subproject studies the relationship between labour immigration and welfare state support. We do so using several empirical approaches: We construct local labour supply shocks based on the socio-economic structure of the municipality, we study how regional immigration affects voters differently depending on where they are located in the income distribution, and we study how immigration influences natives' propensity to unionize. The third subproject studies whether immigration attitudes are influenced by the growing use of the internet as voters' main information source. We rely on the staggered expansion of broadband internet to study whether the expansion of internet broadband is associated with more or less polarization in attitudes. The fourth subproject studies whether immigration has directly influenced government redistribution policies. It does so by examining whether the immigration issue have polarized low income voters to a stronger extent than high income voters, which in important theoretical models imply that governments will redistribute less income for rich to poor. The results from the project will be policy relevant and of high interest to the research community.