Three essays in market microstructure

The purpose of my dissertation is to provide some empirical evidence and theoretical insights concerning several important innovations in financial markets. In the first chapter of my dissertation, I consider the change in the quality of the market for the FTSE 100 Index futures contracts after the London International Financial Futures and Options Exchange (LIFFE) transferred trading in these contracts from the open outcry to the electronic trading.; I measure market quality as the variance of the pricing error, based on the Hasbrouck's (1993) model. Open outcry market is found to have higher market quality. Due to the specifics of the floor trading, open outcry market is able to rapidly incorporate information into the prices at the times of intensive information arrival. I also find that degree of information content of trades is higher in the open outcry market. Inventory control is more costly in the electronic market.; In the third chapter I investigate what determines the profits of the liquidity providers (limit order traders) at the LIFFE after automation. Contrary to locals, liquidity providers at the open outcry exchanges, limit order traders at the LIFFE cannot anticipate the order flow and front-run it because there is no physical trading location (pit). Their trading profits are determined by the inventory control considerations and asymmetric information, and resemble those of the designated liquidity providers at the dealer markets. Therefore, an electronic futures market operates more like a traditional exchange where the liquidity providers only charge competitive fee for providing immediacy.; The necessity to study market efficiency and new developments (such as creation of new trading venues and mechanisms) induced and made popular common factor models of Hasbrouck (1995) and Gonzalo and Granger (1995). The purpose of the second chapter of my dissertation, is to show the relationship between the two common factor models, and to provide the insights into which model helps to better understand how markets work. The two models complement each other and provide different views on the price discovery. Hasbrouck model focuses on the market's contribution to the variance of the innovations to the efficient price—a random walk, also the common factor. Gonzalo and Granger focus on the error correction mechanism, with the market adjusting to the equilibrium contributing less to the common factor. When the residuals between the markets are uncorrelated, the two models attribute price discovery consistently. When the substantial correlation exists, the models produce different results. Gonzalo and Granger measure of price discovery does not incorporate correlation among the markets.