Stock liquidity and value relevance of earnings : evidence from a natural experiment

This thesis explores the effect of stock liquidity on the value relevance of earnings, employing the Securities and Exchange Commission's 2016 Tick Size Pilot Program (TSP) as a natural experiment to investigate this relationship. This study leverages the exogenous shock to stock liquidity induced by the program to address endogeneity issues that have historically complicated the analysis of stock market dynamics. By examining the causal effects of increased liquidity on the market's response to earnings announcements, the research provides new insights into how liquidity impacts the usefulness and reliability of accounting information from the shareholders' perspective. The findings indicate that stock liquidity positively and significantly affects the value relevance of earnings, suggesting that more liquid markets may improve the efficiency of capital allocation and investor decision-making. However, while intended to increase liquidity, the TSP inadvertently reduced it in treatment firms, diminishing the value relevance of earnings within these groups. The study further analyses the role of institutional investors, revealing that reduced liquidity discourages both transient and non-transient institutional investor participation, which in turn lowers relevance to earnings value. The study employs several robustness checks to establish causality. These results have important implications for regulators and standard setters, highlighting the potential consequences of liquidity adjustments for financial statement utility in equity valuation. Moreover, this thesis contributes to the academic discourse by filling a gap in the existing literature and providing empirical evidence on the direct relationship between stock liquidity and value relevance. This insight informs future policy decisions and helps market participants better understand the integral role of liquidity in the valuation of firms and the strategic management of accounting information.