On the dynamics of tracking indices by exchange traded funds in the presence of high volatility

Purpose – The purpose is of this paper is to investigate whether the tracking ability of exchange traded funds (ETFs) is lower in highly volatile periods, and to shed more light on the factors behind the tracking error. Design/methodology/approach – The authors apply the Error Correction Model that incorporates a short‐run adjustment mechanism on domestic US ETFs that follow industrial indices. Findings – It was found that tracking errors attained pronounced levels during 2008 compared to 2006 and 2007, mainly in ETFs from the real estate and banking and finance sectors. In addition, tracking error is positively correlated with the daily volatility of the ETF, while trading volume has a limited effect on reducing tracking errors. Practical implications – The paper sheds more light on the relationship between securities and their fundamentals, and contributes to the literature on the information transmission mechanism for dually‐listed securities. Originality/value – The paper uses the co‐integration tests and the error correction model (ECM) to test the long‐run relationship between returns on domestic exchange trade funds (ETFs) and the returns on the underlying indices. In particular, the ECM is used for ETFs for the first time.