The Myth of Positive Excess Return of ETFs in China: Decomposition and Attribution of ETF Tracking Errors

Abstract

We adopt 72 passively managed equity exchange-traded funds (ETFs) from the Chinese A-share market to examine the size and time-varying characteristics of their tracking errors, identify their determinants, and gain insight into the persistent and positive excess returns of China’s equity ETFs. We observe larger deviations in ETF closing price returns than in net asset value (NAV) returns from the index return. Furthermore, the tracking errors of the delisted ETFs are significantly larger than those of the surviving ones. Our regressions indicate that the common factors affecting NAV and closing price tracking errors are fund expenses as well as liquidity and volatility of the index. The cash drag and leverage ratio correlate with NAV tracking errors, whereas the liquidity of ETF and fund investment in futures and security lending affect closing price tracking errors. Additionally, many determinants asymmetrically influence the tracking errors in bull and bear markets. Moreover, it is more difficult for ETFs to track benchmarks in bearish markets when the tracking capabilities across ETFs are more dispersed. We further decompose the NAV tracking errors into three parts based on the capital asset pricing model (CAPM). The results reveal that the initial returns of initial public offerings (IPOs) are significant for ETFs’ positive excess returns, suggesting inefficiency in the A-share market.

Keywords: Exchange-traded funds, tracking errors, CAPM, excess return

JEL Classification: G11, G12, G23