Periodic Reporting for period 2 - DESPO (De-industrializing Societies and the Political Consequences)
Berichtszeitraum: 2021-08-01 bis 2023-01-31
The aim of this project is to reveal how a person’s individual, family, and local community experiences of manufacturing decline transform the way they participate in politics and their political attitudes over the course of their life. More specifically, the project answers the following questions: How and why has long-term manufacturing decline affected the political participation, party loyalty, and political attitudes of: a) industrial workers who have been displaced or experienced job insecurity, b) spouses and children of industrial workers who have been displaced or faced job insecurity and c) residents of local areas affected by chronic manufacturing decline?
While much of the research has considered the political impact of deindustrialization as occurring though occupational exposure (for instance job loss, job displacement or job insecurity), this project considers the political impact of deindustrialization through diffuse exposure. Diffuse exposure can occur through family and kin networks, communities and neighborhoods, and also, more generally, growing up in an ‘era’ characterized by deindustrialization. DESPO will focus on the long-term consequences by studying a rich time frame which spans five decades of manufacturing decline and its political aftermath (1965-2015) in the United States, United Kingdom and Germany. It will examine an exhaustive series of political attitudes and behaviours: a) if and how people vote; b) what people believe and think about their political system; and c) how strongly people are attached to and identify with political parties.
(see Figure A)
DESPO has three main objectives:
1) to explain how deindustrialization causes an individual, household, class, and local contextual basis for political change by exploiting its local, industrial, and time variation
2) to establish a conceptual and empirical framework for how the uneven experience of deindustrialization across social groups constitutes a basis for political differentiation
3) to construct a unique database of multi-dimensional measures of deindustrialization for small geographic units by repurposing large scale administrative data of business registries for novel use
While much of the research has considered the political impact of deindustrialization as occurring though occupational exposure (for instance job loss, job displacement or job insecurity), this project considers the political impact of deindustrialization through diffuse exposure. Diffuse exposure can occur through family and kin networks, communities and neighborhoods, and also, more generally, growing up in an ‘era’ characterized by deindustrialization. DESPO will focus on the long-term consequences by studying a rich time frame which spans five decades of manufacturing decline and its political aftermath (1965-2015) in the United States, United Kingdom and Germany. It will examine an exhaustive series of political attitudes and behaviours: a) if and how people vote; b) what people believe and think about their political system; and c) how strongly people are attached to and identify with political parties.
(see Figure A)
DESPO has three main objectives:
1) to explain how deindustrialization causes an individual, household, class, and local contextual basis for political change by exploiting its local, industrial, and time variation
2) to establish a conceptual and empirical framework for how the uneven experience of deindustrialization across social groups constitutes a basis for political differentiation
3) to construct a unique database of multi-dimensional measures of deindustrialization for small geographic units by repurposing large scale administrative data of business registries for novel use
Sub-project 1: Measuring Local Deindustrialization
Our work on measuring deindustrialization began with a new conceptualization and operationalization of place-based deindustrialization that considers decline, loss and dispersion. To measure this, we construct a database of deindustrialization at local level in the United States between 1970 and 2018. Due to some inconsistencies and missing data, we designed a validation process to ensure the quality of the administrative data for each county (n = 3109). After harmonizing them on a 2010 county basis using area and population-based crosswalk, we developed a decision tree model to ensure the reliability of the data for each county (employment, establishment, and wages) between 1975 and 2020. Afterwards, we validated our time series with other available data.
Subject 2: Family Exposure to Deindustrialization
We use data from the Youth-Parent Socialization Panel, a longitudinal survey across three biological generations (1965– 1997) in the United States. The analysis statistically estimates whether the respondent has voted in a national election in that year. In addition to the main predictor of interest, working-class occupation, the model includes parent and respondent controls as well as including year dummies. Then, we estimate a regression to model the relationship between a working-class family background and informal political participation in the second and third generation.
Subproject 3: Community Exposure to Deindustrialization
We link individual panel data from Understanding Society from 2009 to 2017 with the English Index of Multiple Deprivation at neighborhood level. We use data from Understanding Society, a yearly panel study of more than 40,000 households in the UK which began in 2009, and the English Index of Multiple Deprivation. Our merged data are suitable because they allow us to: track individual exposure to changing community deprivation across 10 years (2010 to 2019); define community at a granular level; accurately measure deprivation as well as various forms of civic involvement.
We also have two other ongoing studies in-progress. One study examines the role of place-based manufacturing decline in electoral outcomes in US presidential elections (1980-2016), especially the 2016 election. The other study looks at the relationship between regional manufacturing decline and satisfaction with Democracy in Europe (1980-2018).
Our work on measuring deindustrialization began with a new conceptualization and operationalization of place-based deindustrialization that considers decline, loss and dispersion. To measure this, we construct a database of deindustrialization at local level in the United States between 1970 and 2018. Due to some inconsistencies and missing data, we designed a validation process to ensure the quality of the administrative data for each county (n = 3109). After harmonizing them on a 2010 county basis using area and population-based crosswalk, we developed a decision tree model to ensure the reliability of the data for each county (employment, establishment, and wages) between 1975 and 2020. Afterwards, we validated our time series with other available data.
