The Role of ETFs in Asset Pricing, Mutual Fund Performance, and Market Prediction

Abstract

This thesis investigates the various roles that the information provided by Exchange Traded Funds (ETFs) could play in asset pricing and market prediction. The empirical analysis contains three parts: The first part extracts information from the US ETFs market and constructs explanatory returns to price the Fama-French portfolios. It aims to provide a parsimonious model (the ETF-factor model) that is able to compete with the five-factor model of Fama and French (2015) and the q-factor model of Hou, Xue, and Zhang (2015). The second part applies the ETF-factor model, along with other conventional pricing models, to measure US equity fund performance. In addition, it attempts to develop relative pricing models as passive benchmarks for measuring US fixed-income fund performance by using information from bond ETFs. The purpose of the third part is to develop a new measure of Chinese investor behaviour that has predictive power for the Chinese market by using the information provided by respective ETFs.
The results suggest that ETFs deserve more attention in academic research. In line with conventional financial theory, ETFs’ market dramatically increases the investment universe and securitizes illiquid assets. It comes as no surprise that the risk factors developed from ETFs have explanatory power for a cross-section of stock returns. In addition, a proxy for the bond market can be developed from bond ETFs. This avoids the subjective selection of the bond index as a passive benchmark and can provide a unique pricing model for bonds. Furthermore, research on ETFs contributes to the behavioural finance literature. Investor sentiment is a very important concept in behavioural finance. This thesis finds evidence that the investor behaviour that uses information from ETFs explains and predicts the Chinese market. In addition, it could lead to a profitable high-frequency trading strategy in actual trading.
Overall, this thesis researches ETFs from a new perspective. It does not view the ETFs as an investment vehicle but consider ETFs as a type of fundamental asset in the economy. The findings of this thesis contribute to the literature of asset pricing, behavioral finance, and market prediction, and identifies new areas for future research.