Efficient Markets, Behavioral Anomalies, and the Evolving Practice of Investment Management

Abstract

The efficient market hypothesis (EMH) has framed the study of financial markets for more than half a century, yet its empirical status remains contested. This paper provides a focused synthesis of the theoretical foundations and empirical evidence on market efficiency and investment management. We first revisit the mean-variance foundations of Markowitz (1952), the capital asset pricing model of Sharpe (1964) and Lintner (1965), and the arbitrage pricing framework of Ross (1976), before turning to the EMH of Fama (1970, 1991). We then review well-documented anomalies—size, value, momentum, calendar, and volatility effects—and the behavioral explanations of Kahneman and Tversky (1979), De Bondt and Thaler (1985), and Shleifer and Vishny (1997). We examine the evolution of factor models from Fama and French (1993, 2015) to the machine-learning approach of Gu, Kelly, and Xiu (2020), and we assess the active-versus-passive debate in light of Sharpe (1991) and Fama and French (2010). We conclude that markets are neither perfectly efficient nor wholly predictable: prices deviate from fundamentals, yet the deviations are costly and risky to arbitrage. The practical art of investment management therefore lies in recognizing the limits of arbitrage while respecting the costs and risks of exploiting it.