Does the timing of AI exposure matter? Generational composition, human capital, and productivity

This study examines whether generational differences in exposure to digital technologies and artificial intelligence are associated with productivity performance. Using Penn World Table, World Development Indicators, Worldwide Governance Indicators, and United Nations World Population Prospects, the study builds a balanced panel of 72 countries over 2000-2023. The empirical strategy decomposes the working-age population into Gen X, Millennials, and Gen Z and estimates fixed-effects models linking generational shares, human capital, and the post-2016 AI diffusion period to labor productivity. Generational variables are working-age population shares rather than employment shares; results should therefore be interpreted as demographic-potential associations rather than direct evidence on individual technology use. The findings show that generational composition alone is not positively associated with productivity. However, the triple interactions between generational shares, human capital, and the post-AI period are positive and statistically significant. The strongest association is found for Millennials, suggesting that this cohort combines digital adaptability with accumulated labor-market experience. Results are robust to using total factor productivity, controlling for institutional quality, excluding internet users, replacing year fixed effects with a linear time trend, fully saturating the interaction specification, and alternative AI-timing thresholds. The findings suggest that AI-era productivity gains depend on economies’ demographic and human-capital capacity.