While, on average, portfolios composed of value stocks have the tendency to provide higher average and median monthly returns than portfolios composed of growth stocks for all three price-multiples on which portfolios are classified, the results are, overall, too small and statistically insignificant to define it as a true value premium. Another point of interest is the finding that both global value and portfolios do not outperform the market or the risk-free rate securities. This means that, during the financial crisis, no superior return could be earned for investing in value and growth stocks. This failure to reject the null-hypothesis provide an answer to the main research question; what class of stock, value or growth, offered the highest return during the financial crisis of 2007-2010. From a statistical point of view, the answer to my research question is that neither value stocks nor growth stocks offered the highest return. From a practical viewpoint, value stocks offered the highest return by only a few basis points.
The portfolio return in association with systematic risk, as measured by Jensen’s Alpha and Treynor, indicate that portfolios composed of value stocks show higher return per unit of risk than portfolios composed of growth stocks. For Jensen’s Alpha, this means that the compensation of value portfolios, by an extra Alpha (return) versus the return and risk existing in the market, was higher than the compensation for growth portfolios. In case of Treynor, value stocks provide a higher return than growth stocks in relation to what the return would be in a riskless investment per unit of (systematic) market risk. However, the compensation in rate of return for portfolios composed of value stocks is only higher by a few basis points as compared to portfolios composed of growth stocks. Overall, this means that the outcomes obtained from the two-sample one-tailed t-test and two-sample one-tailed Mann-Whitney test are too small to statistically conclude that statistically portfolios composed of value stocks provide higher average and median monthly Alphas and Treynors than portfolios composed of growth stocks during the financial crisis. Therefore, the null-hypothesis, that there is indifference in average and monthly outcomes in risk-adjusted measures, cannot be rejected. From a statistical viewpoint, I therefore fail to reject the null-hypothesis and conclude that there is no statistical difference between value and growth stocks regarding the outcomes on the risk-adjusted measures. This failure to reject the null-hypothesis provides an answer to the first sub-question; whether value stocks offered a higher return per unit of systematic risk than growth stocks during the financial crisis of 2007-2010. From a statistical point of view, value stocks do not offer a higher return per unit of systematic risk, as measured by Jensen’s Alpha and Treynor, than growth stocks. From a practical point of view, the outcomes on Alpha and Treynor are higher than for growth stocks which insinuate that higher return per unit of systematic risk for value stocks.