Subject 2: Family Exposure to Deindustrialization
We use data from the Youth-Parent Socialization Panel, a longitudinal survey across three biological generations (1965– 1997) in the United States. The analysis statistically estimates whether the respondent has voted in a national election in that year. In addition to the main predictor of interest, working-class occupation, the model includes parent and respondent controls as well as including year dummies. Then, we estimate a regression to model the relationship between a working-class family background and informal political participation in the second and third generation.
Subproject 3: Community Exposure to Deindustrialization
We link individual panel data from Understanding Society from 2009 to 2017 with the English Index of Multiple Deprivation at neighborhood level. We use data from Understanding Society, a yearly panel study of more than 40,000 households in the UK which began in 2009, and the English Index of Multiple Deprivation. Our merged data are suitable because they allow us to: track individual exposure to changing community deprivation across 10 years (2010 to 2019); define community at a granular level; accurately measure deprivation as well as various forms of civic involvement.
We also have two other ongoing studies in-progress. One study examines the role of place-based manufacturing decline in electoral outcomes in US presidential elections (1980-2016), especially the 2016 election. The other study looks at the relationship between regional manufacturing decline and satisfaction with Democracy in Europe (1980-2018).
Main Results Achieved
(see Figure B and C)
Community Exposure:
Our results indicate that the social isolation that commonly occurs in deprived neighborhoods is a strong mechanism in reducing participation. Social cohering plays a substantively large positive role on civic and political membership, suggesting that the more people feel attached to their community the more likely they are to participate in local associations but also actively engage with national ones such as political parties and trade unions. The lack of social cohesion functions as a mechanism whereby residents feel less closely attached to the neighborhood and each other and we speculate that this could alter the prism through which residents perceive the benefits of participation. As participation requires an investment of time and energy, individuals may be less likely to invest these finite resources if they feel less connected to those around them.
We also uncover evidence to support a normative mechanism, whereby deprivation has a cultural influence on social life. Our study finds that neighborhood deprivation depresses norms of civic obligation which, in turn, lowers propensity for engagement. On the flip side, adhering to norms regarding the obligation to participate civically has an important positive role on all forms of membership, with a strikingly large effect on political membership. Weaker norms about civic participation being the "right" thing to do, alters the incentive structure for participating. As there are less perceived social benefits from this behavior, people are less likely to see a value in such activities.
Family Exposure:
In a study published recently, we have also found that occupational structures of industrial society are still politically relevant and that inequalities in political participation are a legacy in the biological descendants of working-class families of the 1960s. By conceiving of working class not only as those who hold working-class jobs, we are able to study the family members whose life trajectories have also been affected by manufacturing decline. The findings do not dispute the unequal political participation of individuals with working-class occupations but compliment this with explaining how these inequalities are passed down through families. The project finds evidence of the social reproduction of social class in political participation. Such a process maintains the political relevance of the working class in post-industrial society as features of industrial societies remain embedded in social structures such as the family.
(see Figure B and C)
Community Exposure:
Our results indicate that the social isolation that commonly occurs in deprived neighborhoods is a strong mechanism in reducing participation. Social cohering plays a substantively large positive role on civic and political membership, suggesting that the more people feel attached to their community the more likely they are to participate in local associations but also actively engage with national ones such as political parties and trade unions. The lack of social cohesion functions as a mechanism whereby residents feel less closely attached to the neighborhood and each other and we speculate that this could alter the prism through which residents perceive the benefits of participation. As participation requires an investment of time and energy, individuals may be less likely to invest these finite resources if they feel less connected to those around them.
We also uncover evidence to support a normative mechanism, whereby deprivation has a cultural influence on social life. Our study finds that neighborhood deprivation depresses norms of civic obligation which, in turn, lowers propensity for engagement. On the flip side, adhering to norms regarding the obligation to participate civically has an important positive role on all forms of membership, with a strikingly large effect on political membership. Weaker norms about civic participation being the "right" thing to do, alters the incentive structure for participating. As there are less perceived social benefits from this behavior, people are less likely to see a value in such activities.
Family Exposure:
In a study published recently, we have also found that occupational structures of industrial society are still politically relevant and that inequalities in political participation are a legacy in the biological descendants of working-class families of the 1960s. By conceiving of working class not only as those who hold working-class jobs, we are able to study the family members whose life trajectories have also been affected by manufacturing decline. The findings do not dispute the unequal political participation of individuals with working-class occupations but compliment this with explaining how these inequalities are passed down through families. The project finds evidence of the social reproduction of social class in political participation. Such a process maintains the political relevance of the working class in post-industrial society as features of industrial societies remain embedded in social structures such as the family.