Momentum in Australian style portfolios: risk or inefficiency?
Academic Research | 29 October 2019
The existence of momentum in the cross-section of stock returns has received widespread empirical support from academics and practitioners alike. This paper examines a different type of momentum, ‘style’ momentum, in the Australian stock market. In contrast to momentum in individual stocks, style momentum is the return continuation of portfolios that are sorted on stock characteristics like price-to-book (often called a value strategy) and size. These style portfolios are a proxy for managed portfolios that apply a consistent investment philosophy. The authors find significant profits by investing in style portfolios that have performed well recently and selling style portfolios that have performed poorly recently.
The existence of style momentum in the Australian stock market is an interesting finding; however of equal importance are the sources of these profits. While most discussions on the sources of an anomaly focus on either risk or behavioural reasons, this paper takes a different tact and asks, how much of the momentum profits can be explained by risk and how much can be explained by behaviour? The authors decompose the style momentum profits into three sources; variance (supportive of a risk explanation), autocorrelation (supportive of a behavioural explanation) and cross-sectional variance (supportive of market frictions). The authors show that the majority of the style momentum profits come from autocorrelation (between 56% and 83% depending on the method used to construct the style portfolios), variance accounts for between 48% and 28% and cross-sectional variation between -2% and 4%. These results suggest a substantial amount of inefficiency in the Australian stock market that can be exploited by the use of a momentum strategy.
Click on the below link to read more about this research:
Chan, H., & Docherty, P. (2016). Momentum in Australian style portfolios: risk or inefficiency? Accounting & Finance, 56(2), 333-361.