In recent decades, various cities in Sweden have implemented school closures as a desegregation strategy. Using simulation models calibrated with administrative data for all primary and lower secondary schools in Stockholm, Sweden, we assess the potential impact of school closure on ethnic school segregation. More specifically, we study how the characteristics of the school to be closed, the local opportunity structure for the displaced students, and the student reallocation criteria influence the effects of school closures on school segregation. Our findings show that the change in ethnic school segregation is highly dependent on reallocation criteria and local opportunity structures. Moreover, they demonstrate that the current practices associated with school closures in large urban areas (i.e., closing minority-dominated schools in minority-dominated neighborhoods) are likely to be ineffective in reducing school segregation, especially when students are reallocated to their nearest school or to schools whose composition resembles that of their former schools.

Drawing on individual-level full-population data from Sweden, spanning four decades,we investigate the joint growth of income inequality and income segregation. We study Sweden asa “least likely” case comparison with the United States, given Sweden’s historically low levels ofinequality and its comprehensive welfare state. Against the background of U.S.-based scholarshipdocumenting a close link between inequality and segregation, our study provides an importantinsight into the universality of this relationship. Using entropy-based segregation measures, weanalyze trends and patterns of income segregation between and within income groups along differentsociodemographic dimensions—migration background and family type. Our findings reveal thatgrowing income inequality in the last 30 years has been accompanied by a sharp uptake in incomesegregation, especially for the bottom quartile of the income distribution who are facing increasingisolation. Income segregation is most extensive for individuals with children in the household, amongwhom it has increased at a higher rate than those without children. Interestingly, income segregationis lower among non-Western minorities than among majority-group Swedes. We conclude thatchanges to the welfare state, liberalization of the housing market, and rapid demographic changeshave led Sweden onto a path that is difficult to distinguish from that taken by the United States

As a result of the negative consequences and persistence of school segregation, its causes have received a great deal of scholarly attention across a range of disciplines. However, the existing research has tended to overlook those aspects of the segregation process that lie beyond the choice of the individual. This thesis concerns itself with the way in which opportunities are influenced and thereby constrained by hard-to-change macro structures such as the spatial distribution of individuals and organizations, individuals’ social network characteristics, and population composition. To this end, the four chapters presented in the thesis focus on the key actors involved (i.e., parents, teachers, and organizations), their decisions, and how their actions are moderated by the structures in which they are embedded. All four chapters use Sweden as the empirical case, and they utilize the rich data provided by Swedish population registers.

Essay I analyzes the role played by parents’ ethnicity-related school preferences in the ethnic segregation of the school system, and shows that opportunity structures are an important moderator of the effects of preferences on school segregation. By combining statistical analyses of the school choices made by all parents of compulsory school students in the Stockholm region during the years 2008 to 2017 with large-scale empirically calibrated agent-based simulations, the study shows that preferences tend to be trumped by opportunities. Although parents have ethnicity-related preferences, and although these preferences vary between different ethnic groups, parental preferences have little impact on the extent of school segregation. The main drivers of school segregation are rather to be found in the ethnic segregation of the housing market and the geographic location of schools.

Given the importance of residential segregation for school segregation, Essay II examines how residential segregation has evolved in Sweden over the last three decades. Building on entropy-based segregation measures, the study analyzes patterns of income segregation between and within income groups along different socio-demographic dimensions —migration background and family type. The findings show that the rise in income inequality witnessed over the last 30 years has been accompanied by a sharp increase in income segregation, especially for those in the bottom quartile of the income distribution. Moreover, the results show that income segregation is more extensive, and has increased at a higher rate, among individuals with children than among individuals with no children.

Essay III examines school segregation from an organizational point of view and focuses on “school closure” as a desegregation mechanism. Using large-scale simulation models calibrated using data on all schools and students in Stockholm municipality, the study examines how the closure of a school affects segregation levels, and how this effect varies with the characteristics of the closed school and the criteria used to allocate students from the closed school to other schools. The analyses show that the degree to which closing a school changes the level of segregation varies considerably between schools, and that the ethnic composition of nearby schools is an important moderator of the effect of a school closure.

Essay IV examines network dependencies in the school choices of teachers in Stockholm’s upper secondary education market during the years 2000 to 2010. Using stochastic actor-oriented models, this study focuses on the factors associated with teachers’ labor market mobility between public- and private-sector schools. Controlling for various school characteristics, such as student and teacher composition, and the work environment, the results show that networks—defined as the affiliation networks formed via links to former and current co-workers—are important for the within- and between-sector school mobility of teachers.

In summary, the four studies together show that individuals’ choices are considerably constrained by their opportunities. A segregated residential market determines the school opportunities available to parents, and the differentiation between public and private schools affects the workplace decisions of teachers. Overall, differences in opportunities moderate the potential effects of individual preferences and tend to reproduce existing segregation patterns. 

Superlinear growth in cities has been explained as an emergent consequence of increased social interactions in dense urban environments. Using geocoded microdata from Swedish population registers, we remove population composition effects from the scaling relation of wage income to test how much of the previously reported superlinear scaling is truly attributable to increased social interconnectivity in cities. The Swedish data confirm the previously reported scaling relations on the aggregate level, but they provide better information on the micromechanisms responsible for them. We find that the standard interpretation of urban scaling is incomplete as social interactions only explain about half of the scaling parameter of wage income and that scaling relations substantively reflect differences in cities sociodemographic composition. Those differences are generated by selective migration of highly productive individuals into larger cities. Big cities grow through their attraction of talent from their hinterlands and the already-privileged benefit disproportionally from urban agglomeration.