This study reconsiders the determinants of flows into US growth funds, focusing in particular on the dynamics of the impact of past performance on flows.We model the flow-performance relationship at the monthly frequency, allowing for dependence of the sensitivity of flows to past performance on size and age of the fund.The dynamics of the impact of past performance is modelled using polynomial lag structures.Performance from 6 to 8 months ago seems to have thestrongest impact on net flows to US growth funds.We observe that performance during the most recent quarter is less important than performance during the remaining three quarters of the first year, suggesting that some investors react to fund performance with a certain lag.Specifications based on average past performance at annual or quarterly frequency are strongly rejected.The first three years of past performance history account for about 90 percent of the total impact of past performance on flows.The well-documented convexity of the flow-performance relationship appears robust to allowing for dependence of this relationship on size and age of the fund.The return on systematic risk factors has a small additional impact on top of the impact of risk-adjusted returns.
|Place of Publication||Tilburg|
|Number of pages||28|
|Publication status||Published - 2002|
|Name||CentER Discussion Paper|
- investment trusts
- return on investment