Dow Dividend Strategy Historical Returns

Some things you should know about the returns shown in the table below:

• The returns shown are the compound annual returns for the period 1972-2007, a total of 36 years.
• The Std. Dev. column is the standard deviation of the 36 annual returns. No claim is made that these are "normally distributed." However, higher standard deviations indicate greater variability in the annual returns.
• The returns are based on PRICE CHANGES ONLY. Dividends are not included. This is different from the results presented by most others. I did this mainly as a matter of convenience, but there are other reasons to exclude the dividends. Historically, dividends have accounted for about half the return from investments in common stocks. However, in recent years, this has not been the case. Due to tax laws, which changed in 2003, and other considerations, many companies have decided not to increase their dividends in line with increases in earnings. Instead, some companies will purchase their own shares in the market, which is an alternative method of distributing profits to shareholders. Until this general dividend policy changes, focusing on capital gains, which is the change in price, seems to be a better way to evaluate strategies in the current environment.
• The analysis below, and those presented by other sources, ignore commissions. This is another reason not to include dividends in the calculations. One could say that the dividends will pay the commissions. However, with the advent of deep, deep discount brokers, it is possible that most investors' can pay commissions that are far less than the dividends received, even at today's relatively low level of dividends.
• Tax consequences are not included since they vary by investors.
• Any spin-offs or distributions, such as MMM's spin-off of Imation in 1996, are treated by assuming that the shares received are held until the end of the year and sold then. In cases when there have been special large cash dividends, these dividends are added to the price of the stock at the end of the year.
• With the exception of the "Foolish Four," it is assumed that an equal dollar amount is put into each stock in the strategy using the closing prices on the last day of the year. Returns are based on the closing prices on the last day of the following year. As a practical matter, most investor will not be able to make exactly equal investments in each stock.

The strategies in the table are:

• BTD5: (Beat the Dow 5) - The O'Higgins and Downes strategy of buying the five lowest priced of the ten highest yielding Dow stocks.
• High10: Buy the ten highest yielding stocks in the Dow.
• High5: Buy the five highest yielding stocks.
• PPP: (Penultimate Profit Prospect) - Buy only the second lowest priced stock in the High10 list.
• MF4: (Motley Fool's "Foolish Four") - Buy two lots of the PPP (40%) and 20% each of the third through fifth lowest priced stocks in the High10 list.
• LY: (Last Year) - Buy the last year's BTD5 stocks. For example, for 2003, buy the stocks on the BTD5 list for 2002.
• 2YR: (Two year combined) - Buy the stocks that are on this year's or last year's BTD5 lists. Since 2 or 3 stocks normally stay on the list from year to year, this will usually be 7 or 8 stocks. Execution of this strategy will likely require some rebalancing by buying or selling some shares of the stocks that are on the LY list.
  

    Strategy  Return  Std. Dev.   Best    Worst


    BTD       11.9%   18.4%        60.8%  -20.6%
    High10     9.3%   15.4%        48.1%  -12.6%
    High5      9.0%   18.6%        61.2%  -13.0%
    PPP       11.5%   46.5%       183.4%  -53.6%
    MF4       12.5%   25.4%        88.4%  -24.6%
    LY        12.0%   18.9%        60.9%  -19.0%
    2YR       12.1%   18.1%        52.2%  -16.3%


    DJIA       7.8%   15.8%        38.3%  -27.5%
    S&P 500    7.7%   16.2%        34.1%  -29.7%

From the data in the table, it is evident that strategies based solely on yield are less desirable as a rule than the others. The High10 strategy does have the advantage of being the least variable and having the worst year with the smallest loss. The High5 strategy does not seem to have any advantages. The MF4 strategy of the Gardners is clever in that now is doing better than the PPP, which has not done very well in recent years and has given up its lead in the return column above, but with considerably less volatility, which most people consider to be a risk measure. If one is willing to trade off some of the return of MF4 for lower volatility, then the 2YR strategy is an attractive alternative to the BTD5 strategy. (The last sentence was written several years ago. Since then the 2YR strategy has done worse than the BTD5 more often than not and the advantage has virtually disappeared.)

Note that all of the strategies shown have had a higher returns than the Dow or S&P. Moreover, except for the PPP, the worst year with any of the strategies was not as bad as the worst year experienced by the two averages. The larger standard deviations of the strategies (except for High10) is mainly due to the superior returns. In this case, greater variability does not mean greater risk, with the exception of the PPP.