Current research on the value premium, the outperformance of value stocks in comparison with other (growth) stocks, mostly involves explaining it through either behavioral or efficient-market mechanisms. However, in their November 2005 paper entitled “Does the Value Premium Really Exist in the UK Equity Market?”, Panagiotis Andrikopoulos, Arief Daynes, David Latimer and Paraskevas Pagas challenge its existence. Their study focuses on eliminating any possible effects of survivorship bias, look-ahead bias and the method of calculating returns in comparing the performance of value and growth stocks of United Kingdom firms. They classify value versus growth via four selection factors (low for value, high for growth): book-to-market value, earnings-to-price ratio, dividend yield and weighted average sales growth. Using a new database of 2006 UK equity issues fully listed at any time during 1987-1996, they find that:
- There is some weak evidence of a value premium in the UK stock market during 1987-1996.
- The value premium is larger for stocks that satisfy two selection factors rather than just one.
- Value premiums for equal-weighted portfolios tend to be larger than those for market-weighed portfolios, indicating a modest but persistent small-firm effect.
- The value premiums discovered are much smaller than those reported in earlier studies, and are in general economically and statistically insignificant.
- The results of the paper favor semi-strong form market efficiency.
In summary, the empirical foundation for value investing may not be as sound as generally thought.
