A wonderful paper scheduled for April publication in the Journal of Finance takes a new twist on the Dogs of the Dow strategy. The paper, by Boudoukh, Michaely, Richardson, and Roberts is titled, “On the Importance of Payout Yield“. SmartMoney magazine also has a good overview of the paper in their “Stockscreen” column.
In an earlier post we examined a few Dogs of the Dow strategies. One of the problems with the original Dogs strategy (buy the top 10 yielding stocks in the Dow, rebalanced yearly) is that the performance has deteriorated in the past decade or two. Below we will examine this new strategy based on payout yield (and leave a comment if you have a name suggestion for this strategy which I will track on Stockpickr).
Dividends are only one way of returning capital to shareholders. Share repurchases are another such method (see MSFT), and since they are not taxed like dividends, it can be argued they are a more efficient way of returning profits. Buybacks represent about half of all shareholder payouts, and have increased steadily since the early 1980’s. There is a structural reason for this, and is due primarily to the SEC instituting rule 10b-18 in 1982 – providing a safe harbor for firms conducting repurchases from stock manipulation charges. See Grullon and Michaely  for more info on the impact of Rule 10b-18.
The authors examined the payout yield and net payout yield, whose formula is:
Payout Yield = $ spent on dividends + $ spent on share repurchases
(Net payout is simply subtracting the $ raised through new share issues to the above formula)
The authors find that “the widely documented decline in the predictive power of dividends for excess stock returns is due largely to the omission of alternative channels by which firms distribute and receive cash from shareholdlers.” Additionally, while dividend yield has lost its predictive ability over time, the payout yield has remained a robust indicator for excess stock return.
From 1983 – 2003 the various strategies returned:
The current names will be tracked here:
(in descending order of NPY)
Previous research showed that the dividend yield process changed remarkably during the 1980’s and 1990’s, but that the payout yield (dividends plus repurchases over price) changed very little. As such, we investigate the empirical implications of using various measures of payout yield rather than dividend yield for asset pricing models. We find that the widely documented decline in the predictive power of dividends for excess stock returns is due largely to the omission of alternative channels by which firms distribute and receive cash from shareholders. Statistically and economically significant predictability is found in the time series when payout (dividends plus repurchases) and net payout (dividends plus repurchases minus equity issuances) yields are used instead of dividend yield. In the cross-section, we find that payout yield contains information about expected stock returns exceeding that of dividend yield and that the high minus low payout yield portfolio is a priced factor. Finally, we show that trading on this characteristic leads to excess
profits that can not be explained by the traditional risk factors.