(Author's Note: After a lengthy discussion with a reader (mjs_28s), I have asked that some terminologies originally used in this article be changed in favor of the common usage. These changes are reflected in the text below.)
When I wrote "Dogging the Dow: Examining the 'Dogs of the Dow' Strategy," followed by "Dogging the S&P 500: PIC-ing Another Kennel," I introduced my own strategy, which I now refer to as the "PIC" strategy (for "Potential - or Preferred - Investment Criteria"), and put together portfolios of companies chosen by my criteria to compare against the portfolios chosen by the "Dog" strategy. Since I have been maintaining weekly updates on my Instablog, I was going to pass up a mid-year update on the comparisons; however, during the past few weeks (actually, the three months beginning about mid-May) the markets - and the portfolios - have had their ups and downs, so to speak, and this has made certain features of my experiment interesting, perhaps uncovering a weakness to the Dog strategy that may not be obvious at face value.
The Dow Jones 30 Industrials Portfolios
The Dogs of the Dow consist of those ten companies that offer the highest dividend yield (traditionally based on the dividends and share price of a given company on December 31). The strategy involves dividing an amount of money (say, $10,000.00) equally among the ten Dogs, letting the investment ride untouched for a full year, at which time the portfolio is adjusted to accommodate any changes among the Dogs. Since its "introduction" in 1972 it has managed to yield an average of 14% to the Dow's average of 11%.
The "PIC" strategy involves using a set of criteria - outlined in the first article above - to select a portfolio of companies from the Dow that have what would appear to be the solid fundamentals (in deference to the Dog strategy, as of December 31) that would indicate a well-managed company. I originally called the portfolio the "Pedigrees of the Dow," but - in keeping with the strategy - I now call it the "PICs of the Dow." My intention here is to determine which is a better indicator of performance over the course of a year - a company's dividend yield or a selection of its fundamentals.
As it turned out, for this year there are five companies that appear in both portfolios: Intel Corporation (INTC); Johnson & Johnson (JNJ); McDonald's Corporation (MCD); Merck & Co. (MRK); and Pfizer Inc. (PFE). The other five Dogs were: DuPont (E.I.) De Nemours (DD); General Electric Company (GE); Hewlett-Packard Co. (HPQ); AT&T Inc. (T); and Verizon Communications (VZ). As for the PICs, the remaining companies were: Cisco Systems (CSCO); Chevron Corp. (CVX); International Business Machines (IBM); 3M Company (MMM); and Microsoft Corp. (MSFT).
The Comparison, Year-to-Date
As mentioned above, the comparison between the Dogs and the PICs involves a cost basis of $10,000 per portfolio, divided equally between each of the ten companies in that portfolio. The number of shares per company per portfolio will depend upon the share price; I allow for fractional shares even though that may not be possible in actuality. I also keep track of two data: performance, which is in keeping with the traditional measure of performance for purposes of accurate comparison with the actual index (in this case, the Dow); and total return, which includes dividends paid during the existence of the portfolio (in response to those readers who felt this is only fair since the Dog strategy is based on dividend yield).
The following table reflects the performance of the PICs and the Dogs for the first 32 weeks of this year (through August 9, 2013):
Clearly, the Dog portfolio is substantially ahead of the PIC portfolio. How they compare to the Dow in terms of their total return to the Dow's performance is shown in the following chart:
Both portfolios perform better than the Dow as a whole when dividends are included. In the following chart, we can compare the portfolios against the Dow in terms of performance alone:
Here, the Dogs are still clearly outperforming the Dow, but the PIC portfolio is only marginally ahead of the index; indeed, for the past four weeks the PICs and the Dow have been "leapfrogging" each other.
The Hewlett-Packard Difference
My expectation for the PIC portfolio was that it would perform somewhat significantly better than the Dow, but 29 BPS is hardly "significant" - it would barely qualify as "marginally" better. The Dogs, on the other hand, have the Dow beaten quite convincingly thus far, and this would seem to support the general expectation that the Dogs of the Dow generally perform (much) better than the Dow. But there is something particular in the way the Dogs are accomplishing this - one of the Dogs is Hewlett-Packard (or, "HP").
Why should HP's role in the Dogs' performance be "particular"? Consider the following chart:
HP is performing at nearly 90% - far and away higher than that of any other member of the Dow Industrials. Indeed, running second to HP is Boeing Co. (BA) at 42.03%. If we were to abstract HP out of the Dog portfolio, the remaining nine companies would be performing at 16.87% - well below the Dow and the PICs. The total return for the nine-stock portfolio would be 18.93% - 80 BPS lower than the total return of the PICs.
