Mandal Bidisha, Batina Raymond G, Chen Wen
School of Economic Sciences, Washington State University, Pullman, WA, USA.
Health Econ. 2018 May;27(5):877-886. doi: 10.1002/hec.3636. Epub 2018 Jan 8.
We use system-generalized method-of-moments to estimate the effect of gender-specific human capital on economic growth in a cross-country panel of 127 countries between 1975 and 2010. There are several benefits of using this methodology. First, a dynamic lagged dependent econometric model is suitable to address persistence in per capita output. Second, the generalized method-of-moments estimator uses dynamic properties of the data to generate appropriate instrumental variables to address joint endogeneity of the explanatory variables. Third, we allow the measurement error to include unobserved country-specific effect and random noise. We include two gender-disaggregated measures of human capital-education and health. We find that gender gap in health plays a critical role in explaining economic growth in developing countries. Our results provide aggregate evidence that returns to investments in health systematically differ across gender and between low-income and high-income countries.
我们运用系统广义矩估计法，在1975年至2010年期间127个国家的跨国面板数据中，估计特定性别的人力资本对经济增长的影响。使用这种方法有几个好处。首先，动态滞后相依计量模型适合处理人均产出的持续性。其次，广义矩估计法利用数据的动态特性生成适当的工具变量，以解决解释变量的联合内生性问题。第三，我们允许测量误差包含未观察到的国家特定效应和随机噪声。我们纳入了两种按性别分类的人力资本衡量指标——教育和健康。我们发现，健康方面的性别差距在解释发展中国家的经济增长中起着关键作用。我们的结果提供了总体证据，表明健康投资的回报在性别之间以及低收入和高收入国家之间存在系统性差异。
