Return to search

Value and size investment strategies: evidence from the cross-section of returns in the South African equity market

Value and size related equity investment strategies are supported by a large body of empirical research that shows a persistent premium, both longitudinally and crosssectionally. However, the competing rational and behavioural finance explanations for the success of these strategies are a subject of debate. The rational explanation is that the premium earned on value shares or shares of small companies can be attributed to higher risk. Behaviouralists argue that such shares are not riskier and attribute the premium to cognitive errors and biases in human decision making. The purpose of this study is to determine, firstly, whether the value and size premium exist in South Africa during the period July 2006 to June 2012, which includes one of the worst equity market crises in history. Secondly, this study sets out to determine whether the premium earned on value and size strategies are adequately explained by the principles of rational finance theory. To provide evidence regarding the existence of the value premium and size effect, returns are analysed, cross-sectionally, on portfolios of shares sorted by value and size. For evidence of a rational explanation, returns are regressed on value and size variables, and the relative riskiness of value and small companies is analysed. The results show evidence of a value premium in portfolios of small companies, but not big companies. The size effect is found not to be statistically significant. While regressions do show significant relationships between value and size variables and returns, these variables are found not to be associated with higher levels of risk. The conclusion is that the evidence does not support a rational, risk based explanation of the returns

Identiferoai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:rhodes/vital:874
Date January 2013
CreatorsBarnard, Kevin John
PublisherRhodes University, Faculty of Commerce, Accounting
Source SetsSouth African National ETD Portal
LanguageEnglish
Detected LanguageEnglish
TypeThesis, Masters, MCom
Format147 p, pdf
RightsBarnard, Kevin John

Page generated in 0.0023 seconds