The impact of the one child policy on China's infant mortality from 1970-1989: A quasi-experimental study

Background China's One Child Policy (OCP), enforced in 1980, was the largest fertility control policy in human history, and its effects are still widely debated. Previous analyses of the OCP have focused on population reduction and sex ratios, among others. This study evaluates the impact of the OCP on infant mortality in China using quasi-experimental methods. Methods Country-level panel data for China and 106 other countries were extracted from the World Bank data repository for the years 1970-1989. The primary outcome was infant mortality rate (IMR) defined as deaths among children <1 year of age per 1000 live births. The impact of the OCP was estimated using uncontrolled and controlled interrupted time series (ITS) designs that control for within group time varying characteristics. Models were further adjusted for a comprehensive set of covariates and corrected for first-order autocorrelation. Countries included in the control groups were identified using the method of synthetic controls and propensity score matching (PSM). Valid control groups reduced threats from selection bias and unobserved time invariant confounders. Results Adjusting for GDP per capita, GDP growth, women's education, and population density, IMR decreased on average by 2.5 deaths per 1,000 annually before the OCP was enforced in China. The pre-OCP trends in IMR moved in tandem across China, synthetic and PSM controls, validating the parallel trends assumption. After the OCP, the trend in IMR in China increased by 2.24 (95% CI 1.39-3.09) per year, compared to the IMR trend in the pre-OCP period, comparing differences in China to synthetic controls. Results were similar and confirmed across PSM (1.96, 0.56-3.37) and uncontrolled ITS (2.14, 1.39-2.89) models. No significant immediate level change in IMR was observed in the controlled ITS models. Conclusions The reduction in IMR slowed down significantly after the OCP was enforced in China. The slowdown is likely due to a relative increase in the ratio of the IMR of females to males suggesting that practices such as female infanticide, abandonment, and neglect, stemming from a strong son preference were primary contributors. Coercive policies to reduce fertility can have unintended consequences.