Automation, Workforce Age Structure, and Firm Productivity: Evidence from Austrian Linked Employer-Employee Data

Abstract

The aim of this study is to estimate the age–productivity profile of Austrian firms using a linked employer–employee dataset for the years 2013–2022. The OLS and FE estimates indicate a highly significant relationship between workforce age structure and labour productivity. Across both estimation methods, we find an inverted U-shaped age–productivity profile. We also account for capital intensity and the share of automation-related assets (ADRA). The estimation results show that firms with greater capital intensity and higher levels of automation consistently exhibit higher productivity across the distribution. In addition, the marginal effect of the share of ADRA-related capital is greater than that of the agerelated variables. These findings have important implications for both firm strategy and public policy, highlighting the role of technology diffusion, education, and potentially organisational change in sustaining productivity in ageing societies. The empirical strategy is complemented by panel data methods and robustness checks to account for persistence, unobserved heterogeneity, and potential reverse causality.