Three Essays On Market Microstructure

Abstract

My dissertation consists of three essays. The first essay develops a price impact function when competitive market makers are risk-averse. The essay proves that price change is linear both in current and lagged order flows. The coefficient of the current order flow measures price impact while the coefficient of the lagged order flow represents price reversal. The price impact reflects both informational and inventory costs while the price reversal reflects only an inventory cost. The first order serial covariance is shown to be negative and be proportional to both price impact and price reversal coefficients. The second essay analyzes the risk of merger arbitrage strategy. The information components in the bid-ask spreads of both stocks decrease after the announcement. The inventory components for the target stocks significantly increase after the announcement. The Granger-causality tests show that the unexpected shocks of the bid-ask spreads cause the increase in the merger arbitrage spread. The informational share of each stock after the announcement is similar to the ratio of the market value of each stock to combined firm values prior to the announcement. The third essay investigates the effect of liquidity on the exchange-traded-funds. Both illiquidity and ETF tracking errors are positively related and are persistent. The empirical tests show that illiquid ETFs tend to be more sensitive to underlying index returns or market liquidity. There also exists a positive liquidity premium in the US ETF markets. Finally, the ETF return variance can be decomposed into the NAV return variance plus additional terms associated with no trading probability. Empirical tests show that ETF variances are typically larger than NAV variances when ETFs are not traded actively in the market.