This paper models the People’s Bank of China’s operating procedures in a two-stage vector autoregression model to search for a valid good policy indicator for Chinese monetary policy. The model disentangles endogenous components in changes in monetary policy that are driven either by demand for money or the liquidity management needs arising from foreign exchange purchases. There are four main findings. First, the PBC’s procedures appear to have changed over time, and hence no single indicator represents Chinese monetary policy well for the 2000-2013 time period. Second, its operating procedure is neither pure interest-rate targeting nor pure reserves targeting, but a mixture. Third, a set of indicators all contain information about the policy stance. It is hence preferred to use a composite measure to measure Chinese monetary policy. Finally, we construct a new composite indicator of the overall policy stance, consistent with our model. A comparison with several existing measurement approaches suggests that the composite indices, rather than individual indicators, perform better in measuring Chinese monetary policy.
Available for download here.