Abstract: The increase in employment polarization observed in several high-income economies has coincided with a reduction in inter-generational mobility. This paper argues that the disappearance of middling jobs can drive changes in mobility, notably by removing a stepping-stone towards high-paying occupations for those from less well-off family backgrounds. Using data for two British cohorts we examine how the occupational outcomes of children depend on both initial occupations and occupational upgrading during their careers. We find that transitions across occupations are key for mobility and that the effect of parental income on those transitions has become stronger over time. Moreover, the impact of parental income increased the most in the regions where the share of middling employment fell the most, suggesting that greater employment polarization may be one of the factors behind the observed decline in mobility.

Abstract: This paper estimates industry and occupation exposure to a comprehensive set of emerging digital technologies and assesses their impact on European regional employment. Using a novel, scalable methodology based on advanced natural language processing techniques (sentence transformers), we measure technological exposure using semantic similarity between patents and standardized international classifications. Using an instrumental variable shift-share approach, we find that higher regional exposure yields net employment gains. Explicitly accounting for complementarities between digital technologies, we estimate their individual effects and classify technologies as labor-saving or labor-augmenting, based on their impacts on aggregate employment. We identify distinct patterns of technological impacts across different skill groups, and we rationalize them within a task-based theoretical framework. Our findings highlight that focusing narrowly on specific technologies such as AI or robots, without accounting for complementarities across the broader digital technology landscape, can significantly understate the broader, positive effects of digital transformation on employment.

Abstract: This paper estimates the exposure of US occupations and industries to emerging digital technologies and their impact on US commuting zone (CZ) employment. Building upon the natural language processing approach introduced by Prytkova et al. (2024), we estimate the exposure of O*NET-SOC occupations and NAICS industries, thereby extending the open-access TechXposure database to the US context. Using this new data source, we apply a shift-share design to instrument the CZ exposure to emerging digital technologies and estimate their employment impact across CZs between 2012 and 2019. We find that digital technologies have an overall positive net impact on US employment. However, the impact varies among different worker demographics: while there is a noticeable decline in employment for core working-age (25-44) and non-college-educated workers in more exposed CZs, we observe employment increases for younger (16-24) and older (45-64) workers, as well as for those with a college education.

Abstract: Tuition fees are a critical source of revenue for universities, yet how student demand responds to changes in fees remains poorly understood. Using administrative data from one of the largest UK universities between 2019 and 2025, we estimate the price elasticity of demand for both undergraduate and postgraduate degrees. Our analysis distinguishes between the application and enrolment stages, accounts for persistence in demand across cohorts, and incorporates fee data from competitor institutions to estimate cross-price elasticities. We find that postgraduate students are substantially more price-sensitive than undergraduates, with estimated elasticities of -0.27 for applications and -0.13 for enrolments. Undergraduate demand is largely price-inelastic. Elasticities vary sharply across countries: applicants from emerging markets such as India, Indonesia, and Turkey display positive application elasticities, consistent with tuition functioning as a signal of quality, while students from Europe and the Americas exhibit conventional price sensitivity. Subject-level variation is more muted: demand for engineering and other STEM disciplines is effectively inelastic, consistent with high expected earnings, while other subjects display stronger negative elasticities. We also document strong persistence in demand across cohorts within countries, suggesting peer-driven information spillovers. Finally, we find limited responsiveness to competitors' tuition at the application stage but positive cross-price elasticity at enrolment, indicating substitution effects once offers are received. These results provide the most comprehensive and recent evidence on tuition responsiveness in UK higher education, highlighting how price sensitivity differs across stages, markets, and subjects.

Abstract: This paper explores how life events change values and social identity when both are endogenous, that is when individuals identify with a social group based on shared values. Life events may introduce new information that shifts a value central to their social identity, misaligning individuals with their current social group's values. Consequently, individuals may align with a new group, affecting previously unchanged values and creating spillover effects. Using cohort data, I find that life events, such as parenthood or sickness, significantly alter values and social identity. Overlooking the interdependence between values underestimates the extent to which life experiences affect individuals.

Abstract: This paper analyzes how the impacts of ICT, Software and Databases, and Robots on European regional labor markets (1995-2017) vary across technology life cycle phases. Motivated by theories predicting shifting skill biases between early adoption and maturity, we first identify major technological breakthroughs and delineate their life cycle phases (early versus maturity) based on investment growth patterns. Using a shift-share instrumental variable approach, we estimate phase-specific impacts of regional technology exposure on employment and wages. While confirming that effects differ significantly across phases, we find only partial support for standard skill-bias predictions during early adoption. Our results highlight the importance of analyzing dynamics within specific technology life cycles to understand the heterogeneous short-term labor market adjustments often obscured in aggregated long-run analyses.

Abstract: This paper examines the labor market adjustments to four automation technologies, that is robots, communication technology, information technology, and software and database, in 227 regions across 22 European countries from 1995 to 2017. By constructing a measure of technology penetration, we estimate changes in regional employment and wages affected by automation technologies along with the reallocation of workers between sectors. We find that labor market adjustments to automation technologies differ according to i) the technology involved, ii) the sector of penetration, iii) the sectoral composition of the region, and iv) the region's technological capabilities. These adjustments are driven largely by the reallocation of low-paid workers across sectors.

Abstract: This paper argues that the decline in the labor income share since the 1970s is a consequence of the emergence of a relatively larger generation, the Baby Boomers, compared to other cohorts. I develop an OLG model in which public policy is endogenously shaped by the population's age structure through voting. When young, Baby Boomers vote to increase unemployment benefits to mitigate unemployment risk, raising the value of their outside option in wage bargaining and enabling them to negotiate higher wages. Firms respond by substituting labor with capital to limit workers' rent appropriation, causing a decline in the labor share. Once the Boomers retire, this effect reverses but is offset by capital accumulation driven by the Boomers' high savings rates, fueled by higher wages, further reducing the labor share. Calibrated for France and the United States, the model's simulations replicate the observed decline in labor share and labor market dynamics. The model predicts that, from 2020 onward, approximately one percentage point of labor income share will shift to capital income every 20 years, on average, through the end of the 21st century.