[Séminaire conjoint] The Political Legacy of Coal Mining

Deindustrialization has been one of the defining economic and political processes of the last 50 years. However, while existing literature systematically explores variations in the timing and intensity of deindustrialization, it largely remains silent on the potential long-term influence of differences in the preceding industrialization process. We argue that the timing, speed, and shape of the original industrialization are an important determinant of the political effects of deindustrialization. This argument builds on a rich tradition in economic history and political science, where scholars have long emphasized the lasting economic and political legacies of industrialization. Against this background, we develop a theory on the long-term effects of (de)industrialization on political alignment, proposing the guiding thesis that variations in the industrialization process lead to important differences in the political responses to deindustrialization. We test our theory using detailed historical and fine-grained geographical data from the Ruhr area. The region experienced notable variation in industrialization and deindustrialization, driven by the geography of coal deposits: the South developed earlier thanks to the easier accessibility of coal, while the North, with larger sites and more (migrant) workers, followed later, resulting in more disruptive deindustrialization.

This geographical variation offers a unique opportunity to explore differences in industrialization and deindustrialization within a largely homogenous institutional and cultural context. Empirically, we match the geolocation of nearly 1,000 historical mining shafts with polling station level electoral data in fourteen cities from the interwar period to the present. First, exploiting the varying depth of coal deposits for cross-sectional causal identification, we study if communities historically shaped by coal mining exhibit higher support for the radical right and lower voter turnout in recent federal elections. Second, we build a neighborhood-level panel and use staggered difference-in-differences to examine whether the gradual closure of coal mines in the second half of the twentieth century triggered electoral realignment.

Speaker 

Lukas Haffert, University of Geneva 

Discussant 

Jens Carstens, Sciences Po, CEE