Cross-Sector Style Analysis Of Global Equities Based On The Fama And French Three-Factor Model

Abstract

Although the ability of the Fama and French (1993) 3-factor model in explaining style-based portfolio returns have been widely tested, no such test has been conducted on sector-based portfolios. The study conducted by Hsieh and Hodnett (2011) indicate that the resource sector yields significant abnormal returns under the capital asset pricing model (CAPM) over the period from 1999 to 2009. In addition, the book value-to-market ratio and market capitalization are found to have pervasive effects on the pricing of sector returns for global equities. Motivated by this insight, we undertake to test the ability of the Fama and French (1993) 3-factor model in explaining the variations in the global sector returns. Our test results indicate that the market risk premium is the most significant factor that drives the returns in all sectors under review. Although the positive abnormal returns of the resource sector dissipates under the 3-factor model, the industrial sector and the information technology (I.T.) sector yield abnormal returns under the 3-factor model. Unlike the empirical findings on the style portfolios, the signs and statistical significance of the exposures to the value and size risk premiums are not consistent across all sectors. This finding suggests that sector exposures are more unique and distinctive compared to the style portfolios. It could be argued that since most of the style portfolios are directly related to the value and size anomalies, any factor model that incorporates risk premiums on these anomalies would significantly explain the style portfolio returns. However, the ability of such factor model in explaining returns on portfolios formed using methodologies other than style anomalies, such as sector portfolio returns, would be questionable. Taking into account the rising global integration, sector allocation might be more effective in terms of global active portfolio management or international diversification than style allocation and country allocation.