Research

Recent advances in automation have raised concerns about job displacement and increased interest in social protection policies. However, public support for such measures is not uniformly distributed across cases of job loss. This study argues that gender norms, rather than economic vulnerability alone, explain support for social protection in response to automation-driven layoffs. Using a survey experiment in South Korea, we show that automation-driven job loss increases support for an ex-ante protective measure (e.g., Automation Tax) only when male workers are affected. This selective protection reflects the male-breadwinner model, which views male labor as more essential to household income and male job loss as more socially disruptive. The disparity in social policy preferences by the laid-off worker’s gender profile is pronounced among individuals who hold sexist attitudes. Our findings reveal how gendered beliefs about labor value shape social protection preferences, highlighting identity-based biases in responses to economic change.

The climate crisis is an urgent challenge, yet public support for fossil fuel taxation remains low. Existing research highlights how material interests, such as concerns about the regressive nature of consumption taxes like fuel taxes, contribute significantly to this limited support. This perspective aligns with the literature on the emerging ‘eco-social divide,’ which highlights tensions between environmental and equity concerns, yet potential synergies remain understudied. While social policies may ease resistance by protecting displaced workers, we know little about whether they actually do. We examine what drives public support for a fossil fuel tax designed to reduce carbon emissions but likely to displace workers, focusing on the mediating role of the welfare state. Drawing on original survey data from the ‘Eco-Social’ UniMi project (10,500 respondents across seven European countries), we find that income anxiety and dissatisfaction with national social policies significantly reduce support for fossil fuel taxation. Moreover, individuals facing high employment risk are less supportive of carbon taxation when dissatisfied with social policies, but their support increases as satisfaction with the welfare state improves. These findings suggest that confidence in the welfare state can offset economic concerns driving opposition to climate policy, thereby easing the ‘green backlash.’

Many of the richest social data come as open-ended, write-in responses. Occupational data are a prime example, yet the difficulty of coding such free-text entries has long limited research. We address this challenge by fine-tuning a large language model (LLM) to classify millions of write-in occupation fields into a standardized taxonomy. Treating classification as a text-generation task, we supplement standard occupational descriptions with a large corpus of alternative job titles, generating synthetic data for infrequent occupations. The resulting classifier matches or exceeds the accuracy of the US government’s benchmark tool, especially in the long tail of rare occupations. We use this method to construct the Occupational Partisanship in the United States (OPUS) dataset, which assigns partisan scores to nearly all occupations in the O*NET standard classification. Two findings illustrate its potential: (i) occupation predicts partisan identity as well as, or better than, county of residence; and (ii) majority-female occupations lean consistently Democratic. Our approach provides a low-cost, generalizable framework for classifying write-in data and a new empirical lens on the political structure of the American labor market.

Do citizens judge job displacement differently depending on whether it is driven by automation or immigration, and does institutional context shape these evaluations? This paper develops an institutional-conditioning framework arguing that coordinated industrial relations create shared legitimacy frames through which publics interpret economic adjustment. An institutional analysis of the UK and Sweden shows that Britain’s weak unions and voluntarist bargaining enabled employers to rely on low-cost migrant labor, whereas Sweden’s stronger labor coordination limited dependence on migration and promoted technology-oriented adjustment through retraining and social partnership. A survey experiment (n = 1,050) finds that Swedish respondents distinguish sharply between the two sources—viewing immigration-led displacement as less legitimate and more deserving of regulation than automation-led displacement—while British respondents show no such distinction. Exploratory analysis suggests individual union membership is associated with more favorable evaluations of automation-led displacement but not immigration-led displacement. Taken together, the findings suggest that institutions of industrial relations shape both how economies adjust to labor market pressures and how citizens assign legitimacy to economic transformation.