In this paper we use a new data set describing households with and without twin children in China to quantify the trade-off between the quality and quantity of children using the incidence of twins that for the first time takes into account effects associated with the lower birthweight and closer-spacing of twins compared to singleton births. We show that examining the effects of twinning by birth order, net of the effects stemming from the birthweight deficit of twins, can provide upper and lower bounds on the trade-off between family size and average child quality. Our estimates indicate that, at least in one area of China, an extra child at parity one or at parity two, net of birthweight effects, significantly decreases the schooling progress, the expected college enrollment, grades in school and the assessed health of all children in the family. We also show that estimates of the effects of twinning at higher parities on the outcomes of older children in prior studies do not identify family size effects but are confounded by inter-child allocation effects because of the birthweight deficit of twins. Despite the evident significant trade-off between number of children and child quality in China, however, the findings suggest that the contribution of the one-child policy in China to the development of its human capital was modest.