Incentives are essential to promote labor productivity. We implemented a two-stage field experiment to measure effects of career and wage incentives on productivity through self-selection and causal effect channels. First, workers were hired with either career or wage incentives. After employment, a random half of workers with career incentives received wage incentives and a random half of workers with wage incentives received career incentives. We find that career incentives attract higher-performing workers than wage incentives but do not increase productivity for existing workers. Instead, wage incentives increase productivity for existing workers. Observable characteristics are limited in explaining the selection effect.