بازده سهام و جریان اعتماد سرمایه گذاری در بازار مالی ژاپن: رویکرد سیستم

To study dynamic and causal relations between stock returns and investment trust flows in Japan, we employ a system method which utilizes information from the stock, bond, and money markets. The empirical evidence from SURECM, and Granger (1969) and Sims (1972) causality tests in the system method indicates that investment trust flows are weakly exogenous and stock returns cause net fund flows, 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 a driving force behind rallies in Japanese financial markets.

This paper investigates the dynamics and the possibility of a causality between stock returns and investment trust flows in the Japanese financial market using a system method.1 The degree to which prior stock market returns influence investor demand for mutual fund shares and to what extent this demand drives returns have important implications for the stability of the U.S. stock market. Substantial efforts have been made to detect the relationships explained by security returns and mutual fund flows of the U.S. market, however the Japanese mutual fund market which is one of the most advanced and largest mutual fund industries in the world has not been extensively researched and therefore, relatively little information is known on the behavior of cash flows of investment trusts and its relation to security returns. While mutual funds have grown to become a dominant vehicle for savings in the U.S. over the past decade, its Japanese counterpart, the investment trust sector – a term that includes both closed-end and open-end funds – has grown more modestly. It is our interest to see whether the fund flows have been a driving factor of equity returns in the Japanese market. In analyzing the relations between stock returns and mutual fund flows, there are two different approaches, a microapproach and a macroapproach. The microapproach focuses attention on how mutual funds flows are analyzed on individual bases. Among the studies using the microapproach, Ippolito, 1992, Hendricks et al., 1993 and Grinblatt et al., 1995, and Sirri and Tufano (1998) show that investors tend to move cash into the funds that had the highest returns in the preceding years and find little support for the assumption that flows drive performance. On the other hand, the macroapproach analyzes large scale movements of money into and out of the market without regard to which fund it goes into or comes from. Hence, the research at the macrolevel has centered on the relationship between stock market returns and aggregate mutual fund flows. Warther (1995) pioneers study at the macrolevel and finds evidence of a positive relation between flows and subsequent returns but a negative relation between returns and subsequent flows. His results support the popular belief that fund inflows and returns are positively related. Empirical research including Remolona et al., 1997, Edelen and Warner, 2001, Goetzmann and Massa, 2003 and Boyer and Zheng, 2004 indicates that in general a high positive correlation exists between aggregate mutual fund flows and stock returns. Further, the positive correlation between aggregate mutual fund flows and stock returns is supported by theoretical approaches such as the price pressure theory, the information revelation hypothesis, and investor sentiment.2 The high positive correlation between fund flows and stock returns, however, does not necessarily imply that the former causes the latter and vice versa because there might be other possible reasons for the causal relationship.3 Furthermore, it is not clear whether the stock market is driven by mutual fund inflows and outflows due to the fact that prior studies on stock returns and mutual fund flows employ overly simple regression approaches such as least squares methodologies with logged differenced or normalized data, which removes the unit root or permanent component of the data and therefore avoids the complications related to unit roots and spurious regressions. Since business cycle activity comprises both temporary and permanent components, the removal of the permanent component loses valuable long term information concerning the evolvement of short-term movements. To examine whether the monthly net fund flows have been a driving factor of equity market returns in the recent increase of the Japanese market we employ a seemingly unrelated regression error correction model (SURECM), and two causality tests proposed by Granger (1969) and Sims (1972) in a system method. More specifically, when markets have a close relationship with one another, a seemingly unrelated regression methodology is presented to account for cross equation correlations among markets. Furthermore, we make use of information in the variance–covariance matrix of residual to improve the efficiency of the statistical estimates. This study is distinctive from the existing literature in that the system method utilizes information from the stock, bond, and money markets to improve the efficiency and provides more economically reasonable estimates. Our empirical evidence from the system method shows unidirectional positive causality from stock returns to net fund flows. Further, if there is a deviation from long-run equilibrium, the stock returns force the deviation to go toward the long-run equilibrium, implying that the investment trust flows are weakly exogenous and do not respond to eliminate the deviation from long-run equilibrium in the Japanese market.

To investigate the dynamic and causal relations between stock returns and investment trust flows in the Japanese market at the macrolevel, this study employed a system method which combines information of the stock market with information of bond and money markets. Even though there is no standard way to interpret the empirical results, our findings based on the system method can be explained in several ways. The empirical evidence from weak exogeneity of the net trust flows and the positive causality from stock returns to net investment trust flows can be explained by the investor sentiment. Several empirical studies have shown the significance and the profound impact of investor sentiment on the U.S. equity market. Our empirical findings are not likely to support the popular notion of mutual fund flows as a driving force behind rallies in the Japanese security markets. Instead, the empirical evidence shows that at the aggregate level, there is evidence that investors react to securities performances at monthly frequency, and investors psychology or expectations lead to move their money to the securities that yield higher returns to rebalance their investment portfolios in the short-run. This type of behavior seems to be rational for those who try to maximize their individual returns; in addition, this investment strategy disciplines fund managers and aligns their interests with those of investors as well. Our empirical findings do not support the popular notion of investment trust flows as a driving force behind rallies in security markets. The most important element explaining investment trust flows seems to be security performance in the Japanese financial market.