My research is mainly about historical trends in voting behavior and public opinion, with a particular interest in matters of class. Feel welcome to contact me if you have questions or want to discuss any of the projects.

The political attitudes of the self-employed in a historical perspective (with Giedo Jansen)

The political attitudes of the self-employed are often on the margins of class analysis, limited to brief, standard observations of their conservative tendencies. Recent scholarship, however, has argued that the self-employed have become a more socially and economically heterogenous group over time, and thus more politically heterogenous. So far, however, these hypotheses have not been empirically investigated. We will conduct the first more extensive historical analysis of the political attitudes of the self-employed, using data from the Swedish National Election Studies from the 1960s to the present.

The political effects of the murder of Olof Palme (with Elisabeth Lindberg)

The murder of Swedish prime minister Olof Palme in 1986 will probably never be definitively solved. One reason that the case continues to engage seems to be underlying assumptions that the murder had considerable political consequences for the Swedish welfare state and for economic policy. In this project, we will scrutinize these ideas, and discuss them in relation to political, economic and historical research. Our preliminary argument is that Palme, had he lived, would actually not have made much of a difference for the development of the Swedish Social Democratic Party in the last years of the 20th century. In part due to his age and possibly decreasing motivation. In part due to his lack of expertise on economic and social policy. In part due to his career-long position as part of the party establishment, which defined both his thinking and his repertoire as a politician. There are good reasons to doubt both that Palme could have changed the course of the party, and that he wanted it.

The causes of lower unionization in later cohorts in Sweden (with Patrik Vulkan, funded by Arena Ide)

Unionization in Sweden has declined since the mid-1990s. In an earlier article, we have established that a significant part of this decline is due to generational replacement. Later cohorts (born after ca 1970) do not join unions to the same degree. We also found that the idea that these generational differences are caused by increasing individualization in later cohorts is not empirically plausible. In this project, we will test furtherhypotheses about the reasons for these cohort differences. The analyses will include further elaboration of differences in sub-groups, with a particular interest in the working class and non-European migrants. We will also look into gender-related, educational, and sectoral differences

Class identification in Sweden: historical trends and the role of mixed-class marriages (with Aino Tiihonen and Moa Frödin Gruneau)

The aim of this project is to create a time-series for the prevalence of class identities (working class and middle class) in Sweden over time, and to analyze the causes of their historical decline. We will combine the Swedish National Election Studies from University of Gothenburg with the Welfare State Surveys from the Department of Sociology at Umeå University, and create a time-series of class identity in Sweden that covers almost 60 years. Our main working hypothesis is that an increasing share of mixed-class marriages has made more people hesitant about their class identity.

Cultural conservatism as cause, rather than consequence, of economic difficulties

A popular hypothesis in the literature on radical right parties is that voting for these parties, and the underlying anti-immigrant attitudes, are a consequence of some form of economic difficulties. This project investigates whether causality also runs in the opposite direction; that anti-immigrant attitudes, and cultural conservatism more generally, can be a cause of economic difficulties. There are several possibilities for such a connection, but this project investigates the more specific hypothesis that cultural conservatism leads to difficulties in the educational system, and eventually in the labor market.

Longer project description from an application to the Swedish Research Council.

The causes and consequences of library closures: a quantitative approach (with Andreas Hedberg)

Since 1995, 25% of all public libraries in Sweden have been closed. Op-ed authors have coined the term 'library death' (biblioteksdöden) to the describe this trend. In this project we will create the Swedish Public Libraries Dataset (SPLD) with data on all public libraries in Sweden over an extended period of time (location, opening hours etc.). We will then combine SPLD with other data sources, and analyze the causes and conseqences of library closures. Our explanatory hypotheses refer to demographic, fiscal, and political factors. On the outcome side, we will look into the consequences for reading habits, voter turnout and trust.

Ticket-splitting and local party organizations (with Filip Sjöstrand)

We study the differences between parties’ results in local and national elections in Sweden. Our point of departure is the finding that while there is a strong correlation between these results, the relationship is often not 1:1. We will test the hypothesis that this can be explained by the local organization: In municipalities where parties have little support, they have difficulties maintaining their organization and finding local candidates to stand for election. In municipalities where parties have strong support, they are more likely to recruit strong candidates that attract additional voters. These mechanisms may be able to explain a considerable share of the increase in ticket-splitting over time, as many parties today have difficulties maintaining their local organizations.