Stingy Investor Contact - Subscribe - Login
  Home | Articles | Screens | Links | SNW | Rothery Report


Simple Screens

High Dividend Yield
  High Yield DJIA30
  High Yield TSX60

Low P/E
  Low P/E DJIA
  Low P/E TSX60

Low P/B
  Low P/B DJIA
  Low P/B TSX60

Dividends at Risk
  Dividend Risk DJIA30
  Dividend Risk TSX60

Value Ratio
  Value Ratio DJIA30
  Value Ratio TSX60


The Value Ratio Approach

In the May 1999 edition of the Canadian MoneySaver, I reviewed several stock-picking strategies that revealed the positive effect that one fundamental factor can have on investment returns. This article reviews some of my research into a slightly more complicated two factor approach which I refer to as the Value Ratio (VR) approach. While stocks selected solely on the basis of high dividend yield (yield) or low price-to-earnings ratio (P/E) have performed well in the past, a combined approach with the VR method may achieve better performance. For comparative purposes the yield-based Beating the TSE method is also presented and some of the potential pitfalls associated with using the VR approach are discussed.

The current investigation is limited to the 35 stocks that are in the TSE 35 index. This index is composed of large, sound and well known Canadian companies. The approach could be easily applied to other indices but the TSE 60 is less than a year old and the TSE 300 contains a few issues of questionable soundness.

The value ratio approach is based on selecting stocks that have low ratios of P/E-to-yield (thus the term Value Ratio). In algebraic terms the value ratio of a stock is given by:

Value Ratio = (P/E) / (Dividend Yield) = (Price/Share * Price/Share)/(Earnings/Share * Dividends/Share)

As the formula demonstrates, stocks with low value ratios have a combination of low prices, high earnings, and high dividend yields. Intuitively, these characteristics should be viewed as quite favorable to the buyer of securities. The VR approach is initiated by sorting the stocks in the TSE 35 by VR each year (with the prior removal of any stocks that have no dividends or no earnings). A VR portfolio is created by purchasing equal dollar amounts of a number of stocks with the lowest VRs. The portfolio is held for a year, sold and a new one is made based on a fresh VR sorting. The historical results of applying this technique to portfolios with one, five and ten stocks are presented in Table 2. All three portfolios beat the TSE 35 total return, although the ten-stock portfolio did not fare nearly as well as the others. It should be noted that, in the one-stock case, I am using the term portfolio rather loosely since one stock can hardly be said to constitute a portfolio.

Comparative results for the Beating the TSE approach are included in Table 2 courtesy of a study undertaken by Jennifer Hadrevi (See dogs). Her results may be slightly different from those calculated by David Stanley in his Canadian MoneySaver articles since the portfolios shown in Table 2 are rebalanced on December 31 whereas most of David's figures are based on a May 27 rebalancing. The ten-stock Beating the TSE portfolio is composed of the ten highest yielding stocks in the TSE 35. The one-stock portfolio is created by sorting the ten high yielders by price and then selecting the second lowest priced stock. The four-stock portfolio is created by sorting the ten high yielders by price and then selecting the second through fifth lowest priced stocks with double amounts of the second lowest. Readers of David Stanely's fine articles and the "foolishly inclined" ( should recognize these common variants. On the whole, the Beating the TSE approach has proven profitable in the past. It also beats the VR approach in the ten-stock portfolio case, but it is not nearly as good when it comes to smaller portfolios.

