Two essays in empirical finance

This dissertation contains two empirical essays in finance. The first essay empirically examines the assumption that investors can buy and sell any amount of a firm's equity without specifically affecting the price of the asset. The second essay examines the question: who better predicts the market value of a factor of production, the manager (sophisticated trader) or game prices (unsophisticated traders)?;The first essay measures the liquidity of the 105 most traded stocks at the Tel-Aviv Stock Exchange. We estimate the elasticity at the opening stage directly from the demand and supply curves. We show that the demand, the supply and the excess demand are inelastic. We estimate the price impact of each order as the difference between the equilibrium price and the hypothetical price without each specific order. We find significant differences between the demand and the supply curves. Our findings are consistent with the notion that buy orders are more informative than sell orders.;The second essay uses a field experimental approach and investigates who better predicts the ex-post market value of a baseball player's performance, the management team (sophisticated traders) or the game's prices (unsophisticated traders) and what affects this predictive ability. Since in the market microstructure literature we divide investors into liquidity traders and informed traders, such a comparison is of interest. We compare the salaries paid to the players and pre season game prices (as a measure of current year performance) to estimates of ex post performance. Not surprisingly, we find higher correlation between the game pre season prices and the ex post prices than between the salary and the ex post prices. However, we find that the naive prediction—future performance is equal to past performance—outperforms both the ex ante game prices and the salary. Moreover, in a multiperiod regression of a measure of ex post performance on past performance, salary, ex ante game price, and some dummy variables, only past performance has a significant coefficient. We conclude that both the ex ante game prices and the salary contain no predictive ability over and above that included in past performance.