Essays on Information Acquisition and Asset Pricing

Abstract

In this dissertation, I explore different mechanisms by which information is generated in financial markets, and whether these mechanisms can account for empirical anomalies that models without information choice have difficulty explaining. In the first chapter, I survey the theoretical literature on perfectly competitive asset markets, with a particular focus on rational expectations models with endogenous information acquisition. In the second chapter, ``The Distribution of Information, the Market for Financial News, and the Cost of Capital", I present a rational expectations model with a competitive market for financial news that provides an explanation for why stocks with a higher degree of information asymmetry tend to earn higher expected returns. I demonstrate that when a small fraction of investors hold a large fraction of a firm's private information, few investors demand a copy of firm-specific news in equilibrium. As a result, each investor must incur a larger share of the fixed cost of news production to obtain a copy, which deters investors from learning more about the firm and therefore raises their required risk premium. This result hinges crucially on the ability of investors to share in the fixed cost of news production, which suggests that the financial news media plays an important role in determining how the cost of capital varies with the inequality of information across investors. In the third chapter, ``Learning About Noise" (with Oleg Rytchkov), we study theoretical implications of endogenous acquisition of non-fundamental information in financial markets. We develop a rational expectations model with heterogeneous information and multidimensional costly learning and demonstrate that i) investors specialize in information acquisition, that is, those who are endowed with high (low) quality information about fundamentals learn only about fundamentals (noise), ii) learning about fundamentals increases the asymmetry of information, whereas learning about noise decreases it, and iii) the opportunity to learn about noise unambiguously increases price informativeness.