이 연구는 Gruber(1996)와 Zheng(1999)이 제시한 펀드투자자의 펀드선정능력, 즉 스마트머니 효과에 대하여 국내 시장을 대상으로 Keswani and Stolin(2008)의 연구방법론을 사용하여 실증 분석하였다. 분석결과를 요약하면 첫째, 분석기간 중 개인투자자의 자금증가율이 기관투자자보다 크고 개인투자자와 기관투자자간의 자금흐름의 상관관계는 매우 낮다. 둘째, 국내 펀드투자자가 스마트하다는 스마트머니 효과(smart money effect)의 존재에 대한 어떠한 증거를 발견하지 못했다. 셋째, 펀드투자자의 펀드선정능력의 정도를 분석하면 시장에서의 인기펀드가 비인기펀드보다 오히려 성과가 저조하다. 넷째, Zheng (1999)의 정보효과(information effects)를 분석한 결과, 펀드자금흐름정보를 활용하여 역의 정보효과가 존재하며 다섯째, 시장국면별 분석과 조건부성과 측정모형을 적용한 결과도 스마트머니 효과의 존재를 발견할 수 없었다.
이 연구의 의의는 다음과 같다. 첫째, 펀드자금흐름의 두 가지 측정방법, 즉 펀드별 순자산․수익률에 기초한 내재적 측정방법과 펀드별 직접 측정방법을 비교분석하였다. 둘째, 개인투자자(공모펀드)만의 기존연구에서 기관투자자까지 연구대상에 포함하여 분석을 시도하였다. 셋째, 펀드매입과 환매를 일별 데이터를 사용하여 보다 정교하게 자금흐름을 산출하여 스마트머니 효과를 분석하였다.
The smart money hypothesis states that investor money is “smart” enough to flow into the funds that will outperform in the future; that is, the investors have an ability to identify superior managers and invest accordingly. The first studies to address this issue (Gruber, 1996; Zheng, 1999) find that, indeed, funds that receive greater net money flows subsequently outperform their less popular peers. This pattern was termed the “smart money effect.” Also, more recent research (Keswani and Stolin, 2008) finds a robust smart money effect in the United Kingdom and the U.S. The effect is caused by buying (but not selling) decisions of both individuals and institutions. On the other hand, Sapp and Tiwari (2004) could not detect the smart money effect when they applied a more appropriate performance evaluation procedure that takes the stock returns momentum effect into account.
To understand better how different types of investors make their fund buying and selling decisions, we briefly present evidence on the determinants of mutual fund money flows in Korea. Our dependent variables are net flows and their components that are expressed as proportions of fund value at the end of the month. We would expect lagged fund cash flows and fund returns to be the primary determinants of the fund cash flows. Control variables are logarithms of fund total net assets (TNA) and fund age, as well as fund total expenses (Chevalier and Ellison, 1997; Sirri and Tufano, 1998). Our results, based on the time series of cross-sectional regression coefficient estimates, indicate that our flow variables are persistent. Coefficient estimates for lagged flows and TNA are always positive and significant. But those for past performance, fund age, and fund expenses are negative and significant.
In this paper, we examine the smart money issue with Korean daily data (January 2002 to May 2008). Owing to data constraints, all of the above studies work with the aggregate money flows to funds. All investors are aggregated, and sales are offset by repurchases. Furthermore, not having access to exact net flows, these papers approximate such flows using fund TNA and fund returns. But our data allow us to conduct a stronger test for the smart money effect by using daily data on exact fund flows, and to gain greater insight into investors’ decisions by considering separately the sales and purchases of individual and institutional investors in the Korean fund market. So we observe exact flows rather than approximations based on fund values and fund returns.
To examine the smart money controversy, a simple way is to evaluate the performance of all “new money” put into mutual funds by investors. A natural benchmark against which to measure the success of these new investments is the performance of “old money”, that is, of assets already in place before the latest round of investments. We characterize our fund portfolios using what Zheng (1999) calls “the portfolio-level approach.” Specifically, each month we conduct a Carhart (1997) four-factor regression for every fund portfolio to obtain our four estimated factor loadings. Our new money portfolios don’t deliver higher alphas than old money fund portfolios. In other words, new money is not, in fact, smart. We also use a different methodology above, which involves sorting funds into positive and negative flow groups. The results are similar.
On the other hand, in order to examine the pervasiveness of Korean investors’ ability to select superior funds, we compare equally weighted groups of popular and unpopular funds. This approach curtails the influence of funds with extreme flow observations. To understand which flow components drive this result, it is desirable to apply the same methodology to all the flow variables comprising net flows. Specifically, each month we sort funds using our measures of normalized money flows into high flow portfolios and low flow portfolios. We then compare the risk-adjusted performance of equally weighted high and low flow portfolios. The results show that the average alphas of the high flow portfolios are not larger than those of the low flow portfolios. This means that we no longer find this smart money effects. Only the result from the trading strategies confirms adverse Zheng (1999)’s information effect, that is, relatively investors can beat the market by investing in negative money flow, low money flow funds.
To verify that the smart money effect holds throughout the empirical period and appraising methodology, we use alternatives approach. First, we repeat our analysis separately for the first and last halves of our Jan. 2002 to May 2008 study full period. Indeed, the contributions of the two subperiods are of similar results. Second, we show portfolio-level approach results using Ferson and Schadt’s (1996) conditional performance evaluation. Specifically, we follow Wermers (2003) and Kacperczyk et al. (2005) in implementing the conditional version of the Carhart (1997) model with the risk-free rate and term structure premium and credit spread representing the conditioning set of publicly available information. The results are qualitatively similar to those under the unconditional portfolio approach. There is no smart money effect on the basis of net flows, inflows and outflows. The results once again confirm that inflows and outflows don’t give rise to the smart money effect in the Korean fund market.
