Voting as a measure of social inclusion for natives, immigrants and descendants in Sweden Pieter Bevelander International Migration & Ethnic Relations, Malmo & Ravi Pendakur Graduate School of Public & International Affairs, Ottawa

Social Inclusion and Voting in Sweden Background Sweden extended local and provincial voting rights to immigrants with resident status in the 1970s. Policy rationale: –increase political influence, interest and self-esteem among foreign citizens. Citizenship acquisition: –Nordics – after 2 years –Refugees – after 4 years –All other immigrants – after 5 years

Social Inclusion and Voting in Sweden Goals: To what extent do differences in voting participation across immigrant and Swedish-born categories exist? –are they explained by Contextual factors (municipal characteristics) Personal characteristics: –Demographic –socio-economic –human capital attributes and immigration related factors Is citizenship an important factor in determining voter participation?

Social Inclusion and Voting in Sweden Previous Research Human capital attributes are key drivers of voting behaviour –Age, education and labour market characteristics are powerful determinants in explaining voting behaviour (DeSipio 1996; Bass & Casper 2001) Less research done on voter turnout and minority status. –Minorities have lower voting participation rates. Not always reduced over generations (Ramakrishnan & Espenshade 2001) –Asians are less likely to vote than native-born (Bass & Casper 2001; Lien 2004) –Immigrants vote less. Canadian-born minorities display similar voting rates to majority Canadian-born (CES) (Chiu, et al 1994)

Social Inclusion and Voting in Sweden Data Swedish 2006 electoral survey –contains information on individual electoral participation for all those eligible to vote in national, provincial and municipal elections. Matched to: –Registry data from Statistics Sweden, which contains information for every Swedish resident. –Municipal level data from census Sample: –70,932 residents in Sweden, About a third are immigrants (23,678). More than half of the immigrants (12,790) are not citizens but have the right to vote in municipal and provincial elections.

Social Inclusion and Voting in Sweden Model Contextual Factors: –Log of city population –Log of immigrant population –% minorities on council –City employment rate Human capital attributes –Age –Education In Sweden outside Sweden –Marital status –Labour market status –Income quintile –Housing tenure Minority identifiers –Immigrant status (age at immigration) –Immigrant status of spouse –Immigrant status of parents –Place of birth –Years since immigrating Voted in 2006 municipal election Citizenship Eligibility Years since arrival Interaction between the two

Social Inclusion and Voting in Sweden Results from OLS regression (all) demographic characteristics

Results from OLS regression (all) socio-economic characteristics Social Inclusion and Voting in Sweden

Results from OLS regression (all) immigrant characteristics

Social Inclusion and Voting in Sweden Results from instrumented regression (all immigrants including those born in Sweden) NOTE: filled in bars are significant at 0.05

Results from 9 instrumented regressions showing impact of citizenship acquisition Social Inclusion and Voting in Sweden NOTE: filled in bars are significant at 0.05

Social Inclusion and Voting in Sweden Conclusions After controlling for demographic and socio-economic characteristics, acquisition of citizenship makes a real difference to the probability of voting. Years of residency: –Participation takes time – long term immigrants are more likely to vote than short term immigrants. But effect is small compared to that of citizenship. If we want to encourage inclusion/integration policies for immigrants we should be really careful about changes to citizenship policy and immigration intake