Do bank-affiliated funds perform better than the others: the higher moment approach

ABSTRACT

In this study, we examine the difference in mutual fund performance between the bank-related and the non-bank-related mutual funds in emerging markets. We further improve the empirical testing model to match the environment of the high-volatility and high-reward market – the emerging market. Specifically, we introduce co-skewness as an additional important risk factor in this study. Therefore, our model specification matches the non-normality of return distribution in the market. Furthermore, according to the information advantage hypothesis, we provide evidence of the superior market timing ability of the high-performance bank-related fund.

KEYWORDS:

• Mutual fund performance
• timing ability
• higher moment
• bank-affiliated fund

Acknowledgments

We would like to thank the editor, and the reviewers of Asian-Pacific Journal of Accounting & Economics for the helpful and insightful comments to enhance the contribution of our study.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. We comply the data from the international investment fund association (IIFA) and Investment company institute (ICI). The worldwide mutual fund manages 11,579.90 billion USD of asset and its expands to 49,284.28 billion USD at the end of 2017.

2. See detail in Mehran and Stulz (2007), Massa and Rehman (2008), Hao and Yan (2012) and Berzins, Liu, & Trzcinka.

3. Sources: Morningstar Direct database, as of December 2018.

4. The error term was obtained from ${\epsilon }_{t+1}=r_{t+1}-a_{t+1}-{\beta }_{t+1}\left(r_{m,t+1}\right)$. See Harvey and Siddique (2000).

5. In order to confirm that our results were not driven by the co-skewness risk factor formation procedure, in unreported tables, we formed $S^{-}$ from 5%, 10%, 15%, 25%, and 30% of the most negative co-skewness stocks, respectively. The results remained unchanged and are available upon request.

6. In the unreported table, we applied the Jarque-Bera test technique to do the normality test. We found that 177 of 253 fund returns departed from the normality distribution. Indeed, 123 bank-related funds and 54 nonbank-related funds are departed from the normality distribution.