ESSAYS ON MUTUAL FUNDS

Open Access
Author:
Wang, Ying
Graduate Program:
Business Administration
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
May 06, 2008
Committee Members:
  • Jingzhi Huang, Committee Chair
  • Jean Helwege, Committee Member
  • James Alan Miles, Committee Member
  • Quang H Vuong, Committee Member
Keywords:
  • Market Liquidity
  • Portfolio Holdings
  • Government Bond
  • Market Timing
  • Mutual Funds
  • Market Volatility
Abstract:
This dissertation explores three issues regarding mutual funds. The first chapter examines the ability of government bond fund managers to time the market, based on their holdings of Treasury securities during the period 1997-2006. We find that, on average, government bond fund managers exhibit significantly positive timing ability at the one-month horizon, under a holdings-based timing measure. In particular, fund managers specializing in Treasury securities are more likely to better time the market than general government bond fund managers. We also find that more successful market timers tend to have relatively higher Morningstar ratings, larger fund flows, lower expense ratios, higher Sharpe ratios, and higher concentrations of holdings of Treasury securities. In the second chapter, we study whether mutual fund managers successfully time market liquidity. Using a model assuming that fund managers attempt to maximize fund shareholders¡¯ utility by timing market exposure, we motivate liquidity timing from the fund manager¡¯s perspective. Fund managers reduce market exposure in illiquid markets and increase market exposure in liquid markets. Relative to survivors, the negative performance of non-surviving funds is emphasized when the model includes market wide liquidity indicating that liquidity timing is an important component in evaluating the performance of active fund managers. In the third chapter, we study the dynamic relation between aggregate mutual fund flow and market-wide volatility. Using daily flow data and a VAR approach, we find that market volatility is negatively related to concurrent and lagged flow. A structural VAR impulse response analysis suggests that shock in flow has a negative impact on market volatility: An inflow (outflow) shock predicts a decline (an increase) in volatility. From the perspective of volatility¨Cflow relation, we find evidence of volatility timing for recent period of 1998¨C2003. Finally, we document a differential impact of daily inflow versus outflow on intraday volatility. The relation between intraday volatility and inflow (outflow) becomes weaker (stronger) from morning to afternoon.