China has undergone significant deviations from the command economy established in 1949, achieving unprecedented success in sustaining high growth. Consequently, numerous scholars have devoted their attention to studying China's economic development. In this study, we contribute to this literature using a new technique to examine China's economic development. This technique utilizes nonparametric directional distance function in the framework of neoclassical growth theory. It avoids the assumptions imposed on parametric production functions that are often short of practical justifications. Our analysis shows that capital and energy inputs account for 75.7% and 20.8% of China's GDP growth during 1985–2020. These figures characterize a capital-driven growth model that is unlikely to be sustainable by conventional wisdom. However, our empirical evidence, obtained through an interactive-fixed effects model, indicates that it is not unreasonable to say that growth in China is government-led rather than capital-driven; the former could be sustained by desirable reforms under strong national leadership while the latter is not.