If one were to try to substitute another company for HP and still bring the Dogs' performance above that of the Dow and the PICs, it would take a company that currently has a performance of 30% or better, which would put it in the top-ten-performing members of the Dow. If we keep the dividend requirement, Cisco (which has the fifth-highest performance in the Dow) would be the most likely candidate, and replacing HP with Cisco would give the Dogs a performance of 18.44% and a total return of 20.44%, which would lead both the Dow and the PICs.
Put briefly, for all intents and purposes, the only reason the Dogs of the Dow are doing as well as they are is because HP is part of the portfolio. And who would have imagined, last January, that HP would be on track - in mid-year - to be a two-bagger (and better)?
HP and the Pups
The Pups of the Dow ("Puppies of the Dow," "Small Dogs of the Dow," or "Flying Five") is a variation on the Dog strategy wherein the five lowest-priced Dogs are the only holdings in the portfolio. It has been quite a success, its average of more than 20% return per year since 1973 surpassing both the Dow and the Dogs.
This year's Pups are HP, GE, Intel, Pfizer and AT&T, and their performance is exceptional at 26.45%, while their total return is 28.56%. Again, however, it is totally by virtue of having HP that the Pups are doing so well - abstracting HP out of the equation, the other four Pups have a performance of 11.10% and a total return of 13.24%. Filling the fifth slot with the next-lowest-priced company (Merck) gives a performance of 12.52% and total return of 13.24. If we took the Cisco substitution described in the previous section, Cisco would be the fifth Pup, and the share performance would be 15.39% and total return would be 17.45% - not bad, but still below the Dow overall.
The S&P 500 Portfolios
When I started to put the PIC portfolio together for the S&P I found that I needed to add to the set of criteria I was using - the modified (and what currently constitutes the PIC) criteria are discussed in the second article ("Dogging the S&P …") listed at the top of this article. The Dogs of the S&P were taken from a list developed by StockMarketPundits who kindly allowed me to use his material in preparing the list of Dogs.
While there is no overlap between the Dogs and the PICs of the S&P, three PICs of the Dow also appear in the PICs of the S&P: Cisco, Johnson & Johnson and Pfizer. The remaining PICs of the S&P 500 are: Analog Devices Inc. (ADI); Baxter International Inc. (BAX); KLA-Tencor Corporation (KLAC); Eli Lilly & Co. (LLY); Mattel Inc. (MAT); Seagate Technology PLC (STX); and Texas Instruments Inc. (TXN).
The Dogs of the S&P 500 are: Cliffs Natural Resources Inc. (CLF); CenturyLink, Inc. (CTL); Exelon Corporation (EXC); Frontier Communications Corp. (FTR); Altria Group Inc. (MO); Pitney Bowes Inc. (PBI); Pepco Holdings, Inc. (POM); Reynolds American Inc. (RAI); R.R. Donnelley & Sons Co. (RRD); and Windstream Corporation (WIN).
The Comparison, Year-to-Date
The S&P comparison is structured the same way as that of the Dow, $10,000 per portfolio, divided equally, fractional shares, etc. The following table shows how the two portfolios stand as of August 9:
In this case, the PIC portfolio leads the Dogs by a good margin if not quite as substantial as the lead the Dogs have over the PICs in the Dow. With regard to the S&P 500 performance, and how it compares to the portfolios we can look at the following chart:
Both the PICs and the Dogs lead the index, just as they do in the Dow comparison. When we look at share performance, however, things are a bit different:
In terms of share yield, the Dogs underperform the S&P 500 by 133 BPS. Moreover, while the PICs of the S&P have kept close to the index itself over the past few weeks, the Dogs of the S&P have clearly been underperforming the S&P substantially since mid-May. Let's take a closer look at the past couple of months.
The May Malaise
The following chart focuses on the last 19 weeks of performance in the S&P 500 comparison (from 10 May through 9 August):
Early in May various market information sources began cautioning that a "correction" was due (and many are still cautioning investors about the upcoming "correction"). The week ending May 17 was a high point for the market (not just the S&P, but the Dow, as well), and the following week began a five-week drop by the end of which the PIC portfolio lost 416 BPS, the S&P 500 lost 526 BPS, and the Dog portfolio lost 837 BPS.
In three weeks (the week ending on July 12) the PIC portfolio's performance bounced back to the tune of 751 BPS - settling down by 145 BPS to its level on August 9. It took the S&P six weeks (through the week ending August 2) to jump 822 BPS, then dropping 128 BPS last week. The Dog portfolio, however, jumped 1481 BPS in that same six-week period, before slacking off to lose 5 BPS the week ending August 9.
