Wage regressions by language groups.

At least half of the world population is bilingual, but lifetime financial gains from early exposure to multiple languages are largely unknown. In this study, we analyze individual earnings of bilinguals in the US using 15 years of Census data and an augmented wage model, accounting for cognitive, manual, and interpersonal skills derived from O*NET job task descriptors via sparse principal component method. Using unconditional quantile regression, we find evidence that language skills mostly benefit individuals at the lower end of the earnings distribution. While our analysis does not establish causality, it underscores the potential of early language acquisition to mitigate income inequality by enhancing the employment prospects of low-income individuals. We also highlight the favorable cost-benefit ratio of language acquisition in childhood, when learners face no monetary opportunity costs and can achieve greater levels of fluency.