DGI Vs. The Annualized Real Return From Dollar Cost Averaging Into The Market

|
Includes: ROST, SPY
by: Dividend Math Guy

Summary

Investors frequently compare their past returns to the past returns of the market.

Past returns are no guarantee of future returns.

The historical real return from dollar cost averaging into the market has been fairly consistent.

To outperform the market going forward, it seems reasonable to select stocks that have a high probability of beating these historical real returns.

Dividend growth investing provides an opportunity to beat the market going forward.

Indexing vs. other strategies

There have been several articles on Seeking Alpha lately attempting to answer the question of whether dividend growth investing beats a passive index strategy in total return. In this article, for example, the author points out that the SPDR S&P 500 ETF Trust (NYSEARCA:SPY) has compounded at an annualized rate of 15.67% for the last five years, while most dividend growth portfolios have not. Aside from whether this is a cherry-picked time frame (it starts at the beginning of the current bull market, just after SPY fell much further than most dividend growth stocks did, for instance), it is not particularly relevant to our current needs as investors in that it provides no reasonable expectation about future returns.

As a young investor, my primary goal is accumulating enough assets to fund my eventual retirement. I identify as a dividend growth investor because I feel that the dividend growth strategy gives me the highest chance of achieving my goals, but I cannot deny that an investor who earns a higher total return during the accumulation phase is better positioned going into retirement regardless of what strategy is used to obtain that return. Nevertheless, I believe that intelligently executed dividend growth investing has a reasonable chance of outperforming the market going forward, and in this article I lay out why.

To do so, I will do the following two things.

  1. First, I will explain what I expect the future market returns to be over a long time span.
  2. Second, I will show how intelligently selecting dividend growth stocks gives a high probability of beating these returns.

Future market returns from past market returns, 1928-2014

I think the best way to predict the future is by studying the past. Since I am going to make many investments over the course of many years leading up to my retirement, and since inflation distorts market values and investment results, I am most interested in knowing what the historical real (inflation-adjusted) market returns from dollar cost averaging over long time periods (20+ years) have been. However, I was not able to find any information about this on the internet, so I have been forced to do the calculations myself.

Historical 1-year nominal stock market returns (price change + dividends) since 1928 can be found here.

The real return from dollar cost averaging into the market

Suppose an investor dollar cost averaged into the market over a long time period (20 years or more), what would his/her real (inflation-adjusted) returns have been?

To answer this question I looked at three different dollar cost averaging scenarios intended to mimic how real people invest, over every 20+ year stretch from 1928 to present.

  1. Scenario 1: The investor invests a constant amount of money (say $5000) each year.
  2. Scenario 2: The investor invests $5000 during the first year and varies the amount invested each following year according to the inflation rate. If the inflation rate from year 1 to year 2 is 10%, for instance, in year 2 $5500 is invested. On the other hand if the inflation rate from year 1 to year 2 is -10% (that is, 10% deflation), then in year 2 $4500 is invested instead.
  3. Scenario 3: The investor invests $5000 during the first year and varies the amount invested each following year according to the inflation rate, except that he never increases his contribution by more than 5% from one year to the next. (The idea here is that it is difficult to keep up with high inflation.)

Surprisingly, the results were virtually identical for all three scenarios, so I'll only record here the results from Scenario 1. For market returns, I used the link above, and for inflation, I used the government's official cpi-u data. Here are the results.

Table 1: Annualized real return (as a percentage) from dollar cost averaging into the market for N years, starting in year Y

