The ‘Dogs of the Dow’ reasoning sounds sensible, the process is simple and the result are being reported as proof that it works. However, appearances are not what they seem. The basis (genes) is bad, the process (breeding) is improper and the results (papers) are false. Here’s the explanation…
Don't Miss: Enter the I4U News NES Classic Giveaway!
False papers (performance results)
Let’s cut to the chase: Performance – because investors show leniency in judging weak rationale when there are positive results attached. In addition, with virtually all recent reports giving good marks to the Dogs of the Dow’s performance, we need to address that first.
Various numbers are being tossed about, but the main scorecard reads: Dogs of the Dow +30.3% vs. Dow Jones Industrial Average +26.4%. Winner: Dogs! Wait — not so fast…
The Dog portfolio’s construction is a mismatch to the DJIA’s structure, making the return comparison apples vs. oranges. Instead of the price-weighted construction used for the DJIA, the Dog portfolio equal weights its ten stocks. The effect? That different weighting scheme is the entire reason for the performance difference. Here are the proper return comparisons, using the DJIA’s price-weighting:
- Dogs (price-weighted, excluding dividend income) +24.5%
- DJIA (as reported, excluding dividend income) +26.4%
Even including dividends, the Dogs lag behind:
- Dogs (total return, including dividend income) +28.8%
- DJIA (total return, including dividend income) +29.6%
Below are the 2013 returns (including dividends) for all 30 DJIA stocks, separating them into the Dogs of the Dow 10 issues and the remaining 20.
Special note: Hewlett-Packard. The entire difference between the Dogs’ equal-weighted and price-weighted performance can be explained by Hewlett-Packard, the top DJIA performer, up ~100% for the year. Using price-weighting, the Dog position would have been 3.54%, instead of the equal-weighting 10%. Compounding the effect was the DJIA removal of Hewlett-Packard on September 20. It was up about 50% at the time, and its weighting in the Dog portfolio had grown to 12.5%. Instead of following suit, the Dog portfolio held the stock, did not even rebalance, and so enjoyed a hefty return as the stock rose a comparable amount from there to year-end. (We will address DJIA changes and rebalancing below.)
So, there is no performance proof in 2013 that supports the Dogs of the Dow’s supposed superiority. Does this one-year underperformance proves the Dog approach lacks merit? No, because it’s too short a period to judge any strategy. However, it does free us up to turn to the more important evaluation of the Dogs of the Dow underpinnings: Its basis and process.
Bad genes (basis of Dog approach)
The Dog approach started with a good idea: Invest in stocks that are currently being ignored or disliked by investors. It’s how value and contrarian investors attempt to capture added return at reduced risk – i.e., buy low.
However, figuring out which stocks are temporarily down vs. which will wallow or fall further is a challenging process. No simple screening rules can separate one from the other. But the creator of the Dog approach claimed otherwise and then “proved” it.
- First, to control the risk of getting caught in a real loser, he focused only on the 30 DJIA stocks – the blue-chips.
- Second, he used dividend yield as the proxy for undervalued and underappreciated – i.e., value stocks.
- Third, to keep things simple and give the appearance of better diversification, the ten holdings were equally weighted.
This simplistic methodology leaves so many unanalyzed questions.
- What’s the payout behind the dividend yield? (The higher payout, the less chance of an increase and the greater chance of a decrease)
- What’s the growth outlook, especially for cash flow and earnings? (It’s from these that dividends are paid)
- What’s been the price performance? (I.e., has the stock actually underperformed and become a value – and, if so, why?)
- And so forth, on through the questions about management strategy, competitive position, financial structure, stock valuations beyond yield, etc.
But the creator forged ahead, jumping to positive performance results to override these concerns.
So, where did the performance come from? Not from actual portfolio returns but from back testing – in this case, 20 years showing that the equal-weighted Dog approach produced better returns than the price-weighted DJIA.
(Note: Throughout my career, I have seen many back-tested investment schemes come and go. Each has had reasonable sounding rationale, although most often it was created after the performance results were “discovered.” The ease with which data can be analyzed and re-analyzed means great looking charts, tables and results can be created that are simply the result of happenstance. Analysis of historical data is fine, but using “data mining” to find links and patterns for creating an investment plan is incorrect and dangerous.)
So, we can now say the basis of the Dogs of the Dow approach is overly simplistic and that the back testing using a different weighting scheme is no proof at all. But there’s more…
Improper breeding (the process for producing the results)
Now we can address the process of building and managing the Dog portfolio. Compared to wise portfolio construction and management, there are a number of flaws:
Dependence on the calendar year. The portfolio is built only once a year, using year-end dividend yield data. (Plus, all positions no longer in the top ten yielders are automatically sold.) But there is no magic in the annual calendar, other than seasonal and tax effects. Stocks happily rise and fall when conditions warrant, not when the New Year arrives. Also, most dividend increases occur within the next year, making the previous year-end yield inoperative.
Freezing the portfolio names for the year. In 2013, the DJIA replaced three stocks on September 20. Out were the weaker companies: Alcoa, Bank of America, and Hewlett-Packard. In were the stronger blue-chips: Goldman Sachs Group, Nike, and Visa. By ignoring these changes and keeping the heavily-weighted Hewlett-Packard position, the Dog portfolio broke its own rule of focusing only on the DJIA’s blue-chip stocks for risk control.
No rebalancing. With its limited portfolio of only ten holdings, the Dog portfolio should periodically rebalance to keep the risk controlled. The 12.5% allocation to Hewlett-Packard on September 20 is a good example. Even without the DJIA replacement, the 1/8 weighting skewed the risk significantly. Moreover, not rebalancing means missing the opportunity to buy the underweight positions at relatively better prices.
Ignoring basic portfolio construction analysis. For 2014, as in 2013, AT&T and Verizon Communications are two of the Dog portfolio holdings. Thus 20% is in telecom. Other concentrations are 30% technology (Cisco Systems, Intel and Microsoft), 20% major drugs (Merck and Pfizer) and 30% other (Chevron, General Electric, and McDonald’s). Note that none of the DJIA’s five financial stocks (American Express, Goldman Sachs, J.P. Morgan, Travelers Cos and Visa) made it, even though J.P. Morgan has the second lowest forward P/E ratio. Such concentrations mean the Dog portfolio is taking major sector/industry bets, raising the risk level without any analysis to support doing so.
Below is the 2014 Dogs of the Dow sector allocation (using both price- and equal-weighting) compared to that of the DJIA. Clearly, the Dogs’ omission of 20 DJIA stocks produces a radically different look. The equal weighting further shifts things about. .
The 100% dependence on dividend yield with no human intervention. By using yield only, the Dog portfolio creates anomalies, such as not holding J.P. Morgan, as discussed above. Other 2014 Dog portfolio anomalies include holding two top 2013 performers, Microsoft (+44%, including dividends) and General Electric (+38%). Their dividend yields may put them into the top ten, but their performance argues against their being value or contrarian positions. On the other hand, the 2014 Dog portfolio excludes the two weakest DJIA 2013 performers: Caterpillar (+3.4%) and IBM (-0.2%). Their yields don’t qualify them, but that performance and the commentary surrounding the companies certainly put them into the value-contrarian camp.
The bottom line
The “Dogs of the Dow” investment scheme has a weak basis, a poor process and, measured properly, unsupportive performance results. Therefore, we can take it off our list of things to think about.
A last thought about focusing on dividend yield at this time – 2014 looks to be a year in which growth rules. Economic, business, consumer and financial trends all point in that direction. With widespread interest in stock investing, a higher dividend yield in 2014 will likely indicate an uninteresting laggard, not an attractive bargain.