Performance of Small- and Large-cap stock portfolios- The importance of market anomalies across business cycles

Abstract

This Master´s thesis investigated the importance of the market anomalies size (market capitalization), value (Book-to-Market ratio) and momentum (lagged short-term momentum) for equity returns of small- and large-cap composite stock portfolios. The study focused on two contrasting stock markets (NASDAQ OMX and NYSE) across domestic business cycles over the time-period 2006 to 2021. Several studies focused on asset pricing have during the last decades demonstrated that the original Capital Asset Pricing Model (CAPM) has limited capacity to explain and predict cross-sectional and temporal variations in expected equity returns. Equity returns have been suggested to be influenced by market anomalies such as size, value, and momentum. In this thesis, the econometric approach included single- (CAPM) and multi-factor modelling (Fama-French Three-factor model and Carhart Four-Factor model) of composite stock portfolios based on market capitalization. Overall, there was a size-effect where the small-cap portfolios outperformed the large-cap portfolios, as well as the OMXSGI and NYSE Composite market benchmarks over the aggregate sample period. However, the relative stock performance and general patterns in equity returns of the two composite stock portfolios and market benchmarks varied within and between years. The financial indicators Sharpe Ratio and Jensen´s alpha suggested a higher risk-adjusted equity return for the small- compared to the large-cap stock portfolios, as well as for the market benchmark in a temporal isolation determined by the domestic business cycles in Sweden and the USA. Alpha-values for the small-cap portfolios seemed enhanced during boom- (more positive) or bust- (more negative) periods. In contrast, alpha-values for the large-cap portfolios were generally more at par with the market performance, especially during bust-periods. Over the aggregate sample period, however, alpha-values for both stock portfolios and stock markets seemed to converge towards zero. The multi-factor model regressions supported a positive size premium (SMB) for the small-cap portfolios and a negative size-premium for the largecap portfolios. During boom-periods, the modelling approach indicated an enhanced size premium for the small (positive) and large (negative)-cap stock portfolios. Although no clear patterns were observed for value (HML) and momentum (UMD), the multi-factor modelling revealed positive valueeffects on equity returns particularly during bust-periods and negative effects during boom-periods, irrespective of market capitalization. The UMD-coefficients seemed more pronounced (larger/smaller) from modelling of shorter time periods related to the domestic business cycles, compared to the aggregate sample period (2006-2021). Despite the observations of a size effect, particular for the small-cap portfolio on NASDAQ OMX, the financial mechanisms that controlled equity returns appeared complex. The significant temporal variations between the equity returns of the two portfolios and the market benchmarks, along with multi-factor econometric modelling, directly implied that additional financial factors other than size, value and momentum may be important for the observed equity returns. In fact, market capitalization may serve as a proxy for one or several undisclosed financial factors correlated with size.