Stock returns and mutual fund flows in the korean financial markets: a system approach

ABSTRACT

This paper investigates dynamic and causal relations between stock returns and mutual fund flows in Korea using a system method that utilizes information from the stock, bond, and money markets. For this purpose, we employ the Dynamic Seemingly Unrelated Regression, the Seemingly Unrelated Regression Error Correction Model, and two causality tests in a system method to account for cross-equation correlations among markets that have a close relationship with one another. Furthermore, we use the information in the variance-covariance matrix of residual to improve the efficiency of the statistical estimates. The empirical evidence from the system method indicates that fund flows do not respond to eliminate deviations from long-run equilibrium, and stock prices cause net fund flows in the Korean market, implying that investors move their money to the securities that yield higher returns to rebalance their investment portfolios in the short-run. Thus, our findings do not support the popular notion of mutual fund flows as the driving force behind rallies in the Korean financial markets.

KEYWORDS:

• Stock returns
• mutual fund flows
• causality
• cointegration
• DSUR
• SURECM

JEL CLASSIFICATION:

• G11
• C30

Acknowledgements

This paper has been greatly improved by the constructive comments and suggestions of Mark P. Taylor (editor) and an anonymous referee. We are also grateful for helpful comments from Moon Jung Choi, Woon Gyu Choi, Jong Ku Kang, Geun-Young Kim, Jaerang Lee, Jin-Su Park, and Joon Myung Woo. We would also like to thank Jin Woong Lee for his outstanding research support. The views expressed herein are those of the authors and do not necessarily reflect the official views of the Bank of Korea.

Data availability statement

The data described in this article are openly available in the Open Science Framework at DOI:10.17605/OSF.IO/TPA6U.

Disclosure statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

¹ See Warther (1995), Boyer and Zheng (2004), Baker and Wurgler (2007), and Qureshi et al. (2019), among others, for details.

² See Qureshi et al. (2019) for details about dynamics of mutual funds and stock markets in Asian economies.

³ See Warther (1995) and Baker and Wurgler (2007) for details.

⁴ See KFIA report (KFIA 2014) for details.

⁵ See World Bank report for details.

⁶ See EFAMA report (EFAMA 2014) for details.

⁷ Since monthly aggregate fund flows are available in both datasets, we choose monthly as a time frequency for comparing the results between the two data sources and they are seasonally unadjusted.

⁸ For details, see Jotikasthira, Lundblad, and Ramadorai (2012) and Fratzscher (2012).

⁹ See Jotikasthira, Lundblad, and Ramadorai (2012) and Fratzscher (2012) for details.

¹⁰ According to the Global Financial Stability Report by IMF, the EPFR data include bond funds that hold bonds with a wide range of maturities. Therefore, the effect of short-term rates on bond flows is obscured by possible differing effects on long-term bonds. The converse appears also to be true, as using long-term rates in the regressions does not change the results. Thus, whereas different interest rates along the yield curve may affect flows into bonds of different maturities, their effect on total flows into bonds of all maturities is not statistically significant in these data.

¹¹ For details, see Qureshi et al. (2019).

¹² To see dynamic relationships between stock price and mutual fund flows, we include other financial markets including bond and money markets in our system method to minimize any issues regarding model misspecifications, omitted variables, simultaneity, and cross-equation correlations which may cause serious endogeneity problem.

¹³ It is SURECM when SUR is applied to Equations (3)(3) $\mathrm{\Delta }{Y}_{it}={\mu }_{0i}+{\lambda }_{0i}{\hat{Z}}_{it-1}+\sum _{j=1}^k{\delta }_{1,ij}\mathrm{\Delta }{X}_{it-j}+\sum _{j=1}^k{\delta }_{2,ij}\mathrm{\Delta }{Y}_{it-j}+u_{1,it}$(3) and (4(4) $\mathrm{\Delta }{X}_{it}={\mu }_{1i}+{\lambda }_{1i}{\hat{Z}}_{it-1}+\sum _{j=1}^k{\varphi }_{1,ij}\mathrm{\Delta }{X}_{it-j}+\sum _{j=1}^k{\varphi }_{2,ij}\mathrm{\Delta }{Y}_{it-j}+u_{2,it}$(4) ). As mentioned above, i represents stock, bond, and money markets.

¹⁴ For maximum lags, we use Hall’s general to specific method.

¹⁵ See Elliot (1998) for details.

¹⁶ We perform an ADF test on the residuals without intercepts from Equation (1)(1) $Y_{it}={\alpha }_i+{\beta }_i{X}_{it-1}+\sum _{i=1}^3\sum _{j=-p}^p{\theta }_{ij}\mathrm{\Delta }{X}_{it-j}+{\epsilon }_{it}$(1) and use the critical values from Engle and Yoo (1987). The critical values for the test of cointegration are −3.37 and −3.03 at 5% and 10% levels, respectively.

¹⁷ Since there are no qualitative differences between the 3-year and the 5-year government bonds, we report only the 5-year government bond here. The empirical results of the 3-year government bonds are available upon requests.

¹⁸ We test the causality hypothesis that stock returns (net equity fund flows) do not lead net fund flows (stock returns).

¹⁹ This is based on the results with the ETFR database.

²⁰ This is the well-recognized observation in King and Levine (1993) and Levine (1997) regarding financial development and economic growth in the long run.

²¹ For details, Baker and Wurgler (2007).