Dividend-based investment strategies are often hailed as an easy and simple way of identifying value investments. The actual mechanics of a dividend strategy do intuitively seem to point more to an identification of undervalued stocks, even though they do not directly contain any elements that are typically connected with value investments. But how much do dividend-based and value-based selection methods really differ, and how similar are the results? Which method is more effective in identifying true value investments, and can one effectively be replaced by the other? How much of the returns shown by the respective indexes are actually attributable to the value factor?
In this article, we will examine the difference in the characteristics of the stock selection made by value and dividend schemes using the results of the application of the index rules for a standard dividend index as well as a typical value index. As a basis, we compare all results to a standard market-cap-weighted benchmark. To ensure a consistent data approach, all fundamental data used in this article have been sourced from Bloomberg with the same effective date. This ensures that no inconsistency arises due to the differing data cutoff dates within the index methodologies when comparing the results of one selection method with the other based on the respective selection criteria. Furthermore, we compare the factor breakdown of index returns for a value index versus a dividend index using a sophisticated multifactor model.
We will take a short look at the selection rules governing the indexes. While selection of dividend indexes is primarily focused on selecting stocks with a high dividend yield, the selection mechanics of value indexes are more complicated, as different fundamental factors need to be considered. The three indexes mentioned in Figure 1 serve as representatives for the general group of dividend indexes, value indexes and general market indexes, respectively. Different indexes or investment schemes do of course differ in various aspects of the methodology; however, the key selection variables and mechanics are common among these schemes, allowing a broader comparison with similar strategies. Figure 1 outlines the key selection criteria for the three indexes.
In this article, we do not aim to compare the performance characteristics of these indexes and their resulting portfolios, as multiple differences in the methodologies dilute the value of such a comparison. The dividend index, for example, uses a different weighting scheme, and—with 30 components—has significantly fewer components compared with the value index, which is much closer to the original benchmark that it is modeled on. This can also be easily observed by looking at Figure 2, as well as some key data for the three indexes in Figures 3 and 4.