The effect of cash transfers on maternal health seeking: Evidence from Ecuador
研究厄瓜多尔有条件现金转移项目对孕产妇就医行为的影响，发现项目使机构分娩增加10%、私人机构分娩增加20%，但未显著提升产前保健利用率，且出生体重略有下降。
Governments often aim to increase health service uptake using financial incentives. In the case of maternal healthcare, where the quality gradient between informal and formal care is quite steep, these incentives have the potential to improve birth outcomes significantly. However, whether these programs encourage care ultimately depends on the elasticity of demand for care. We examine the effects of one such program in Ecuador that aimed to incentivize care through conditional cash transfers for antenatal care and lump-sum grants to facilitate delivery. We employ a difference-in-discontinuities approach, exploiting a discontinuity created by the program’s targeting method to identify its causal impact. Our findings reveal a 10 percent increase in institutional deliveries, a 3.5 percent increase in trained delivery care, and a 20 percent increase in births at private facilities. We do not see evidence of a complementary increase in the uptake of antenatal care services, which was already high before the program. Our results also suggest a marginal decrease in birth weights as a result of the program, which we attribute to selection into measurement. We show that our difference-in-discontinuity approach improves precision compared to traditional regression discontinuity designs, but does so at the cost of increased specification error. 1 1 We thank John Hoddinott, John Cawley, Chris Barrett, Mahesh Karra, and two anonymous reviewers, whose careful feedback has greatly improved the quality of this manuscript. We thank Nancy Flores of the Ecuadorian Ministry of Health for her valuable assistance with program details. All data used in the project was de-identified and is publicly available at the URLs in the bibliography. This research did not receive any specific grant from funding agencies in any sector, and the authors have no conflicts of interest to disclose. • A large-scale transfer program in Ecuador, which combined conditional and unconditional but labeled transfers, led to moderate improvements in institutional child delivery, particularly among private sector care. • In the context of a shift of birth location, program effects on children’s anthropometrics are likely to be biased by selection into measurement. • Results suggest that difference-in-discontinuity designs display greater sensitivity to bandwidth selection than more traditional regression discontinuity estimates.