Portfolio Strategies based on Fundamental Weighting: An Empirical Study of the Swedish Stock Market

Purpose: The aim of this study is to investigate if portfolios whose composition is based on fundamental values (FVs) outperform a market value (MV)-weighted benchmark portfolio in terms of mean-variance efficiency on the Swedish stock market. We also seek to answer if an investment strategy, which focuses on the difference in composition between these two weighting methods, can further enhance the assumed benefits by the use of active extension. Methodology: A quantitative study with a deductive approach has been performed. Data from non-financial firms listed on the Swedish stock market 1980 to 2009, have been studied in order to compose portfolios based on FVs. A MV-weighted portfolio has been constructed as a benchmark. Theoretical perspectives: Our thesis is based on studies which argue that the market often fails to price stocks at their true value. Hence, a MV-weighted index will not be mean-variance efficient and consequently an inappropriate proxy for the theoretical market portfolio of the CAPM. Our theoretical perspectives are further based on studies which imply that historical FVs may indicate a firm’s ability to generate future value. The benefits of active extension are also discussed. Empirical foundation: FV‐weighted portfolios outperformed the MV‐weighted portfolio on the Swedish stock market during the period 1983‐2008. Portfolios only including stocks that are the most overweighted in the FV‐weighted portfolios relative to their weight in the MV‐weighted portfolio, have performed even better. Our statistical evidence is relatively weak however, in particular for the latter claim. In some cases it may have been possible to further improve the return to risk ratio through shorting and taking additional long positions, but we have found little to no statistical evidence to prove this. Conclusions: Our results indicate that the market has made systematic errors in the valuation of stocks, putting too little emphasis on fundamental values. This is evident during the TMT‐bubble in particular, but to a lesser extent also during the other time periods of the study