Heterogeneous information in market microstructure

Zhang, Xiangguo (2023). Heterogeneous information in market microstructure. University of Birmingham. Ph.D.

Preview

Zhang2023PhD.pdf
Text - Accepted Version
Available under License All rights reserved.
Download (1MB) | Preview

Abstract

This dissertation studies topics in the areas of information in financial markets.I model the competitive financial market where investors own heterogeneous information signals about trading assets. This dissertation studies the effects of strategic decisions by investors and learning on asset prices and investors' decision-making in financial markets. I mainly focus on the strategic trading decisions by investors and the impact of imposing market structures with different informational environments. Specifically,The first main chapter studies the quality aspect of information disclosure(i.e., the precision of private information), and the second main chapter analyzes the different investment ability among investors(i.e., looking-forward abilities). Finally, the third chapter analyzes the speed aspect of information, such that, how quickly an insider can process information and act on it before the information disclosure by public sections.

In the first chapter, Market Efficiency, the Cost of Capital, and Information Disclosure Quality. I explore the two-period noisy rational expectations claim in a model with short-horizon traders who have constant absolute risk aversion (CARA) preferences. A coherent framework is presented, showing how, when asset information is heterogeneous and includes both private and public signals, the precision of information disclosure reflects fundamentals, and liquidity trading in perfect competition. Alternatively, as a result of the noisy REE framework with linear expectations equilibrium, the results suggest that higher precision of public information disclosure even under the influence of Keynesian beauty contests still improves the financial market's efficiency level. As long as the competitive and information-related conditions are perfect, and traders aversion risk and risky asset supply have a positive relationship with the cost of capital. Crowding-out effects reduce agents' incentives to protect private information when public information disclosure is sensitive to market participants.

In the second chapter, Level-k Reasoning with Heterogeneous Information Signals. I study a level-k reasoning equilibrium in an asymmetric information environment populated by informed/uninformed agents. Firstly, when private information signal noise is significantly higher than public information signal noise, increasing the level of public information disclosure can lead to an increase in agent's payoff. Therefore, the social planner can improve market efficiency by disclosing more public information when the private signal is highly noisy. Secondly, consider a specification of L0 reasoning with uniformly randomly distribution between public and private signals; Higher-level reasoning agents can have dominant optimal strategies because of their higher prediction levels, but less sophisticated agents can also make better rational strategies under certain conditions. Specifically, this occurs when there are lower expectations of the fundamental state of the economy and a higher symmetric information environment. In these cases, less sophisticated agents may outperform higher-level reasoning agents. Finally, I analyze the Lk reasoning with full and partial disclosure of the public signal. The generalized beauty contests in Morris-Shin(M.S.) model is the particular case of partial disclosure in the Lk reasoning hypothesis when the social planner endogenously fixes the mass of the fraction.

In the third chapter, Unobservable Information Acquisition and Insider trading. I focus on modeling strategic communication of asymmetric information among investors and market makers, considering the costly acquisition of information.Firstly, I extend Kyle's dynamic insider trading model(Kyle, 1985) to incorporate unobservable, costly information acquisition, and the analysis reveals that there is no pure strategy equilibrium where information acquisition occurs after the transaction. This implies that insiders always deviate from the equilibrium by either entering the market earlier or later than the proposed disclosure date. Secondly, I discuss a discrete-time dynamic model of unobservable information acquisition following Caldentey and Stacchetti's (2010) study, which extends Kyle's (1985) continuous-time insider trading model by allowing discrete-time trading and unobservable information acquisition. Thirdly, as part of Kyle's (1985) contest, I consider the continuous-time dynamic insider trading model with no discounting. In this scenario, if the cost of information acquisition is high enough, it is unprofitable for insiders to acquire information. The final section discusses the implications for legal regulation of the SEC penalty of insider trading in U.S. cases, emphasizing how insiders' abnormal profits and illegal trading behaviors are determined by information disclosure.

Type of Work:

Thesis (Doctorates > Ph.D.)

Award Type:

Doctorates > Ph.D.

Supervisor(s):