Window dressing in the Active Share scores in publicly reported portfolios

Using a unique database that includes publicly disclosed fund holdings at the end of quarter as well in all non-publicly months, we found some funds could alter their portfolios months to artificially increase Active Share scores and consequently appear more active take advantage positive relationship between money flows. We show how, consistent with non-informed trades, these erode future performance. However, reach objective increasing Moreover, find window-dresser can be identified by controlling level tracking error. The high low errors have highest levels window dressing worst returns. compared less funds, they are able capture higher JEL CLASSIFICATION G23; G11