Revisit the Real Impact of China’s Send-down Movement on Rural Education

Abstract

Chen et al. (2020) claim a positive causal effect of the send-down movement on rural education during China’s Cultural Revolution. Their result hinges on a highly sensitive key variable: education attained by the treatment cohorts, which is found to be treated with upward bias and inconsistently coded. We show that when these errors are corrected, the claimed positive effect is substantially diminished or even turns negative. Using Chen et al.’s original data and model, our analysis reveals that the send-down movement actually increased rural illiteracy rate, which raises serious concerns about the robustness and policy implications of their findings.