Abstract

We examine time-series characteristics of China’s capital flows during 1998-2014. More specifically, we employ Kalman filtering state-space models to gauge relative importance of permanent and transitory components in China’s overall FDI, equity, bond, other investment, and bank credit flows. Our results show that only in the case of FDI are both gross inflow and net flow dominated by a permanent stochastic level, suggesting that this source of capital is largely permanent. Incorporating covariates into the state-space models, we find that larger difference between RMB onshore and offshore interest rates encourages capital inflows that are dominated by a transitory component. Greater global risk perception, proxied by S&P 500’s volatility index, on the other hand, discourages them. These covariates imply that capital control may not be effective in stemming volatile and speculative flows. Our results on bilateral capital flows between China and US also suggest that these flows are less persistent and more volatile during 1998-2014 than previously found based on 1988-1997 data. Our results bear important policy implications as China engages in further reforms in its domestic financial system and greater integration with the world financial system.