Table 2: Comparison of annual total returns for the Value Ratio, Beating the TSE and indexing approach.
Year Value Ratio Beating the TSE 35 TSE 35
  10 Stock 5 Stock 1 Stock 10 Stock 4 Stock 1 Stock  
1988-1989 15.13 23.03 13.81 14.16 7.41 12.10 11.32
1989-1990 17.91 19.02 22.59 20.01 19.55 32.49 21.89
1990-1991 -10.27 -0.51 5.97 -9.76 -13.66 -24.69 -11.65
1991-1992 33.48 46.86 79.56 30.21 18.36 -12.63 11.01
1992-1993 -1.44 -3.02 -25.93 -2.04 -6.06 -25.93 -3.58
1993-1994 17.44 12.15 28.83 26.47 27.49 28.83 24.34
1994-1995 -1.01 -2.21 1.84 3.24 3.40 5.02 5.52
1995-1996 11.55 14.84 6.54 14.18 10.94 7.56 14.47
1996-1997 31.57 31.02 15.34 38.10 36.62 23.63 30.01
1997-1998 30.75 48.63 72.94 26.66 40.50 72.94 16.38
1998-1999 -3.13 -6.59 -9.68 0.98 -5.99 -15.86 -0.09
1999 to 7/3 7.52 27.47 40.97 8.03 21.51 40.97 15.53
Average : 12.55 17.29 20.57 14.19 13.34 12.04 11.26
High : 33.48 48.63 79.56 38.1 40.5 72.94 30.01
Low : -10.27 -6.59 -25.93 -9.76 -13.66 -25.93 -11.65
Growth : 3.72x 6.06x 6.98x 4.48x 3.95x 2.67x 3.37x
Beats the TSE35 : 6 / 12 8 / 12 7 / 12 8 / 12 5 / 12 5 / 12  

The one stock VR approach exhibits extraordinary returns but with very high volatility. The dangers inherent in a one-stock method are revealed in the historical returns of the one-stock Beating the TSE portfolio. In the period between 1990 and 1993 someone following this method would have lost over 50%. It might be considered a nearly Herculean feat to stick with a method after such a ghastly run. Although it hasn't happened, the one-stock VR method cannot be said to be immune from such occurrences. Therefore, both one-stock methods should be considered only by plungers or new investors who only want to buy one stock for trial purposes.

The objection might reasonably be raised that the healthy returns posted by the VR approach may be due to the stellar returns of the best VR stock (i.e. the 1991 & 1997 one-stock returns). To investigate this possibility two portfolios were created. The first, with average gains 16.68% per year, was composed of the second through fifth lowest VR stocks. The second, with average gains of 15.10% per year, was made by selecting the third through fifth lowest VR stocks. This result reduces the fear that the figures presented in Table 2 are overly biased by the amazing 70-80% one-stock returns of 1991 and 1997.

Average annual results for portfolios with one through to ten stocks with the lowest VRs are presented in Table 3. The clear trend to lower returns for larger portfolios may reflect the inclusion of less attractive value stocks in the larger portfolios. A portfolio size of ten represents almost one third of the TSE 35 and it seems unlikely that a third of this index represents good value at any given time. For investors desiring larger portfolios the TSE 60 is an obvious alternative.

Table 3: Average annual returns versus VR portfolio size.
#Stocks 10 9 8 7 6 5 4 3 2 1
Average 12.46 12.57 14.22 14.62 15.11 17.56 17.89 20.39 21.23 21.06
1Yr Max 33.48 33.24 35.65 40.80 42.99 48.63 50.34 62.77 66.62 79.56
1Yr Min -10.27 -9.66 -9.07 -9.38 -7.43 -6.59 -9.57 -8.43 -14.96 -25.93
Growth 3.72x 3.77x 4.42x 4.54x 4.77x 6.06x 6.23x 7.78x 8.07x 6.98x

It is now time to throw a bit of cold water on the VR approach with the immortal words of Benjamin Graham:

"The moral seems to be that any approach to moneymaking in the stock market which can be easily described and followed by a lot of people is by its terms too simple and too easy to last." - The Intelligent Investor.
To my mind this quote nicely encapsulates the problem with stock picking methods in general. Additionally, the figures presented here are subject to a possible start and stop date bias. That is, the results may be different when different portfolio rebalancing dates are chosen. Furthermore, one would also like to test the approach with a more substantial data record, but unfortunately the TSE 35 hasn't been around for very long. As well, there is always the very real danger that past results may be a poor predictor of future returns. A few other potential problems and uncertainties come to mind but Graham's objection is perhaps the strongest. Nonetheless, the strength of value investing has persisted for very long periods and the behavioral factors that give rise to it are unlikely to dissipate. As a result, I expect that the VR approach should be of value to investors, at least until too many hop on the bandwagon.

About Us | Legal | Contact Us
Disclaimers: Consult with a qualified investment adviser before trading. Past performance is a poor indicator of future performance. The information on this site, and in its related newsletters, is not intended to be, nor does it constitute, financial advice or recommendations. The information on this site is in no way guaranteed for completeness, accuracy or in any other way. More...