N / Y

1928

1929

1930

1931

1932

1933

1934

1935

1936

1937

1938

1939

1940

1941

1942

1943

1944

1945

1946

1947

1948

1949

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

1960

1961

1962

1963

1964

1965

1966

1967

1968

1969

1970

1971

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

20 yrs

2.88

2.99

4.53

6.01

7.03

7.67

6.8

9.69

11.28

11.27

10.01

12.11

12.54

12.12

13.37

11.75

12.32

12.65

12.71

11.16

11.7

11.1

8.75

7.45

7.23

7.37

4.47

-0.15

1.45

2.43

0.84

0.04

0.11

1.21

-0.17

0.92

1.96

1.98

3.71

4.84

4.8

5.56

7.21

6.43

8.03

8.25

8.77

8.51

10.03

10.87

12.53

13.74

14.31

12.62

10.91

7.87

8.79

8.53

7.68

7.87

7.26

2.74

3.72

4.16

3.33

3.85

5.41

21 yrs

2.94

4.29

5.75

6.81

7.73

6.97

9.45

11.02

10.82

9.5

11.79

12.01

11.63

12.92

11.74

12.29

12.5

12.5

10.84

11.7

11.32

9.24

7.98

7.67

7.74

4.96

0.8

1.91

2.64

1.14

0.57

0.38

1.37

0.1

0.99

2.07

1.99

3.54

4.57

4.61

5.23

6.76

6.11

7.64

7.78

8.18

8.09

9.95

10.57

12.02

13.33

13.98

12.46

10.79

8.18

9.01

8.7

8

8.04

7.4

3.25

4.14

4.39

3.78

4.11

5.6

22 yrs

4.19

5.48

6.53

7.49

7.05

9.48

10.73

10.6

9.16

11.22

11.71

11.18

12.42

11.41

12.25

12.47

12.37

10.75

11.37

11.34

9.54

8.48

8.15

8.12

5.44

1.46

2.66

3.01

1.4

0.87

0.85

1.55

0.3

1.19

2.09

2.09

3.47

4.37

4.37

5.04

6.39

5.75

7.28

7.43

7.74

7.57

9.48

10.46

11.67

12.8

13.58

12.26

10.74

8.22

9.23

8.91

8.19

8.32

7.58

3.61

4.52

4.74

4.02

4.48

5.74

23 yrs

5.34

6.25

7.21

6.85

9.44

10.69

10.35

9.03

10.82

11.18

10.94

11.96

11.02

11.92

12.42

12.34

10.73

11.26

11.05

9.66

8.81

8.61

8.55

5.93

2.1

3.17

3.64

1.8

1.12

1.11

1.93

0.52

1.32

2.22

2.1

3.51

4.27

4.2

4.79

6.15

5.45

6.88

7.1

7.42

7.19

8.92

9.99

11.52

12.43

13.07

11.98

10.65

8.33

9.21

9.12

8.42

8.48

7.87

3.99

4.79

5.06

4.36

4.66

6

24 yrs

6.09

6.92

6.61

9.17

10.61

10.33

8.86

10.63

10.81

10.47

11.69

10.65

11.51

12.1

12.31

10.79

11.21

10.96

9.48

8.96

8.92

8.96

6.45

2.73

3.67

4.07

2.46

1.51

1.34

2.13

0.92

1.48

2.3

2.22

3.46

4.27

4.11

4.61

5.87

5.27

6.54

6.73

7.1

6.91

8.5

9.43

11.02

12.25

12.7

11.58

10.47

8.38

9.26

9.11

8.65

8.68

8.05

4.46

5.08

5.28

4.68

4.94

6.09

25 yrs

6.75

6.36

8.86

10.31

10.27

8.91

10.4

10.63

10.14

11.22

10.45

11.14

11.71

12

10.83

11.24

10.93

9.48

8.83

9.05

9.23

6.95

3.41

4.19

4.5

2.92

2.14

1.69

2.3

1.15

1.82

2.41

2.3

3.52

4.19

4.11

4.51

5.65

5.03

6.32

6.42

6.75

6.64

8.17

9.01

10.44

11.73

12.51

11.3

10.17

8.34

9.26

9.16

8.66

8.88

8.25

4.8

5.46

5.53

4.91

5.22

6.29

30 yrs

8.02

9.51

9.83

9.57

10.43

9.52

9.96

10.34

10.46

9.6

10.32

10.27

9.38

9.08

9.24

9.38

7.68

5.14

5.95

6.45

5.23

4.49

4.02

4.19

2.94

3.2

3.55

3.11

3.93

4.45

4.26

4.43

5.29

4.67

5.56

5.61

5.72

5.52

6.76

7.48

8.59

9.58

10.27

9.58

8.84

7.52

8.57

8.77

8.43

8.71

8.55

6.03

6.59

6.72

6.28

6.36

7.13

35 yrs

8.72

9.29

9.72

9.95

9.2

9.57

9.43

8.62

8.27

8.5

8.93

7.71

5.88

6.69

7.09

6.08

5.53

5.31

5.61

4.79

4.99

5.17

4.72

5.17

5.35

4.99

5.04

5.49

4.83

5.57

5.52

5.49

5.24

6.21

6.69

7.59

8.3

8.79

8.2

7.61

6.47

7.3

7.53

7.49

7.86

7.84

6.07

6.75

6.89

6.62

6.94

7.69

Table 2: Statistics from Table 1

N

Mean

Median

Standard deviation

20 yrs

7.19%

7.45%

4.09%

21 yrs

7.25%

7.73%

3.89%

22 yrs

7.3%

7.58%

3.7%

23 yrs

7.33%

7.64%

3.52%

24 yrs

7.35%

8.05%

3.35%

25 yrs

7.35%

8.21%

3.18%

30 yrs

7.21%

7.48%

2.31%

35 yrs

6.95%

6.82%

1.54%

Histogram of 20-year real returns

Histogram of 35-year real returns

From this data several interesting results are revealed.

