Dividend Yield: The Least Useful 'Value' Indicator

|
Includes: DIV, DOD, HDV, HIE, HTD, KBWD, PEY, RDIV, SPHD, VYM
by: Investing Discipline

Summary

I have previously shown that investors who seek stability should focus on volatility rather than dividend yields.

However, value investors focusing on excess returns may be tempted to use dividend yields as a screen for their value portfolios.

In today’s article, I explain why dividend yields are inferior both from the theoretical and empirical standpoints.

Introduction

I have shown in my previous article that low volatility stocks are usually a much better option for investors who seek stable returns. This is often the case with investors who draw down from their portfolios regularly for their spending needs.

High dividend yield strategy, which is the more common approach, has had much more volatile total returns compared to the low volatility strategy. However, it has slightly outperformed the low volatility strategy, albeit with higher volatility. For this reasons, volatility insensitive investors may be tempted to buy high dividend payers in order to achieve higher overall returns over the long term. It would make sense to use dividend yields for value screens as they are simply the inverse of Price to Dividend ratio, meaning that high dividend yields should be a similar signal to low Price to Earnings or Price to Cash Flow.

However, dividends only tell one part of the story. Dividends are only net income that is not reinvested in the business. Price to Book Value, Price to Cash flow and composite z-scores have worked considerably better in backtests, as I have shown in my previous analysis. The reason is that these indicators focus on more broadly defined fundamentals of the underlying businesses.

Analysis

Below are the backtest results that compare high dividend strategy to low Price to Earnings (PE), low Price to Book Value (PTBV), low Price to Cash Flow (PC) and the underlying index. The results are different compared to my previous backtests. This is because instead of initially removing all observations where at least one multiple has a negative value, I have now only removed negative values from the data for each of the multiple separately. This approach better reflects what is typically done in practice when stocks are being screened.

As previously, each of the portfolios is formed of equal weighted baskets of the lowest decile ranked by each of the indicators (the highest decile is used for dividend yield). Re-balancing is performed on a quarterly basis. The chart above shows that the high dividend strategy has underperformed all other value strategies, even including the rather weak PE strategy. Furthermore, while High Dividend Yield decile has outperformed the index, due to its volatility it has done so with the risk/return ratio of 0.56, compared to 0.6 for the S&P 500 index. The table below shows that the annualized returns for the PC and PTBV strategies have been superior, both in the absolute and risk-adjusted terms.

Table 1: Returns and Volatility for selected deciles

High DY

Low PE

Low PTBV

Low PC

S&P 500

Annualized Return

11.08

12.60

17.36

17.88

9.55

Annualized Std Dev

19.74

23.12

25.08

24.94

15.83

Annualized Sharpe (Rf=0%)

0.56

0.54

0.69

0.72

0.60

The table below reminds us that the downside risk is quite high for the value strategies in general. This is why "Value" is not for everyone, especially not for people with relatively short investment horizons and/or regular cash requirements. Importantly, despite their more modest returns, High Dividend yielding stocks have not had less downside risk compared to the superior PTBV and PE strategies.

Table 2: Downside Risk for selected deciles

High DY

Low PE

Low PTBV

Low PC

S&P 500

Semi Deviation

6.98

8.12

8.41

8.22

6.01

Gain Deviation

7.01

8.12

9.60

9.84

4.74

Loss Deviation

7.22

7.47

7.54

7.49

5.93

Downside Deviation (MAR=3.33%)

5.94

6.77

6.53

6.28

5.19

Downside Deviation (Rf=0%)

5.60

6.40

6.19

5.94

4.84

Downside Deviation (0%)

5.60

6.40

6.19

5.94

4.84

Maximum Drawdown

64.08

67.80

64.26

61.06

45.80

Historical VaR (95%)

-9.88

-16.57

-13.19

-11.41

-13.32

Historical ES (95%)

-19.54

-20.24

-20.99

-20.19

-15.82

Modified VaR (95%)

-12.15

-13.69

-13.19

-11.47

-11.47

Modified ES (95%)

-18.67

-18.95

-17.11

-12.29

-16.26

Conclusion

Firstly, dividend yields are not a good indicator of "value" for long term investors. Total returns matter, and measures of total firm profitability such as the Operating Cash Flow better reflect the total return prospects for investors. The backtest conducted shows that selecting companies based on cash flows instead of dividend yields would have led to an outperformance of as much as 6.8% per annum over the last 25.5 years.

Secondly, the improved backtest based on more realistic screening criteria shows that performance of value strategies is even better compared to my initial study. The annualized return of 17.9% for the PC strategy is 8.3 percentage points higher than the index. This means that $100,000 invested in the top PC decile would have resulted in a terminal value of $6,626,534, compared to $1,023,804 for the passive approach.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.