What Determines Investors’ Purchases and Redemptions?

Abstract

This study examines the investment flows of Spanish domestic equity funds. Increasing research is specialized in analyzing the determinants of fund flows, but few studies have been able to investigate buying and selling behavior separately. Unique data from Spain allow us to tackle this issue by using exact information of purchases and redemptions. We find that investment flows are sensitive to past performance, though this sensitivity is more noticeable when using inflows and outflows in contrast to implied flows. Furthermore, our empirical results suggest asymmetric behavior of investors given that purchases are better explained by fund return than redemptions. In general, investors are risk averse and fee-sensitive and do not value greater size of Spanish equity funds.

Keywords:

• Investor flows
• Money flows
• Purchases
• Redemptions
• Spanish equity funds

ACKNOWLEDGMENTS

The authors would like to thank the financial support provided by the University of Zaragoza Grant 268-196 and the Spanish Department of Science Grant EC02009-12819-C0302. Any possible errors contained in this article are the exclusive responsibility of the authors.

Notes

1. At the end of 2006, Spain managed 4.69% of the assets under management in the European UCITS industry according to the European Fund and Asset Management Association (EFAMA).

2. Mutual Fund Fact Book 2007 of the Investment Company Institute (ICI).

3. To include a fund in our domestic equity fund sample, it must belong to this category at the end of the each year given that funds in the sample are revised every year.

4. As far as we know, only the paper of Guercio and Tack [2002] analyzes investor flows when analyzing the behavioral patterns of U.S. pension fund investors.

5. We analyze percentage flows instead of absolute flows since these measures seem to be more appropriate given that they account for the assumption that larger funds will attract larger flows. Moreover, these implied percentage flows are calculated following the approach proposed by Gruber [1996] that assumes that investors in merged funds continue in the surviving fund.

6. Table 2 only shows the results obtained when considering 3 and 12 months. The results achieved when using 24-month periods are quite similar and can be obtained from the authors.

7. The results of the other combinations are similar and could be requested from the authors.

8. Ferruz et al. [2009] also note that the average size of Spanish funds is very small with respect to the average in Europe. This is due to the large number of funds commercialized in Spain.