论文标题
学校建设的学校教育和劳动力市场后果：评论
Schooling and Labor Market Consequences of School Construction in Indonesia: Comment
论文作者
论文摘要
I Revisit Duflo（2001），该（2001年）利用了1970年代印度尼西亚的教育扩展，以估算上学的回报。我纠正数据错误，检查对原始文本中未讨论的两个规范选择的敏感性，调整潜在的偏差来源（治疗单位内的错误相关性，内源性加权，非平行趋势，弱仪器较弱），并随后在生活中后期受影响的队列。减少工资的影响估计是强大的，而多年来在学校度过的估计值并非如此，这使得前者很难将其归因于后者。
I revisit Duflo (2001), which exploits a 1970s schooling expansion in Indonesia to estimate the returns to schooling. I correct data errors, check sensitivity to two specification choices not discussed in the original text, adjust for potential sources of bias (error correlation within the geographic units of treatment, endogenous weighting, non-parallel trends, weak instruments), and follow up on affected cohorts later in life. The reduced-form impact estimate for wages is robust, while that for years spent in school is not, making the former hard to attribute to the latter.
