School of Economics, University of Adelaide, Adelaide, South Australia, Australia.
Health Econ. 2022 Feb;31(2):399-416. doi: 10.1002/hec.4458. Epub 2021 Nov 24.
Using an exogenous measure of natural disasters based on physical intensities, I examine their impact on child mortality. I find that children, mostly, in poor countries are affected - a one standard deviation increase in the natural disaster index leads to an additional 6.77 deaths per 1000 live births in the disaster year. The impact is long lasting, extending beyond the disaster year. I provide suggestive evidence of the potential mechanisms driving these effects, namely, lower GDP and vaccination rates among children, along with an increased maternal mortality and disease incidence in low-income countries due to disasters. Using an endogenous measure of disasters based on damage records from insurance data leads to severe underestimation of the disaster impact. The results are robust to the use of mortality rates from multiple sources, different functional forms, and an extensive dynamic panel specification as well as various other tests.
利用基于物理强度的自然灾害外生指标,我考察了其对儿童死亡率的影响。我发现,主要是贫穷国家的儿童受到影响——自然灾害指数每增加一个标准差,受灾年份每 1000 例活产中就会额外增加 6.77 例死亡。这种影响具有持久性,超出了受灾年份。我提供了一些有说服力的证据,表明可能的机制推动了这些影响,即由于灾害,低收入国家儿童的国内生产总值和疫苗接种率下降,以及产妇死亡率和疾病发病率上升。利用基于保险数据中损失记录的自然灾害内生指标会严重低估灾害的影响。使用来自多个来源、不同函数形式的死亡率数据以及广泛的动态面板规范以及各种其他检验,结果都是稳健的。