Article Preview
TopIntroduction
The relationship between risk and returns lies in the core content of financial theories and regulatory practice. Modern portfolio theory suggests that investors are capable of dispersing the idiosyncratic risk through diversification to the full extent. Modern portfolio theory is a realistic strategy for picking assets with the goal of maximizing total return while maintaining a manageable degree of risk. It maintains two main notes that the entire risk and return character of a portfolio is much more significant than that of the risk and return character of any individual investment, and by knowing this, an investor may design a diverse portfolio of financial assets. To some extent, diversifying is considered to decrease the unsystematic risk without lowering profits. The idiosyncratic risk, which is defined as firm-specific risk, is independent of common market movements and generally represented by the idiosyncratic volatility of an individual stock. The traditional Capital Asset Pricing Model (CAPM) assumes that only the systematic risk, excluding idiosyncratic risk, can affect the expected return of stocks. Therefore, the systematic risk is the only risk factor in the general equilibrium asset pricing model. However, due to practical constraints such as transaction costs, investor preferences and information asymmetry, idiosyncratic volatility describes the changes in stock returns, which fails to be explained in typical asset pricing models. Besides, it exerts an important impact on asset allocation effects.
Currently, full consensus on the relationship between idiosyncratic volatility and expected returns has not been fully reached. It has always been the focus of academic circles and regulators to verify whether idiosyncratic volatility is an important risk factor affecting asset pricing and the correlation between the two. In the case that the perfect portfolio diversification is hard to be achieved. Levy (1978); Merton (1987) believe that idiosyncratic volatility is positively related to expected stock returns, and investors are expected to obtain higher risk premium for compensation. However, Ang et al. (2006) conduct an empirical study on the stocks of listed companies in the U.S. and find that there is a significantly negative relationship between idiosyncratic volatility and expected cross-section returns, which is called the “idiosyncratic volatility puzzle”. Huang et al. (2007); Fu (2009) believe that the above puzzle discovered by Ang et al. (2006) is mainly caused by return reversals. Bali & Cakici (2008) suggest that their findings are sensitive to data frequency, weighting schemes and breakpoints. With the deepening of research, the EGARCH model is adopted by Spiegel & Wang (2006); Eiling (2006) to estimate idiosyncratic volatility. Besides, they find that it is positively related to expected returns, which supports (Merton, 1987). Chua et al. (2007) decompose idiosyncratic volatility into expected and unexpected components. Further testing shows that after controlling for the unexpected idiosyncratic volatility, the expected idiosyncratic volatility is significantly positively related to expected stock returns (Saravanan et al., 2015).