Skill-Based Labor Market Polarization in the Age of AI: A Comparative Analysis of India and the United States (2501.15809v2)

Abstract: This paper examines labor market polarization through a comparative analysis of skill-based employment and wage distributions in India and the United States during 2018-2023, with particular attention to differential automation risks and AI preparedness. Using detailed occupation-level data, automation risk metrics, and a series of statistical tests including wage premium analysis, employment share tests, and wage-employment regressions, we document significant structural differences in labor markets between developing and developed economies. Our analysis yields four key findings. First, we find statistically significant differences in employment distribution patterns, with India showing disproportionate concentration in low-skill employment compared to the US, particularly in occupations with high automation risk. Second, regression analysis reveals that wage premiums differ systematically between the two countries, with significantly larger skill-based wage gaps in India. Third, we find robust evidence of a negative relationship between employment size and wages, suggesting stronger labor supply effects in developing economies. Fourth, analysis of occupation-specific automation risk reveals that developing economies face a "double vulnerability" - concentration of employment in both low-skill occupations and jobs with higher automation potential, complicated by lower AI preparedness scores. These findings provide novel empirical evidence on how development stages influence labor market polarization patterns and carry important implications for skill development and technology adoption policies in developing economies. Our results suggest that traditional approaches to labor market development may need significant modification to account for the differential impacts of AI across development stages.