How to determine China’s true fertility?

Daniel M. Goodkind, U.S. Census Bureau

This paper demonstrates that true levels of fertility in China are best determined by focusing on its annual surveys of population change. A recent cluster of fertility estimates based on the 2000 census are all biased downwards due to exceptional underreporting of births and young children. China’s National Bureau of Statistics (NBS) typically adjusts raw fertility data for underreporting, and our review shows that the annual surveys evince far less underreporting than the 2000 census (or the 1995 bi-census). When we confine our analysis to the annual surveys and correct them for “long-form”-type biases, official NBS data imply that less than 10 percent of births went unreported from 1991-2003, a level we find reasonable, perhaps conservative. Further, we specify how the emerging alternative hypothesis that the NBS over-inflates fertility contradicts over two decades of findings regarding the negative effect of China’s family planning program on the completeness of birth reporting.