How Informative are Reported Holdings? Evidence for U.S. Domestic Equity Mutual Funds

Abstract

It is not uncommon for mutual fund managers to make significant adjustments to their allocations before closing out their portfolios at the end of the quarter. The common wisdom on the street and in the financial press is that part of this activity is due to dressing which has become widespread as investors are becoming increasingly sophisticated in analyzing fund holdings as well as past returns in an effort to detect manager skill. So far, the academic literature on equity mutual funds has provided little empirical evidence to this effect. We analyze the semi-annual holdings and daily net asset values of 4,025 U.S. domestic equity mutual funds over the period from 1997 to 2002 and find strong evidence of increased turnover during the last days of the quarter, consistent with window dressing. In particular, we show that growth funds and funds with poor recent performance are more likely to report misleading holding. Furthermore, the end of quarter trading activity is not easily accounted for by momentum/relative strength strategies and is not associated with strategies that on average provide any added value to investors, even before accounting for expenses. Nor can liquidity costs explain these findings.