Journal of Development Economics · 2024 · Lenin Balza、Camilo De Los Ríos、Raúl Jimenez Mori、Osmel Manzano
中文摘要
本文探讨石油开采对哥伦比亚这一资源丰富的发展中国家本地人力资本积累的影响。我们基于有关石油开采和教育的详细时空数据提供证据,以油井数量衡量学校层面的处理强度。为获得因果估计,我们采用工具变量方法,利用国际油价的外生性和当地石油禀赋的代理变量。结果表明,尽管石油开采不影响高等教育入学,但它会推迟入学决策,因此可能对哥伦比亚的人力资本产生负面影响。此外,石油开采还促使人才配置转向科学、技术、工程和数学(STEM)以及专业学位之外的领域。我们的结果表明,石油开采对中等教育质量和高等教育学业完成均无影响。这些结果在多种模型设定调整下依然稳健,我们强调当地市场和溢出效应是主要的传导渠道。具体而言,我们发现,石油产量提高会增加正规部门工资,但未发现产油地区存在与入读高等教育相关的溢价的证据,同时还存在对其他经济部门的正向溢出效应。
• 我们表明,哥伦比亚的石油开采推迟了高等教育入学,并使学生的偏好转向STEM和专业学位之外的领域。
• 我们进一步表明,在广延边际上,石油开采对高等教育入学和学业完成均无影响。
• 通过聚焦学校层面的效应,我们还提供了提示性证据,表明劳动力市场可能在驱动这一决策。
Abstract
This paper explores the local-level impacts of oil exploitation on human capital accumulation in Colombia, a resource-rich developing country. We provide evidence based on detailed spatial and temporal data on oil exploitation and education, using the number of wells to measure treatment intensity at the school level. To derive causal estimates, we rely on an instrumental variable approach that exploits the exogeneity of international oil prices and a proxy of local oil endowments. The results indicate that although oil exploitation does not impact enrollment in higher education, it may negatively affect human capital in Colombia since it generates a delay in the decision to enroll. Furthermore, it incentivizes talent allocation away from STEM and professional degrees. Our results suggest no effect on either the quality of secondary education or tertiary education completion. These results are robust to a number of specification changes, and we stress the role of local markets and spillovers as the main transmission channel. In particular, we find that higher oil production increases formal wages, without finding evidence of a premium to tertiary education enrollment in oil-producing areas, and there is a positive spillover to other economic sectors. • We show that oil exploitation in Colombia delays tertiary education enrollment and shifts student preferences away from STEM and professional degrees • We further show that there are no effects on enrollment and completion on the extensive margin. • We also provide suggestive evidence that labor markets might be driving this decision by focusing on school level effects.