Feeds:
Posts
Comments

Archive for August, 2010

In an interesting post, Larry MacDonald introduces David Stanley’s Beating the TSX list (BTSX) and then questions whether the list can continue to outperform the TSX. Larry suggests that if many more people adopt the BTSX it cannot continue to outperform the TSX. For evidence he points to the fact that the Dogs of the DOW strategy has underperformed the DOW by 2% annually for the last 15 years. He goes on to suggest that this underperformance is due to the popularity of the Dogs of the DOW, noting that in 2000 an estimated $20 billion (U.S.) was invested using the strategy.

In this post I question the proposition that the reason the Dogs of the DOW has underperformed the DOW is because of the popularity of the strategy.  Intuitively, if the amount of money deployed to the Dogs of the DOW was so large that it was affecting returns then, necessarily it would mean that so many people were buying the 10 highest yielding stocks each year that the prices of those stocks were being pushed higher.  Put another way, the dogs would be getting more expensive. Conversely, the dividend yield of these stocks would decline.  Therefore, as more and more money over the last 15 years was committed to the Dogs strategy, you would expect the average dividend yield of the Dog stocks to approach the average dividend yield of the DOW.

By definition the average dividend yield of the 10 Dog stocks will be greater than the average dividend yield of the DOW index. However, as more money is committed to the Dogs Strategy, the price of the 10 Dogs will be pushed higher driving the dividend yield lower. So the difference between the average dividend yield of the Dogs and the average dividend yield of the DOW stocks should get smaller over time.  To test this hypothesis I have compared the average dividend yield of the Dogs to the average Dividend yield of the DOW for the last 15 years.

Year Yield of DOW on date of purchase Yield of Dogs of DOW on date of purchase Difference 5 yr average of difference
1996 2.28% 3.35% 1.07%
1997 1.91% 3.06% 1.15%
1998 1.80% 2.79% 0.99%
1999 1.79% 2.82% 1.03%
2000 1.57% 3.00% 1.43% 1.13%
2001 1.64% 3.03% 1.39% 1.20%
2002 1.95% 3.48% 1.53% 1.27%
2003 2.47% 4.20% 1.73% 1.42%
2004 2.12% 3.61% 1.49% 1.51%
2005 2.26% 3.82% 1.56% 1.54%
2006 4.77% 6.00% 1.23% 1.51%
2007 2.34% 3.59% 1.25% 1.45%
2008 2.60% 4.27% 1.67% 1.44%
2009 3.81% 6.18% 2.37% 1.62%
2010 2.63% 4.17% 1.54% 1.61%

Here is the same data presented as a graph

Examining the data reveals that the difference between the average yield of the DOW versus the Dogs of the DOW is not getting smaller, but appears to be getting larger. Through 1996-1999 the difference between the 2 strategies was approximately 1% and through 2000-2010 the difference was about 1.5%. This argues against the hypothesis that the popularity of the Dogs strategy is affecting returns and I suspect there are other causes for the underperformance.

Having said this, I believe a more definitive conclusion as to whether the popularity of the Dogs strategy is having an effect on its performance could be made by analyzing P/E ratios as P/E is a better measure of how expensive a stock is. I would have preferred to analyze whether the average P/E of the Dog stocks increased over time and whether it increased with respect to the other stocks on the DOW. However, in the absence of easy access to P/E data I decided to extrapolate based on the dividend yield. Perhaps in a future post I will attempt to see if this argument holds based on P/E data.

Finally, I am not sure the performance of the Dogs of the DOW is a good predictor of the future performance of the BTSX strategy. I believe there are fundamental differences between the TSX 60 and the DOW that make comparing the Dogs with the BTSX difficult. I will explore this in my next post.

Advertisements

Read Full Post »