  1. The annualized real return from dollar cost averaging into the market over a 35-year time period has never exceeded 10%.
  2. The annualized real return from dollar cost averaging into the market over a 35-year time period has almost always exceeded 5%.
  3. An investor who dollar-cost averaged into the market for 20 years starting in either 1955 or 1962 lost money (purchasing power). Investing over any other time period or for any longer length of time provided positive real returns. The poor results for the 20-25 year investment periods beginning around 1960 came about from a trifecta of return-destroying factors. First, the market was pricey, with P/E ratios consistently between 17-25 and frequently above 20, from 1960-1972. Second, inflation was very high from 1974-1982. Finally, around 1974 the market plummeted and then P/E ratio of the market contracted, before finally settling into a P/E of 8 in 1981. It is interesting that it took all three of these factors combined to wipe out any real returns from the market. Were any one of those three factors removed, we would have seen annual real returns in the 5%-7% range for these time periods.
  4. The longer the investment time period, the less influence the starting date has on the real return achieved (of course).
  5. For a 20-year investment time period,
    • 90% of the time the real return was at least 1.12%
    • 75% of the time the real return was at least 3.75%
    • 50% of the time the real return was at least 7.49%
    • 25% of the time the real return was at least 11.05%
  6. For a 25-year investment time period,
    • 90% of the time the real return was at least 2.47%
    • 75% of the time the real return was at least 4.60%
    • 50% of the time the real return was at least 8.12%
    • 25% of the time the real return was at least 10.43%
  7. For a 30-year investment time period,
    • 90% of the time the real return was at least 3.87%
    • 75% of the time the real return was at least 5.31%
    • 50% of the time the real return was at least 7.53%
    • 25% of the time the real return was at least 9.47%
  8. For a 35-year investment time period,
    • 90% of the time the real return was at least 5.00%
    • 75% of the time the real return was at least 5.59%
    • 50% of the time the real return was at least 6.73%
    • 25% of the time the real return was at least 8.41%

How I use these results

I am not concerned with whether I am able to outperform the market over short time periods. Rather, I aim to outperform the market over my investing lifetime. Based on the above data, I expect that dollar cost averaging into the market will produce long-term annualized real returns somewhere between 5%-9%. That average is 7%, so to beat the market going forward I think it is reasonable to aim for annualized real returns of at least 7.5%. With inflation currently around 2% that means I target a minimum of a 9.5% nominal 1-year return on the money I'm investing right now, and with historical long-term inflation around 3%, that means I should target a minimum of 10.5% nominal annual return over the long haul.

An example: Ross Stores

Off-price retailer Ross Stores (NASDAQ:ROST) is one of my favorite low-yield high-growth dividend growth stocks. The company is still expanding rapidly within the US, and sees the store count doubling to 2500 before saturating the US market. The stock's P/E ratio typically fluctuates between 14 and 20. The company typically repurchases a bit more than 3% of its outstanding shares each year and carries very little debt.

Let's contemplate the annualized return we might get from buying ROST at a P/E of 14.5 - which was available as recently as July - and holding it for 20 years. Assuming the store count doubles during this time, the per-store client base grows at 1% per year (roughly in line with the population growth in the US), sales per customer increase at the rate of inflation (say 3% per year), and the company continues buying back 3% of its stock each year, 20 years from now each share will be pumping out 8.1 times as much in profits as it is right now. If the P/E remains unchanged, this amounts to an 11% annualized return from price change alone.

On the other hand, if the P/E expands to 18, each share will be worth 10 times what we bought them for, which amounts to a 12% annualized return from price change alone.

Add in the roughly 1% dividend yield per year and you're looking at a very reasonable chance of obtaining a 12%-13% annualized return over a 20-year time period (that is, a 9%-10% annualized real return), which has a very good chance of beating the market going forward.

My last purchases of ROST were in July at $68/share (at 15.7x earnings) and $63/share (at 14.5x earnings).

This same exercise can be repeated with many of my favorite dividend growth stocks, including Visa (NYSE:V), BlackRock Inc. (NYSE:BLK), Chevron (NYSE:CVX), Kinder Morgan Inc. (NYSE:KMI), and Johnson & Johnson (NYSE:JNJ). When purchased at sensible valuations, I believe all of these have an excellent chance of outperforming the market over the long term.

Summary and conclusions

The past is in the past and comparing recent past returns of your portfolio to the market may not reveal whether your strategy is working as intended. Rather than worrying about the day-to-day fluctuations between your portfolio and the market, to beat the market in the long run, I feel it is sensible to make investments that have a high probability of returning at least 7.5% per year after inflation. Fortunately, there are a number of excellent opportunities in the realm of my preferred investing style to do exactly that.

Disclosure: The author is long BLK, CVX, JNJ, KMI, SPY, V. 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.