I don't believe what I'm calling the "May Malaise" is the correction investors are anxious about; it seems clear that there were three factors principally involved in that drop: weakness in the Chinese economy, an apparently poorly performing Japanese economy, and fears about the impending termination (or gradual diminishment) of the latest round of quantitative easing. Continued fears about QE3 seem to threaten the recovery from the May Malaise. There are times when authorities from the Fed would do better to keep their conjectures to themselves, and leave Mr. Bernanke to be the spokesperson for projections concerning any easing of the easing, so to speak.
A Closer Look at the Hot Dogs
In the Dow competition we saw how the exceptional performance of one company (HP) over an extended period can influence (and even mask) the performance of the portfolio as a whole. In the case of the Dogs of the S&P, however, the exceptional performance of the portfolio takes place over a short period of time (six to eight weeks), and does not seem to be due to the improved performance of just one company.
It turns out that four companies (the "Hot" Dogs) put together a substantial push from June 21 through August 2, their individual and collective performances plateauing from August 2 through August 9: R.R. Donnelley, Cliffs, Pitney-Bowes and Frontier. The individual performances are reflected in the following chart:
[Note: this period represented a significant turnaround for Cliffs Natural Resources in that it had been dropping in share price virtually since the beginning of the year. Although it is still in the negative-performance zone, it has performed at a nearly 50% clip since July 1.]
The remaining six companies (Reynolds, Windstream, Excelon, Pepco, Altria and CenturyLink - collectively the "Not" Dogs) started off the eight-week stretch in mixed fashion, all reaching high levels in the first three to four weeks, then cooling off - again, both individually and jointly. The individual performances are reflected here:
The collective performances of both groups and how they relate to the performance of the Dog portfolio as a whole is shown here:
[Note: the data in the three charts above are relative to the low point reached on June 21, rather than being YTD. This was done so as to highlight the difference in performance between the "Hots" and the "Nots" in this brief period of time; the companies' performances during this period may not be representative of their performance YTD (with the exception of R.R. Donnelley).]
A Little Serendipity Helps
As with HP, R.R. Donnelley has been performing extremely well - as of August 9, it had performed at a rate of 114% - an excellent candidate to finish the year as a two-bagger, and one has to wonder if it could become a three-bagger by year's end. Of course, a correction in the general market could change all of this, but up to now, Donnelley has been a hot performer.
As mentioned above, the Dogs of the S&P 500 were culled from a series of articles by StockMarketPundits. While he lists 15 Dog candidates, I used only the first ten, confirming their status on FINVIZ.com. As luck would have it, on December 11, 2012, R.R. Donnelley was moved to the S&P Mid-Cap 400, and replaced with Garmin Ltd. (GRMN). Technically, then, I should be replacing Donnelley with either Garmin or the next Dog in line according to StockMarketPundits - Lorillard Inc. (LO). The difference in performance (YTD) between the three companies is noteworthy:
Having gone this far with the current lineup of Dogs, I am inclined to keep it that way - changing would seriously (and obviously, adversely) affect the Dogs' performance. Without Donnelley, the Dogs' share performance YTD would be 6.50%, and the total return would be 9.93%. Lorillard would not significantly help the Dogs, and Garmin would harm the portfolio.
I mentioned at the outset that I might have found a weakness in the Dog strategy. Having only conducted this experiment for not quite eight months, any observations I make are made on the basis of very little evidence; nonetheless, there is enough of a peculiarity present that it deserves mention, if only by way of cautionary comment.
Specifically, given the situation with Hewlett-Packard and (in particular) R.R. Donnelley, I can't help but wonder to what degree the success of the Dog strategy is due to pure luck, rather than any profound "truth" to be uncovered in the method by which Dog candidates are chosen. I believe that I understand the reasoning behind the Dog strategy enough to say that it does have plenty to support its history of success: especially as pertains to the Dow, companies with high dividend yields are likely to increase in value, if only because the company or companies involved have gone through a period of stress and are undersold. They should expect to increase in value - again, especially as pertains to the Dow.
There is nothing, however, in the Dog strategy (or the PIC strategy, for that matter) that makes the strategy prone to pick a company or two that are going to become two-baggers during their tenure as Dogs. Without HP, the Dogs of the Dow would be underperforming, in all likelihood, both the Dow and the PICs. Without Donnelley, the Dogs of the S&P 500 would have had a fairly sad performance record to date.
That much being said, there is one truth I have become convinced of: most investment strategies, like any gambling system, are largely dependent upon forces that are beyond the capacity of the strategy to predict - that is, LUCK. One's commitment to a strategy should be grounded on an understanding of what the strategy is based on and the reasons for accepting that as a basis for making investments.