Market Efficiency - A Structural Study

Abstract

I use eight different metrics as separate objective and systematic measures of the efficiency of the market for a stock. I develop a seven-equation (six- for non-Nasdaq stocks) structural model with market efficiency as a function of exogenous factors (transaction costs & constraints, short sales costs & constraints, and dispersion in investor valuations) and endogenous market activities (trading volume, short interest, number of analysts, institutional holdings, shares outstanding, and number of market makers (for Nasdaq stocks)), and each endogenous market activity as a function of the exogenous factors and all other endogenous market activities. I propose a theoretical model that shows that higher trading volume (or another similar market activity) is caused by lower transaction costs, lower short sales costs, and/or higher dispersion of investor valuations, and therefore, that the impact on market efficiency of transaction costs or short sales costs is an empirical question. I apply Three Stage Least Squares and Errors in Variables to estimate the structural system and test the corresponding hypotheses, using panel-based instrumentation strategies for endogenous and inaccurately measured variables. Analyzing Nasdaq and non-Nasdaq stocks separately, I find that transaction costs & constraints have a significant positive (negative) impact on market efficiency for Nasdaq (non-Nasdaq) stocks, whereas short sales costs & constraints and dispersion in investor valuations have ambiguous impacts on market efficiency, that Fama-French Factors are substantially important in affecting these relationships in sign and significance, and that market efficiency is not directly positively associated with trading volume, short interest, number of analysts, institutional holdings, and number of market makers (for Nasdaq stocks), but is directly positively associated with shares outstanding.