Thomas, V., & Gossel, Sean J. (2017) This paper aims to test the efficiency, more specifically semi-strong form efficiency, of the Johannesburg Stock Exchange (JSE) by conducting a portfolio study. Unlike event studies, which aim to assess whether abnormal returns can be attained by employing investment strategies that target specific events, the portfolio study aims to understand whether specific firm characteristics or strategies can be utilised to create portfolios of shares that generate abnormal returns relative to the market. Market efficiency in the semi-strong form assumes that all public information is synthesised and processed efficiently, and as a result, it will not be possible to obtain abnormal returns by pursuing investment strategies that make use of publicly available information. Four common firm specific characteristics namely: the price to book value (P/BV) ratio; the price to earnings (P/E) ratio; the market capitalisation; and the share price momentum were used as the defining strategies with which portfolios were created. In addition to this, random portfolios were also created in line with random walk theory. Jensen’s alpha was computed to determine whether abnormal returns were possible between 2000 and 2016, and the relative portfolio performance was determined using the Sharpe and Treynor ratios. The results demonstrate that an investment strategy consisting of shares with high price momentum outperformed the market during the time by generating significant abnormal returns. In addition to this, it was found that portfolios consisting of shares with the largest market capitalisation were able to generate significant abnormal returns over the entire period and outperform the market. These results provide evidence that firm specific characteristics as well as past share prices can be used to predict future returns and generate abnormal returns, thus providing evidence against semi-strong and weak form efficiency of the JSE